Kathrin Wunsch ANTICIPATORY PLANNING IN CHILDHOOD The development of anticipatory planning and its relationship to executive functions Dissertation Thesis in partial fulfillment of the requirements for the degree of doctor philosophiae (Dr. phil.) at the Faculty of Natural Sciences at the University of Paderborn

Anticipatory planning in childhood The development of anticipatory planning and its relationship to executive functions Dissertationsschrift zur Erlangung des akademischen Grades doctor philosophiae(Dr. phil.) an der Fakultät für Naturwissenschaften der Universität Paderborn Dissertation Thesis in partial fulfillment of the requirements for the degree of doctor philosophiae(Dr. phil.) at the Faculty of Natural Sciences at the University of Paderborn vorgelegt von/ by Kathrin Wunsch November, 2014

The work described in this thesis originated between 2010 and 2014 in the research group of Prof. Dr. Matthias Weigelt. Until October 2011, the work originated in the Sport Psychology and Movement Science Unit at the Institute of Sport Science at Saarland University, from October 2011 until closing of the doctoral studies in the Sport Psychology Unit at the Department of Sport& Health at the University of Paderborn. Die in dieser Dissertation beschriebenen Arbeiten entstanden in der Zeit von 2010 bis 2014 in der Arbeitsgruppe unter der Leitung von Prof. Dr. Matthias Weigelt; bis Oktober 2011 im Arbeitsbereich Sportpsychologie und Bewegungswissenschaft am Sportwissenschaftlichen Institut der Universität des Saarlandes und von Oktober 2011 bis zum Abschluss der Doktorarbeit im Arbeitsbereich Sportpsychologie im Department Sport& Gesundheit an der Universität Paderborn. 1. Gutachter: Prof. Dr. Matthias Weigelt 1 2. Gutachterin: Prof. Dr. Gisa Aschersleben 2 1 Department Sport& Gesundheit, Arbeitsbereich Sportpsychologie, Universität Paderborn Department of Sport& Health, Sport Psychology, University of Paderborn 2 Psychologische Fakultät, Arbeitsbereich Entwicklungspsychologie, Universität des Saarlandes Faculty of Psychology, Developmental Psychology, Saarland University

To my father 27.12.2012

Contents Abbreviations(alphabetical)........................................................................................ I Register of Figures..................................................................................................... II Register of Tables...................................................................................................... V Summary.................................................................................................................. VII Zusammenfassung................................................................................................... VII Chapter 1- General introduction............................................................................ 1 1.1. From intentions to actions ­ Goal-driven motor behavior........................................... 3 1.1.1. The End-State Comfort Effect.......................................................................... 8 1.1.2. The grasp-height effect...................................................................................13 1.2. The role of cognitive control in motor behavior..........................................................15 1.3. Developing voluntary control over the environment...................................................19 1.3.1. Development of cognitive control....................................................................20 1.3.2. Motor development.........................................................................................23 1.3.2.1. Development of motor skills ­ grasping.................................................24 1.3.2.2. Development of motor abilities- anticipatory planning...........................26 1.4. The possible relationship of anticipatory planning and cognitive control....................27 1.5. Motivation of the current research endeavor.............................................................29 Chapter 2- The end-state comfort effect in childhood- a systematic review...31 2.1. Introduction into the systematic review......................................................................32 2.2. Method......................................................................................................................34 2.3. Results......................................................................................................................38 2.3.1. Tasks used to investigate the ESC effect in children.......................................40 2.3.1.1. Bar-transport-task..................................................................................40 2.3.1.2. Handle-rotation-task..............................................................................41 2.3.1.3. Overturned-glass-task...........................................................................41 2.3.2. The ESC effect in children of various populations...........................................42 2.3.2.1. Studies on normally developing children................................................43 2.3.2.2. Studies comparing normally developing children and children with developmental disorders...................................................................................................55 2.4. Discussion................................................................................................................61

2.4.1. Age effects of ESC planning in different tasks.................................................62 2.4.2. Other factors constraining the ESC effect in children......................................65 2.4.2.1. Number of action steps..........................................................................66 2.4.2.2. Required degree of object rotation.........................................................67 2.4.2.3. Precision requirements..........................................................................68 2.4.2.4. Familiarity with the task.........................................................................69 2.4.2.5. Test procedure......................................................................................71 2.4.2.6. Motivation..............................................................................................72 2.4.2.7. Influence of habitual and goal directed factors.......................................73 2.4.2.8. Sample size...........................................................................................74 2.4.3. Parallels of cognitive and motor development.................................................75 2.5. Conclusion................................................................................................................77 2.6. Implementations........................................................................................................78 Chapter 3- Motor planning and executive functions across young ages.........81 3.1. Introduction...............................................................................................................82 3.2. Method......................................................................................................................84 3.2.1. Participants.....................................................................................................84 3.2.2. Tasks and Procedures....................................................................................84 3.2.2.1. Motor tasks used to assess ESC...........................................................86 3.2.2.2. Cognitive tasks used to assess EF........................................................93 3.2.3. Data Analysis/ Scoring...................................................................................99 3.3. Results......................................................................................................................99 3.3.1. The development of ESC................................................................................99 3.3.1.1. Bar-transport-task..................................................................................99 3.3.1.2. Sword-rotation-task.............................................................................103 3.3.1.3. Grasp-height-task................................................................................106 3.3.1.4. Intercorrelations of the ESC tasks.......................................................110 3.3.1.5. Summary.............................................................................................112 3.3.2. The development of EF.................................................................................112 3.3.2.1. The Tower-of-Hanoi-task.....................................................................112 3.3.2.2. The Mosaic-task..................................................................................113 3.3.2.3. The D2-attention-endurance-task........................................................114

3.3.2.4. Intercorrelations of the EF tasks..........................................................115 3.3.2.5. Summary.............................................................................................116 3.3.3. The relationship between ESC and EF..........................................................117 3.4. Discussion..............................................................................................................120 3.5. Conclusion..............................................................................................................123 Chapter 4- General Discussion..........................................................................125 4.1. Constraints on the development of anticipatory planning abilities............................128 4.1.1. Motor reorganization.....................................................................................128 4.1.2. Degree of object rotation...............................................................................130 4.1.3. Precision requirements..................................................................................131 4.1.4. Perceived comfort.........................................................................................132 4.1.5. Different kinds of learning effects..................................................................132 4.1.6. Directedness of the tasks..............................................................................135 4.2. Theory-driven assumptions.....................................................................................135 4.3. A new look at anticipatory behavioral control in children.........................................139 4.4. Criticism on the experimental study.........................................................................143 4.5. Directions of future studies......................................................................................146 4.6. Conclusion..............................................................................................................149 References and Supplements..............................................................................151 References.....................................................................................................................152 Supplements..................................................................................................................170 Statutory Declaration ..................................................................................................210 Eidesstattliche Versicherung.....................................................................................211 Acknowledgments ......................................................................................................212 Danksagung ..............................................................................................................214

Abbreviations(alphabetical) Abbreviations(alphabetical) ABC BRIEF CNS DOF EF e.g. ESC etc. i.e. IMP TEC TOH ZNA anticipative behavioral control Behavioral Rating Inventory of Executive Function central nervous system degrees of freedom executive functions exempli gratia(for example) end-state comfort et cetera(and so on) id est(that is) ideomotor principle theory of event coding Tower of Hanoi Zurich Neuromotor Assessment I

Register of Figures Register of Figures Figure 1. Different outcomes depending on start posture. A) An initially comfortable grasp results in an uncomfortable end-posture. B) Adjusting the grip at the beginning results in a comfortable end posture, enabling the person to pour coffee into the cup.................9 Figure 2. In the upper half, picture A) shows a person grasping the coffee box very low, in order to put it into a high shelf; in picture B), the person adjusts his/her grasp height, grasping the box on the top, when putting it into a lower shelf.......................................14 Figure 3. Proposed model of executive functions(adapted from Anderson, 2002, p. 73)......18 Figure 4. Projected developmental trajectories of the executive domains(adapted from Anderson, 2002, p. 78)..................................................................................................22 Figure 5. Schematic overview of the distinct steps to analyze the reviewed studies.............37 Figure 6. Schematic illustration of the original versions of the three tasks used to examine the ESC effect in children. A) Bar-transport-task(Rosenbaum et al., 1990) with initial position(1) and four possible end positions(2, 3). B) Handle-rotation-task (Rosenbaum et al., 1990). C) Overturned-glass-task(Fischman, 1997) with first turning action(1) and subsequent pouring action(2).....................................................42 Figure 7. Scatter diagram of the results of the three tasks used to assess ESC planning in normally developing children across all studies reviewed here......................................55 Figure 8. Apparatus used for the bar-transport-task. Left the standard start position (orientation of the bar was counterbalanced), right the different sizes of bars and target holes used for different ages...............................................................................87 Figure 9. Both pictures show a critical trial, where the white end of the bar had to be inserted into the target hole, and the possible outcomes. Upper part: An initially comfortable overhand posture results in a finally uncomfortable thumb-down posture. The child therefore does not behave in a manner consistent with ESC. Lower part: An initial grip adaption to a more uncomfortable underhand position results in a finally comfortable end position with the thumb pointing upwards............................................88 Figure 10. Apparatus for the sword-rotation-task. The sword is in the start position (Position 1). Positions are numbered clockwise.............................................................89 Figure 11. Both pictures show a critical trial(Position 2) and the possible outcomes. Upper part: An initially comfortable overhand posture results in a finally uncomfortable posture with the thumb pointing away from the blade. The child therefore does not behave in a manner consistent with ESC. Lower part: An initial grip adaption into a more uncomfortable underhand position results in a comfortable end position with the thumb pointing towards the blade..................................................................................90 Figure 12. Apparatus used for the grasp-height-task. The plunger is in the start position. Shelf heights were scaled to participants body height....................................................92 Figure 13. End-State Comfort in the grasp-height-task.........................................................93 Figure 14. Apparatus used for the Tower-of-Hanoi-task. The Tower shown on the picture needed to be re-built on the peg marked with the black tape.........................................95 II

Register of Figures Figure 15. Apparatus used for the Mosaic-task. The pattern displayed on the picture should be re-build with the several cubes......................................................................96 Figure 16. Set-up of the D2-task according to the IDS-P for children aged 3 to 5 years. Cards had to be sorted due to a specific pattern............................................................97 Figure 17. Set-up of the D2-task according to the IDS for children aged 5 to 10 years. Ducks had to be sketched out due to a specific pattern.................................................98 Figure 18. Set-up of the D2-task according to the D2R adults. All"p's" had to be sketched out due to a specific pattern...........................................................................................98 Figure 19. Percentage of the participants in each group showing end-state comfort in at least two out of three trials in the uncritical(initial overhand grip) and critical(initial underhand grip) conditions in the bar-transport-task....................................................100 Figure 20. Percentage of participants showing end-state comfort in each of the three critical trials in the bar-transport-task...........................................................................102 Figure 21. Percentage of participants in each group showing end-state comfort in the sword-rotation task in at least seven out of the twelve uncritical trials and in at least four out of the six critical trials......................................................................................103 Figure 22. Percentage of the participants showing end-state comfort in each of the three critical trials for Position 2(upper picture) and Position 3(lower picture), respectively. 105 Figure 23. Percentage of the participants showing end-state comfort in the grasp-heighttask for the high target position(light grey bars) and the low target position(dark grey bars). The light grey line indicates means of grasp height in both conditions; the dashed line shows the regression line.........................................................................106 Figure 24. Mean differences in grasp height averaged for the high-target-moves and the low-target-moves. Error bars represent confidence intervals for the distribution of grasp height.................................................................................................................108 Figure 25. Percentage of participants showing end-state comfort in each of the three trials for the high(upper picture) and the low target platform(lower picture)........................109 Figure 26. Scatter diagram of the results of the three tasks used to assess ESC planning in normally developing children in the experimental study...........................................110 Figure 27. Percentage of correctly solved items in the Tower-of-Hanoi-task. The shading of the bars indicates different versions of the test protocol used according to the age of participants..............................................................................................................113 Figure 28. Percentage of points reached depending on the total score in the Mosaic-task. The shading of the bars indicates different versions of the test protocol used according to the age of participants.............................................................................114 Figure 29. Percentage of points reached depending on the total score in the D2-task. The shading of the bars indicates different versions of the test protocol used according to the age of participants.................................................................................................115 Figure 30. Three-phase adaptation of the ABC theory by Hoffmann(2009), adjusted to account for children's performance in ESC tasks. In Phase 1, children automatically select a default grasp(thumb-up grasp), as they are not able to anticipate other III

Register of Figures effects. As the real effect does not match the effect anticipation, no action-effect associations are formed and no contextualization to the situational condition takes place. In Phase 2, children are able to anticipate different action outcomes. Now, the real effect matches the effect anticipation and action-effect anticipations are formed (i.e. initial thumb-down grasp results in final thumb-up posture after rotation), but these are not yet contextualized to the situational condition. In Phase 3, children(and adults) are able to precisely anticipate desired action effects, based on strong actioneffect associations. As the real effect reliably matches the effect anticipation, the action-effect association is now contextualized to the situational condition(i.e. inverted cup)................................................................................................................142 Figure 31. Example for an item-protocol of the Tower-of-Hanoi-task..................................193 Figure 32. Example for correct ducks in the IDS 3-5. All ducks with a yellow snail need to be sorted out, no matter if the sun is shining or not......................................................200 IV

Register of Tables Register of Tables Table 1. Bower's(1977) Phase 1 and 2 Reaching and Grasping Behavior Characteristics (adapted from Payne& Isaacs, 1999, p. 230)................................................................25 Table 2. Overview of the mean percentage of ESC behavior found in the different studies..39 Table 3. Demographic overview of the sample....................................................................85 Table 4. Pairwise Pe arson's correlations between the three motor tasks. Note that two correlations for the adult sample could not be computed due to ceiling effects in the sword-rotation-task......................................................................................................111 Table 5. Pairwise Pearson's correlations between the cognitive and the motor tasks. ........119 V

Summary Summary The end-state comfort(ESC) effect is an important aspect of anticipatory behavioral control. It reflects a person's strategy to avoid uncomfortable body positions at the end of movements. As the focus of previous studies primarily laid on young adults, there are only few studies on the ESC effect in children, which show divergent findings. By means of the systematic review(Chapter 2), possible reasons for these inconsistent findings were provided(e.g. age effects, the number of action-steps, precision requirements, or task differences). One assumption provided in the systematic review was examined in Chapter 3. This assumption implied that motor development relies on the development of cognitive control, mainly on the development of executive functions. Therefore, a test battery was designed, consisting of three motor tasks to measure ESC and three cognitive tasks to measure executive functions. Nevertheless, results were not able to approve the assumption. An important finding was that the performance in the different motor and cognitive tasks were not related to each other, suggesting an interindividually different developmental trajectory for each of them. The focus of future studies should rely on the examination of potential constraints on ESC planning, like those outlined in Chapter 2, which possibly influenced the developmental trajectories of the ESC effect in childhood and caused the inconsistent findings in the studies reviewed. Moreover, causes should be detected for the fact, that the tasks used in Chapter 3 were not related to each other. Another focus should be on the influence of other executive functions, like inhibition, on the development of ESC. VI

Zusammenfassung Zusammenfassung Der End-State Comfort(ESC) Effekt als Aspekt antizipativer Verhaltenskontrolle spiegelt die Strategie einer Person wider, unangenehme Körperpositionen am Bewegungsende zu vermeiden. Bisher existieren nur wenige Studien zum ESC Effekt im Kindesalter, welche divergente Ergebnisse aufweisen. Mit Hilfe eines systematischen Reviews(Kapitel 2) konnten als mögliche Ursachen für diese inkonsistenten Befunde z.B. Alterseffekte, die Anzahl an nötigen Handlungsschritten, der Grad an erforderlicher Präzision, oder Unterschiede im Untersuchungsablauf, deklariert werden. In Kapitel 3 wurde eine weitere Annahme geprüft, welche besagt, dass die motorische Entwicklung auf der Entwicklung kognitiver(bzw. exekutiver) Funktionen beruht. Mit Hilfe einer Testbatterie, bestehend aus je drei Aufgaben zur Messung antizipativer Handlungskontrolle und zur Messung exekutiver Funktionen, konnte diese Annahme jedoch nicht bestätigt werden. Interessant ist außerdem, dass die unterschiedlichen Tests innerhalb der einzelnen Altersgruppen ebenfalls nicht miteinander zusammenzuhängen scheinen, was für eine interindividuell unterschiedliche Entwicklung der einzelnen Fähigkeiten spricht. Der Fokus zukünftiger Studien sollte daher in der Erforschung der in Kapitel 2 dargestellten Ursachen für die inkonsistenten Befunde liegen. Weiterhin sollten Ursachen dafür gefunden werden, dass die in Kapitel 3 verwendeten Sub-Tests keine Korrelationen aufweisen. Ein weiterer Fokus sollte auf dem möglichen Einfluss weiterer kognitiver Funktionen, wie z.B. der Inhibition, liegen. VII

Chapter 1 Chapter 1 1. General introduction 1

General introduction Motor behavior is essential in p eople's everyday lives. In the first minutes of every morning, people implement a countless number of actions. First, one will wake up, open his/her eyes, before one will stretch him-/herself. Then sit up in bed, take out the feet, get up, walk into the kitchen in order to have a coffee, and so on. Obviously, one could continue this enumeration endlessly. Grasping is one of the basic motor skills everyone performs within these daily life activities. It is an observable outcome of motor behavior, which helps to attain action goals. Therefore, simply the action of drinking a coffee in the morning involves several grasping actions. Being in the kitchen, one needs to grasp the handle of the cabinet to open it, grasp a cup from its inside and place it under the machine, grasp the box with the coffee filters to put one into the machine, grasp the coffee box to put some coffee into the filter, press the button, grasp the handle of the refrigerator to open it, grasp for the milk to pour it into the cup, and so on. As one can see, there are a lot of grasping actions involved solely in the first few minutes of a day. Usually, people do not really"think" about what they do or how they grasp, this behavior is like an automatic mechanism in response to a stimulus. There are many ways to grasp the cup and to take it out of the cabinet, and this variety of grasping actions is accounted for in the degrees of freedom problem. Independent of the action that has to be performed, the process of response selection implies the problem that a person involved in this process has to decide for one particular movement or a combination of movements although he or she has many alternative solutions. To get the cup out of the cabinet, no matter in which position it is in there, there are a lot of different solutions, including different hand orientations, movement trajectories, or grip strategies, for instance. This issue is traced back to what Bernstein(1967) referred to as the degrees of freedom(DOF) problem. Rosenbaum(2010, p. 12) gives an example for this problem, namely to"touch the tip of your nose". As in the action of grasping the cup, there are many different ways to tip your own nose. Solely for the movement of the arm there are seven different possibil2

General introduction ities: three DOF in the shoulder, two in the elbow and two in the hand. The combination of these DOF lead to a great number of possible reaching strategies depending on arm position(Rosenbaum, Meulenbroek,& Vaughan, 1996). Taking the fingers into account makes this number even higher. As explained in the following paragraphs, the selection of a movement is due to the anticipation of action goals and to constraints concerning movement economy. Though the focus of the present dissertation thesis is on the acquisition of motor(and cognitive) skills, it is of interest, how the ability of such grasping actions mentioned above emerges. Therefore, a short overview of motor learning theories will be given in the next paragraphs. Motor learning is described as an experience dependent, lasting change of the competence to reach desired effects in different situations through a certain behavior(Mechling& Munzert, 2003, p. 133). In this definition, motor actions(responses) are related with situational conditions(stimuli) on the one hand and with situational modifications(effects) on the other hand. Therefore, motor learning theories can be divided into theories with a focus on stimulus-response associations, or with a focus on action-effect associations. 1.1. From intentions to actions ­ Goal-driven motor behavior Ideomotor approaches state that actions start with intentions, and movements are generated to realize those intentions(Keller et al., 2006). Accordingly, the cause of actions does not only rely on a stimulus(i.e. the cup standing in the cabinet), but on the to-be-attained goal(i.e. take the cup out to have a coffee; Kunde, Elsner,& Kiesel, 2007). The basis for grasping the cup is the visuo-spatial perception of its location and orientation. Where is it in the cabinet, and in which position? Modern theories assume that action and perception are closely related to each other(e.g. the perception-action theory, Gibson, 1966; the dynamic systems theory, Thelen,& Smith, 1994; or the theory of neuronal group selection, Edelman, 1987). 3

General introduction To form an intention to reach any goal, an anticipation of the desired goal is needed(Gollwitzer, Fujita,& Oettingen, 2004). Think about the person standing in front of the cabinet, ready to take out the cup. Surely, this person has often performed and watched others performing the act of taking out the cup. Therefore, the knowledge of this action(or the visualization in a persons' mind's eye) can produce the same effects as the actual production of this action. Considering two people sitting at a table, facing each other, with a cup and a jar of coffee on it, and one of these two people picks up the jar and pours coffee into the cup while the other one watches, then both people would have a very similar representation of the action according to the common coding theory(Hommel, Muesseler, Aschersleben,& Prinz, 2001; Prinz, 1990, 1997). The core postulate of this theory is that perception and action rely on common codes. These codes do not represent actions per se, but instead reflect the perceptual events that these actions produce. Because actions are coded in terms of their perceptual effects, perception and action occupy a common representational domain. Given these information, and think about only one person sitting at the morning table with the cup and the jar of coffee in front. It is likely, that this person has former information of perceived outcomes at his/her disposal. Solely these information can produce the same effect, which would be determined as anticipation then. This person has often seen the action of pouring coffee into the cup and drinking it, and has very often conducted this action him-/herself. Therefore, the knowledge of this action(or the visualization in his/hers min d's eye) can produce the same effects as the actual production of this action. This means, that an action is represented like a perceived action, which is stated in the theory of event coding(TEC; Hommel et al., 2001). This assumption is in the tradition of ideomotor approaches(Lotze, 1852; Münsterberg, 1888; James, 1890; for an overview see Stock& Stock, 2004). 4

General introduction According to the ideomotor principle(IMP; James, 1890), movements of the body become connected with their sensory consequences in a way that the mere image of such consequences receives the power to trigger those movements, which formerly brought them about. In other words, body movements become determined by anticipations of their own sensory consequences, and therefore, it is a premise that one can experience his/her motor abilities only indirectly through perception of achievable sensory effects, which accompany or follow the execution of actions. To this end, people initially have to learn about the sensory effects, which can be produced through motor actions(Elsner& Hommel, 2001). With the integration of motor programs and action effects, a functional unit is built(sensorimotor units). These units represent actions and provide a person cognitive access to voluntary actions: The motor program can be activated through the anticipation of the desired consequences. Therefore, motor activity relies on intentional control. But what exactly do "motor programs" account for? Keele(1968) introduced the term"motor programs" for executive action stru ctures. He defines a motor program as"a set of muscle commands that are structure d before a movement sequence begins, and that allows the entire sequence to be carried out uninfluenced by peripheral feedback"( Keele, 1968, p. 387). In this sense, motor programs involve entire actions, which are programmed previously. This assumption is supported by the fact that simple reaction time will increase when the response movement is of greater complexity(Henry& Rogers, 1960). Therefore, the t heory proposed a"non conscious mechanism that uses stored information(motor memory) to channel existing nervous impulses from brain waves and general afferent stimuli into the appropriate neuromotor coordination centers, subcenters, and efferent nerves [... ] "(Henry& Rogers, 1960, p. 448). Furthermore, it is supported by the observation of anticipatory effects. Regularly, one can observe evidence for the next action step clearly before the subsequent action step is completed. Remember the 5

General introduction example of grasping a cup from the cabinet: If one would register the trajectory of the grasping action, then it would be observable that the to-be-grasped object(here: the cup) is already represented in one's grip strategy, like in the distance of the fingers, for example, which would be greater if one would like to grasp a bigger object(i.e. a coffee cup) instead of a smaller one(i.e. an espresso cup)(Jeannerod, 1981), or in the speed of the movement(Marteniuk, MacKenzie, Jeannerod, Athenes,& Dugas, 1978), which will be lower if the grasping action requires great precision. The grasping hand therefore anticipates the object to-be-grasped, which assumes that the grasping of the object is programmed in advance before it is actually reached. The same can be observed after the grasping action: Previous studies have demonstrated that what an adult is going to do with an object after picking it up affects not only the posture of the fingers, but also the kinematics of the reach toward the object (Marteniuk et al., 1987). To this end, motor behavior needs anticipation. The anticipation of movements is always about predicting the future, or expecting a future event, or imagining a future state or event and can be defined as" a process, or behavior, that does not only depend on past and present, but also on predictions, expectations, or beliefs about the future"( Butz, Sigaud,& Gérard, 2003, p. 4; Hoffmann, 2003, p. 3). To approach the questions about when and where anticipatory behavior is useful, Hoffmann formulated the anticipative behavioral control theory(ABC theory, Hoffmann 1993, 2003; Hoffmann, Stöcker,& Kunde, 2004; Hoffmann et al., 2007), which will be derived below. Most of the behavior in our everyday lives is goal directed. According to the IMP, human behavior is not triggered by stimuli, but by to-be-produced effects. Therefore, behavioral acts have to be connected to the effects they produce; otherwise it is impossible to see how an anticipated effect can affect our behavior. In contrast to behaviorism, the ABC theory assumes that behavioral competence relies 6

General introduction predominantly on the acquisition of action-effect associations rather than stimulusresponse associations(Hoffmann 1993, 2003; Hoffmann et al., 2004; Hoffmann et al., 2007). It is stated, that any voluntary action(like the goal-directed grasping for the cup) is preceded by corresponding effect anticipations. This assumption is in accordance with the IMP mentioned above, which states that body movements become determined by anticipations of their own sensory consequences, and therefore, it is a premise that one can experience his/her motor abilities through the perception of achievable sensory effects. After the performance of the action, like grasping the cup and taking it out of the cabinet, the actual effect is compared to the anticipated effect. This comparison is supported by the principle of reafference(von Holst& Mittelstädt, 1950). This principle denotes a prescribed process in the central nervous system(CNS) for the control of and response to a stimulus. If an efferent signal, which leads to an action, is sent from a superordinate nervous center(motor cortex), then an efferent copy is produced in a more subordinate nervous center(cerebellum). This copy is an internal copy of an efferent, movement-producing signal generated by the motor system. It is then compared with the reafferent sensory input that results from the movement, enabling a comparison of the actual movement with the desired movement, and a shielding of perception from particular self-induced effects on the sensory input to achieve perceptual stability. Together with internal models (Wolpert, Ghahramani,& Jordan, 1995), efference copies can serve to enable the brain to predict the effects of an action. So, if there is sufficient coincidence between what was desired and what really happened(i.e. holding the cup in a comfortable position to pour the coffee), then representations of the action and experienced effects become interlinked, or an already existing link is strengthened(James, 1890). The anticipation of effects is therefore represented as being appropriate to produce a desired action, taking the situational context into account. To sum up, the ABC theory takes the primacy of action-effect learning, as well as the conditionalization of actioneffect relations into account, meaning that the same action can have different effects 7

General introduction depending on environmental circumstances((Hoffmann 1993, 2003; Hoffmann et al., 2004; Hoffmann et al., 2007). Such anticipatory behavior can be observed in a lot of different everyday-life situations. Each time we switch on the radio, we anticipate that it will play some music, or if we press a door handle, we anticipate the door to open. Consider the example above, with the cup standing in the cabinet, it can be placed either with the opening facing upwards or downwards. Grasping the cup with the thumb pointing upwards will end in different end orientations, depending on the initial orientation of the cup. Thus, before reaching for the cup, we should anticipate the outcome of our chosen grasp, depending on the action to-be performed(i.e. pouring coffee into it). This assumption is accounted for in the end-state comfort(ESC) effect. Besides other factors(like the kinematics of the action, speed, or grip force; Marteniuk et al., 1987), the ESC effect is one possible measure, which signifies anticipatory planning in grasping actions. 1.1.1. The End-State Comfort Effect When considering the grasping action described above, it is apparent that the control of this behavior is not haphazardly, but depends on different underlying mechanisms. Depending on the position of the cup in the cabinet(upright or upsidedown), a thumb-up grip would result in two different grasping postures after the cup was taken out of the cabinet(in the same way, if the cup was standing upside-down, a thumb-up and a thumb-down grip would result in different outcomes). These options are depicted in Figure 1 below. 8

General introduction Figure 1. Different outcomes depending on start posture. A) An initially comfortable grasp results in an uncomfortable end-posture. B) Adjusting the grip at the beginning results in a comfortable end posture, enabling the person to pour coffee into the cup. In Figure 1A), grasping the upside-down standing cup with a comfortable thumb-up grasp results in a very uncomfortable thumb-down position after taking it out of the cabinet, making it difficult to pour the coffee into it. But, when grasping with an initially uncomfortable grasp with the thumb pointing downwards(Figure 1B), a comfortable end position is reached, in which the coffee can easily be poured into the 9

General introduction cup. This phenomenon, that people tend to avoid uncomfortable positions at the end of goal-directed movements, is called the ESC effect(Rosenbaum et al., 1990). Besides the above mentioned examples, this effect signifies anticipatory planning skills in many everyday-life tasks. For the first time, the ESC effect was observed in a restaurant by David Rosenbaum. While he sat at the bar, he observed a waiter taking glasses out of a shelf, which were standing upside-down, as the cup in the example above. He noticed, that the waiter always used the initially uncomfortable thumb-down grasp(as depicted in Figure 1A) in order to put the glasses comfortably down onto the bar. Rosenbaum and his colleagues(1990) took this observation into laboratory. They designed the so called bar-transport-task, in which a horizontally oriented bar with one black and one grey end laid on two supports. This bar had to be placed on either a red or a blue target disc, which was placed to the right and to the left of the supports, respectively. Thereby, it was of interest, how participants reached for the bar, depending on the intended target location and position of the bar. Results revealed that right-handed participants chose a comfortable overhand-grip to place the right end of the dowel on the target(irrespective of bar and target color), and that they used an initially uncomfortable underhand-grip when the left end of the bar had to be placed on the target. The flexible selection of the initial grasp type allowed participants to end the object manipulation in a more comfortable thumb-up posture(as opposed to an awkward thumb-down posture), even if this meant to tolerate an awkward posture at the beginning of the action. The ESC effect provides an efficiency constraint on motor planning(Rosenbaum, van Heughten,& Caldwell, 1996). The cognitive mechanisms underlying the effect have been sketched out in the posture-based planning account(Rosenbaum et al., 2007). According to this account, a set of stored postures is scrutinized for their usability in the upcoming action. Then, the best candidate stored posture is optimized 10

General introduction to attain a better goal posture. Once a goal posture is selected, people form a specific movement to attain this goal posture. If an internal simulation of the planned movement shows any obstruction, the planned movement is reshaped by superimposing another, expedient movement on it. This movement is made from the starting posture to a planned"bounce posture", and back to the starting posture again. If a movement is made in this way, from the start posture to the"bounce posture" and back, while the main movement is made from the start posture to the goal posture, the combined movement can have a shape that depends on all, the start posture, the bounce posture, and the goal posture account(Rosenbaum et al., 2007). Several hypotheses have been raised to explain the ESC effect. Besides the "working backwards hypothesis" and the"fatigue hypothesis"(Rosenbaum et al., 1990), as well as the"minimizing time in awkwardness postures hypothesis" and the "exploiting potential energy hypothesis"(Houk& Rymer, 1981; Rosenbaum& Jorgensen, 1992; for an overview see Short& Cauraugh, 1999 ), the"precision h ypothesis" is one possible explanation for the ESC effect(Rosenbaum, Vaughan, Jorgensen, Barnes,& Stewart, 1993). This hypothesis predicts that goal-directed positioning movements at or near the middle of the range of motion is easier than at or near the extremes and that precision increases, while being in a comfortable posture, and movements can be made more quickly within that range(Rosenbaum et al., 1996). The general prediction, which Rosenbaum and colleagues(1993) evidenced in their study, is that the likelihood of people showing the ESC effect is inversely related to the precision requirement of the task. In a study conducted by Short and Cauraugh(1999), this prediction was examined. Participants had to point a horizontally suspended bar with one black and one white end onto one of fourteen different targets on a wall, which differed in height and in target width, resulting in a low precision condition and a high precision condition. The authors found that ESC(depending on awkwardness ratings) increased 11

General introduction with increased precision requirements. Another study on the precision hypothesis conducted by Hughes, Seegelke, and Schack(2012) investigated the influence of precision demands at the start and at the end of movements. Here, participants had to reach for a bar, which was standing in a start-disc, and had to move it to the target disc with either hand. Precision requirements at the start and at the end of the movement were either identical(low-low, high-high) or different(low-high, high-low). Participants showed more ESC planning if the precision requirement at the end of the movement was high. Since its discovery two decades ago(Rosenbaum et al., 1990), a growing body of research has shown the significance of the ESC effect on motor planning in healthy adults across a variety of either unimanual(e.g. Cohen& Rosenbaum, 2004; Rosenbaum& Jorgensen, 1992; Rosenbaum et al., 1990, 1996; Weigelt, Cohen,& Rosenbaum, 2007) or bimanual tasks(e.g. Fischman, Stodden,& Lehman, 2003; Janssen, Crajé, Weigelt,& Steenbergen, 2010; Weigelt, Kunde,& Prinz, 2006; van der Wel& Rosenbaum, 2010). It has also been investigated within different clinical populations, such as autism(Hughes, 1996), cerebral palsy(Crajé, van Elk, van Schie, Bekkering,& Steenbergen, 2010; Mutsaarts, Steenbergen,& Bekkering, 2006; Steenbergen, van Nimwegen,& Crajé, 2007), apraxia(Goldenberg& Hagmann, 1998; Randerath, Li, Goldenberg,& Hermsdörfer, 2009) or developmental coordination disorder(Smyth& Mason, 1997; van Swieten et al., 2010). Moreover, the ESC effect seems to affect action planning(even) in nonhuman animals, such as lemurs (Chapman, Weiss,& Rosenbaum, 2010), cotton-top tamarins(Weiss, Wark,& Rosenbaum, 2007), as well as in squirrel monkeys and tufted capuchins(Zander, Weiss, & Judge, 2013), and in chimpanzees(Frey& Povinelli, 2011). Besides these more general demonstrations of the ESC effect, recent studies started to focus on factors that modulate it, such as habitual vs. goal-directed factors(Herbort& Butz, 2011) or the degree of object rotation required(Hughes, Reißig,& Seegelke, 2011; Hughes et al., 2012; Seegelke, Hughes,& Schack, 2011). A summary of the findings pertaining 12

General introduction to the ESC effect is provided in a recent review article by Rosenbaum and colleagues (Rosenbaum, Chapman, Weigelt, Weiss,& van der Wel, 2012). However, as mentioned above, the ESC effect is not the only phenomenon to examine anticipatory planning skills. Besides ESC, the grasp-height effect is another feature to measure anticipatory planning, which will be addressed in the next section. 1.1.2. The grasp-height effect Given the example of the coffee cup above, showing that people tend to avoid uncomfortable final postures at the end of goal-directed movements, such an anticipation of future body postures can also be observed when this person puts the coffee box back into the kitchen shelf. When placing the box on a high shelf, one would grasp the box very low on its side and very high when it is brought to a low shelf in order to end the action as comfortable as possible, without the need to stretch oneself or to crouch down. Therefore, the future orientation of an object, which should be placed onto targets of varying heights, in order to end the action near the middle of their possible range of movement, is also anticipated. This so called " grasp-height effect " (Cohen& Rosenbaum, 2004), depicted in Figure 2, describes the tendency of people to choose a designated grasp-height on to-be manipulated objects, in order to end the movement in a most comfortable position. Cohen and Rosenbaum(2004) investigated the grasp-height effect, asking their participants to grasp a plunger standing on the middle platform of a shelf and to bring it to another platform(a higher one, a lower one or one right beside the starting platform), and to bring it to the home platform back again. The authors found an inverse linear relation between target height and grasp height, revealing that participants grasped the plunger near its base, when transporting it to a high platform, and near the top, when transporting it to a low platform. But when participants returned the plunger from the target position to the home position, they did not fully display the 13

General introduction grasp-height effect, as they grasped the plunger close to where they had grasped it before. Figure 2. In the upper half, picture A) shows a person grasping the coffee box very low, in order to put it into a high shelf; in picture B), the person adjusts his/her grasp height, grasping the box on the top, when putting it into a lower shelf. 14

General introduction Rosenbaum, Halloran, and Cohen(2006) raised the precision requirements for this task by adding rings of varying diameters to the home platform and to the target platforms. Results showed that when more precision is needed(independent of the start position or the target position), the grasp-height effect was reduced and the grasp heights were lowered. Therefore, the grasp height is chosen to promote control, and not just comfort, but also reflects effects of memory, as was shown when the home platform, rather than the target platforms, was varied in height. Here, the grasp heights for the return moves were close to the position for the home-to-target moves and therefore, the target-to-home moves diverged in grasp-height from the mean grasp height and did not converge toward the mean grasp height. Participants in the study of Cohen and Rosenbaum(2004) may have relied on recall to choose grasp heights for target-to-home moves. Weigelt and colleagues(2007) examined whether these recall strategies rely on postures(full body positions) or on locations(where the object was grasped). These authors used the same task as described above, with the difference, that participants had to take a sidestep after the home-to-target move, and to step either onto a platform, down from a platform or just horizontally. Results showed that participants grasped the plunger close to where they had grasped it in the home-to-target move, relative to the base of the plunger and therefore, used extrinsic rather than intrinsic coordinates for movement execution in these moves. These results can be explained by minimizing the DOF(see Chapter 1). The number of configurations(DOF) to-be stored when remembering a location is less than the number of configurations to-be stored in remembering a configuration of body postures. 1.2. The role of cognitive control in motor behavior Motor behavior is a field of study, which is commonly examined in behavioral research. But only little attention is usually paid to the(cognitive) processes underlying these behavioral outcomes. One domain in which these links between cognition 15

General introduction and action have been explored is the manipulation of objects. The results of studies on motor planning and the ESC effect mentioned above indicate that humans, as well as nonhuman animals, grasp objects in a way that reflects their intentions. The way, in which people(or animals) grasp an object, depends on what they plan to do with this object and on the position at which they grasp an object depends on the height where the object is planned to be positioned. Therefore, grasping actions are all about planning, and planning per se relies on cognitive control, which takes a multiplicity of factors into account. The fact, that cognitive control needs to be an underlying factor of efficient planning, is sustained by clinical studies, for example in participants, which suffer from cerebral palsy. Here, research showed that some of their motor problems do not only relate to movement execution, but also to motor planning (Mutsaarts et al., 2006). Furthermore, concomitance of object manipulation and memory tasks showed a reduced recall ability, which suggests that planning for grasping objects needs cognitive resources(e.g. Spiegel, Koester, Weigelt,& Schack, 2012; Weigelt, Rosenbaum, Huelshorst,& Schack, 2009; Logan& Fischman, 2011). The most important domain of cognition, which might be the driving force for anticipatory planning, is the domain of executive functions(EF). This assumption is made in recent literature(Rosenbaum et al., 2012; van Swieten et al., 2010) and can also be drawn from clinical research, as from cases of brain damage, like cerebral palsy. Here, motor and executive functions are impaired, which suggests a(anatomically) link between both functions(e.g. Mutsaarts et al., 2006; Steenbergen& Van der Kamp, 2004). When looking for a generally applicable definition of EF, one will be stretched very soon to one's limits, as the re is no common definition. But what they all coincide about is that EF is an umbrella term, incorporating a collection of inter-related processes responsible for purposeful, goal-directed behavior(Gioia, Isquith,& Guy, 2001). They refer to higher order, self-regulatory, cognitive processes that operate the monitoring and control of thought and action, and are often described as control 16

General introduction processes(Monsell& Driver, 2000). Processes associated with EF are numerous, but the principle elements include anticipation, goal selection, planning, inhibitory control, attention, attentional and mental flexibility, error correction and detection, the utilization of feedback, and resistance to interference(Dempster, 1992; Welsh, Pennington,& Groisser, 1992; Zelazo, Carter, Reznik,& Frye, 1997). Besides these processes, the executive system organizes learning processes and strategies for problem solving. Therefore, it is actively involved in conceptual thinking processes, decision making, and in the maintenance of anticipatory planning processes, as well as in the implementation of action, the chronological structuring of actions, and in error detection and correction(Andreasen, 2002; Baddeley& Della Salla, 2003; Carlson, 2003; Roberts, 2003). Today, there are several different frameworks trying to account for EF. These have conceptualized EF either as a construct, in which EF are a kind of central executive responsible for multi-modal processing, or as multiple process related systems, that are interrelated, interdependent, and function together as an integrated supervisory control system(Stuss, 2011; Stuss& Alexander, 2000). Results from studies using different EF tasks revealed that they all can be loaded onto three to four factors (Kelly, 2000; Levin et al., 1991; Welsh, Pennington,& Groisser, 1991). Based on these results, Anderson(2000) proposed a model of EF, which describes EF as four distinct domains: 1) attentional control, 2) information processing, 3) cognitive flexibility, and 4) goal setting. He states that these functions work in an integrative manner, in order to execute certain tasks, and that they can be conceptualized as an overall control system when taken together. Anderson's model is depicted in Figure 3. 17

General introduction Figure 3. Proposed model of executive functions(adapted from Anderson, 2002, p. 73). The attentional control unit includes the capacity to selectively attend to specific stimuli and inhibit proponent responses, as well as the ability to focus attention for a long period of time. Furthermore, the regulation and monitoring of actions are included, so that plans can be executed as they were planned, errors can be identified, and goals can be achieved. Impairments relate to people being impulsive, having a lack of self-control, failing to complete tasks, committing procedural mistakes and not correcting them, and responding inappropriately. The information processing domain refers to fluency, efficiency, and the speed of the output. Here, the integrity of neural connections and the functional integration of frontal systems are displayed. Information processing can be measured by the speed, the quantity, and the quality of output. Deficits display itself in reduced output, false responses, hesitancy, and slowed reaction times. Cognitive flexibility refers to the ability to shift between responses, to learn from mistakes, to invent alternative strategies, to divide attention, and to process multiple sources of information concurrently. Here, working memory is also a part of 18

General introduction cognitive flexibility. Inflexible people are generally considered to be rigid and ritualistic, struggling when activities or procedures are changed and failing to adapt to new demands. Impairment in this domain of cognitive flexibility is often associated with preservative behavior, when people continue to make the same mistakes or breaking the same rules again and again. Goal setting relies on the ability to develop new initiatives and concepts, as well as the capacity to plan actions in advance and approach tasks in an efficient and strategic manner. Impairments show up in poor problem solving abilities, disorganization, difficulties in developing efficient strategies, reliance on previously learned strategies, and poor conceptual reasoning(Anderson, 2002). Depending on former studies it is argued, that EF are activated in novel or complex tasks, as they require the individual to formulate new plans and strategies to monitor their effectiveness, whereas in simple or routinized tasks, people perform instinctively, relying on plans that already exist(Shallice, 1990). As studies found different clinical populations to have reduced EF(especially patients with frontal, parietal, and cerebella lobe impairments), it has been assumed, that EF rely on different brain regions(see Nowrangi, Lyetsos,& Munro, 2014, for a review). Having the principles of motor planning and cognitive control in mind, the question remains, how and when these abilities develop. Former studies suggest, that both develop during sensory-motor maturation(von Hofsten, 2004; Stuss, 1992), until they are fully matured in adulthood. In the following Chapter, light will be shed on the development of both, cognitive and motor development during childhood. 1.3. Developing voluntary control over the environment When screening the literature on child development, a bulk of research is found, addressing many aspects of development, like social development(e.g. Ryan & Deci, 2000), emotional development(e.g. Grusec, 2011), and language develop19

General introduction ment(e.g. Cole, Armstrong,& Pemberton, 2010). Besides these aspects, motor development(e.g. Blauw-Hospers,& Hadders-Algra, 2005) and cognitive development (Fischer, 1980) play an important role in our field of studies. Theorists have proposed different theories in the development of each field, but development is not uniform, and does not influence only one of these components at a time. Therefore, it is difficult to indicate theories on motor and cognitive development separately; every theory takes both parts into account, but with different weights. Therefore, theories in the following section will be divided due to their main area of interest. 1.3.1. Development of cognitive control Plenty of different theories account for the development of cognitive control. Based on the assumptions made by these theories(e.g. Piaget, 1936; Vygotsky, 1977), executive functions(EF)-as an area of cognitive control- develop at different rates over the course of development(Anderson, Anderson, Northam, Jacobs,& Catroppa, 2001; Stuss, 1992). Some abilities start to develop very early in childhood, whereas others do in late childhood or adolescence. EF are shaped to life experiences, as assumed by cognitive theories, as well as due to maturation of the brain with building up connections until adulthood. Their development corresponds to the neurophysiological development of brain regions, as when the processing capacity of the frontal lobes and other brain regions increases, the core EF emerge(Anderson, 2002, 2008). This development may not progress in a linear way, but might demonstrate different developmental trajectories, which may occur in spurts. Suggestions, that EF may not operate until the brain reaches maturity in the second decade of life, have been refuted(Golden, 1981), as neuroimaging studies have demonstrated prefrontal activation in infancy(Bell& Fox, 1992). Neuropsychological studies found a developmental trajectory for EF across childhood(Becker, Isaac,& Hynd, 1987; Levin et al., 1991; Passler, Isaac,& Hynd, 1985; Welsh, Pennington,& Groisser, 1991). 20

General introduction Dennis(1989) provided a model of EF development, which he divided into three stages: 1) emerging(early stage of acquisition and not functional), 2) developing(capacity is partially acquired, but not fully functional), and 3) established(ability fully matured). Only the functional skills are measureable using EF tasks, meaning that in the emerging s tage no function can be detected, even if it is already"in pr ogress". When looking back at Figure 3, it is known that cognitive flexibility(in the upper left of Figure 3) develops in early and middle childhood until adolescence. Therefore, preservative behavior is commonly found in infancy and declines upon development until it is rarely found in adulthood(Chelune& Baer, 1986, Levin et al., 1991). Between three and four years of age, children start to be able to switch between two simple response sets, and are able to switch in multi-dimensional tasks between ages 7 and 9(Anderson et al., 2000). In early childhood, the ability to learn from mistakes and to provide other strategies emerges and develops throughout middle childhood. In terms of goal setting(upper right of Figure 3), children as old as four years exhibit simple planning skills(Welsh, Pennington,& Groisser, 1991) and can generate new concepts in terms of conceptual reasoning(Jaques& Zelazo, 2001). Skills of planning and organization develop rapidly between 7 and 10 years of age(Anderson, Anderson,& Lajoie, 1996). Whereas young children often use simple strategies, which are usually inefficient, older children between 7 and 11 years show more strategic behaviors and reasoning abilities(Anderson, Anderson,& Garth, 2001; Levin et al., 1991). This refinement of strategies and improved decision making continues during adolescence(Anderson et al., 2001; Levin et al., 1991). In the attentional control domain(in the lower left part of Figure 3), infants as old as 12 months of age begin to inhibit previously learned behavior and shift to new responses(Diamond, 1985; Diamond& Doar, 1989). At the age of three years, children can inhibit instinctive behaviors well, although they continue to make preservative errors(Diamond& Taylor, 1996). Up to six years of age, improvements in speed and accuracy can be observed in impulse control tasks(Diamond& Taylor, 1996). Children from 9 years and older 21

General introduction can regulate their actions well and show inhibitory behavior(Anderson et al., 2000). In information processing(lower right part of Figure 3), increments in response speed and verbal fluency can be observed between three and five years of age(Welsh, Pennington,& Groisser, 1991), and processing speed continues to improve between 9 and 12 years(Kail, 1986). Efficiency and fluency development is further delayed into adolescence(Anderson et al., 2001; Kail, 1986), although increments are minimal after the age of 15. Based on these findings, Anderson(2002) proposed different developmental trajectories for these domains, which are illustrated in Figure 4. Figure 4. Projected developmental trajectories of the executive domains(adapted from Anderson, 2002, p. 78). With regard to the developmental trends suggested, attention control would be the first EF to have matured to more than 80% at about five years of age, followed by information processing abilities, which are mostly matured by the age of 8 years, and cognitive flexibility, which is only little protracted in development(at about 9 years). Last, goal setting abilities develop, which are almost matured at about 11 years. All 22

General introduction abilities, however, develop further after these ages during adolescence, until they are fully matured in adulthood(Anderson, 2002). If EF do not develop regularly, this is called executive dysfunction(EDF). This disorder represents deficits in one or more areas of EF and a variety of presentations is possible. In children, EDF is observable through poor impulse control, difficulties in monitoring or regulating motor performance, a poor reasoning ability, mental inflexibility, and feedback utilization, difficulties in generating or implementing strategies to achieve any goal, and reduced working memory(Anderson, Bechara, Damasio, Tranel,& Damasio, 1999; Barrash, Tranel,& Anderson, 2000; Eslinger& Damasio, 1985; Grattan& Eslinger, 1991). Besides cognitive development, motor development also plays an important role in the maturation of children. Therefore, theories with a focus on motor development will be introduced below. 1.3.2. Motor development The traditional study of motor behavior was characterized by observations of changes in different motor actions like grasping, rolling, crawling, climbing, and walking(Heriza, 1991). Today, a neuromaturational theory of development is prevalent, based on the assumptions made by Gesell(1933; Gesell& Thompson, 1934), describing normative timetables for motor achievements, and by McGraw(1935), who examined the determinants of these patterns. The dynamic systems approach suggests that new motor skills may emerge from the confluence of many interacting factors, each with its own developmental trajectory(Thelen& Smith, 1994). When talking about motor development, one can distinguish two main areas of interest: the development of motor skills, and the development of motor abilities. Skills are learned and are sort of expertise on a specific motor task, while abilities are 23

General introduction genetically predetermined characteristics that affect movement performance. Abilities are enduring, and therefore difficult to change in adults. They differ from skills in the sense that skills are learned, whereas abilities are a product of both, learning and genetic factors(Fleishman, Quaintance,& Broedling, 1984). In our case, grasping (i.e. for the coffee cup in the morning) is a skill, as grasping efficiently with regard to control precision, manual dexterity or arm-hand steadiness, for example(see Fleishman's taxonomy of motor abilities; Fleishman et al.,1984), needs to be learned early during the ontogenetic development. Anticipatory planning on the other side is an ability, as it is shaped by biological and physiological factors and affected by environmental factors. The rate, at which abilities develop, varies across childhood and adolescence, both within and across individuals(Fleishman et al., 1984), which is largely due to growth and maturation changes, including the development of cognitive control. Both areas will be shortly addressed in the next two sections, using the development of grasping as a motor skill and the development of anticipatory planning as a motor ability. 1.3.2.1. Development of motor skills ­ grasping Arm movements and grasping actions are observable even before birth. After birth, infants have to cope with gravity, and with about four to five months of age children are capable of adjusting their hands in vertical and horizontal positions, depending on the object to-be grasped(von Hofsten& Fazel-Zandy, 1984). Now, they become successful in reaching to objects(Oztop, Arbib,& Bradley, 2006). When reaching towards an object, children aged five to six months begin to close their hands prior to the approach of the object(von Hofsten& Ronnquist, 1988), but the anticipatory opening of the hand, depending on object size, is not adjusted prior than the age of 13 months. The development of anticipatory grasping is due to the development of the ability to visually anticipate the to-be grasped object, which emerges somewhere around 8 months(Adolph, Eppler,& Gibson, 1993). 24

General introduction However, visual information is only needed about the object, not for the way of grasping(Adolph& Berger, 2006). Bower(1977) proposed a phase model of the development of reaching and grasping behavior. He divided the developmental pattern into two phases. The contents of Phase 1 and 2 are summarized in Table 1 below(according to Payne& Isaacs, 1999). Table 1. Bower's(1977) Phase 1 and 2 Reaching and Grasping Behavior Characteristics(adapted from Payne& Isaacs, 1999, p. 230). Phase 1 Phase 2 1. Simultaneous reaching and grasping Differentiated reaching and grasping 2. One-handed reaching Two-handed reaching 3. Visual initiation of the reach Visual initiation and guidance of the reach 4. Visual control of the grasp Tactile control of the grasp The complete anticipation and execution of grip-formations develops over time, until it is fully present in adolescence. Even six to seven year old children are more dependent on visual feedback than adults are(Kuhtz-Buschbeck, Stolze, Boczek-Funcke, Joehnk, Heinrichs,& Illert, 1998). Once infants master the motor components of grasping, they use these skills to extend their own abilities and to bring about rewarding outcomes(like bringing food to their mouth with a spoon). Piaget was the first to describe how older infants separate reaching and grasping into means for achieving ends(Piaget, 1936). He was also one of the leading investigators regarding cognitive development in childhood. 25

General introduction 1.3.2.2. Development of motor abilities- anticipatory planning As mentioned above, planning is observable in a lot of daily activities. People do not only use anticipatory planning skills when they take their cup out of the cabinet for a coffee in the morning, they also use anticipatory planning skills in a lot of other actions. For example, if people are on the way to work, sitting in their car, they automatically try to anticipate what the car in front is doing. This is foresighted driving in order to prevent an accident when the car in front is suddenly braking. Or, if one is going down stairs, he/she will anticipate the step-width and will adapt his or her steplength, so that he/she can comfortably walk down the stairs. Hence, the ESC effect is only one possibility to measure anticipatory planning skills. In recent years, there has been an increased interest in uncovering the developmental pattern of the ESC effect(e.g. Hughes, 1996; Thibaut& Toussaint, 2010; Weigelt& Schack, 2010). Overall, the findings of these studies suggest an increase of ESC planning in preschool children between three and five years(e.g. Weigelt& Schack, 2010) and a continuation of this trend in primary school children until they reach the age of ten(e.g. Crajé, Aarts, Nijhuis-Van der Sanden,& Steenbergen, 2010; Stöckel, Hughes,& Schack, 2012; Thibaut& Toussaint, 2010). When taking a closer look at these studies, however, it is hard to identify a specific pattern of development. Also, there are considerable differences between the tasks used and the procedures followed, which makes a direct comparison between most of these studies difficult. Moreover, some of the findings are controversial, as a number of studies were not able to replicate the ESC effect in children of the same age (Adalbjornsson, Fischman,& Rudisill, 2008; Manoel& Moreira, 2005). Therefore, a systematic review of the research on ESC planning in children is needed. As both, ESC and EF, account for planning, and motor planning is impaired in patients with EDF, it can be assumed that ESC and EF are related in their development. 26

General introduction 1.4. The possible relationship of anticipatory planning and cognitive control Traditionally, the developmental trajectories of intellectual(cognitive) skills and (perceptual-) motor skills were examined separately, but in recent publications a relationship between their development has been suggested(Rosenbaum, Chapman, Weigelt, Weiss,& van der Wel, 2012). There are two principal approaches to examine this assumption: 1) a deficit approach, and 2) a developmental approach. Until today, most studies examined both factors separately, within one of these two principle approaches. The deficit approach has been used to study impairments of motor planning abilities(i.e. ESC) and cognitive skills(i.e. EF). For example, recent studies found in people with hemiparetic cerebral palsy, that motor deficits do not only manifest themselves in motor execution, but also in action planning(Gordon, Charles& Steenbergen, 2006; Steenbergen& Gordon, 2006; Steenbergen, Verrel& Gordon, 2007). From clinical studies on motor planning, it is known that planning skills are also impaired in a range of neurological disorders like in attention deficit hyperactivity disorder(ADHD, Scheres et al., 2004), in autism(Hill, 2004; Hughes, 1996), and in developmental coordination disorder(DCD, Smyth& Mason, 1997). Therefore, results suggest that people with different neuropsychological disorders show a decreased ability for motor execution and motor planning skills. The same approach is taken for the examination of EF. In a variety of studies, EF is used to characterize a diverse set of processes, which can be impaired by damage of the prefrontal cortex and other brain regions(for an overview see Stuss& Benson, 1986). Moreover, it is known that suffering from ADHD may be caused by delayed maturation of the prefrontal cortex or by brain injuries in this area, which is the primary seat of EF(National Institute of Mental Health, 2012). A study of the Johns Hopkins Medical Institutions(2008) also suggests, that autism may be caused be prenatal brain damage. Therefore, it 27

General introduction can be assumed, that poor executive planning skills(not only due to brain injury or brain damage) is linked to poor motor skills. With the second, the developmental approach, both, the development of ESC and the development of EF can be examined. Results suggest, that there is a positive developmental trajectory with the onset of ESC planning abilities found in children ranging from three years(e.g. Jovanovic and Schwarzer, 2011) until 10 years of age(e.g. Thibaut and Toussaint, 2010). As some studies were not able to find ESC planning until the age of 14(e.g. van Swieten et al., 2010), the development seems to be protracted until adolescence, depending on different constraints, like task specifications. For EF tasks, a similar result emerges. Studies on EF development found a positive increase in EF during childhood until adolescence, with some sub-functions maturing early and others maturing later during the ontogenetic development(e.g. Anderson, 2002; Davidson, Amso, Anderson,& Diamond, 2006). However, a review assembled by Rosenbaum, Carlson, and Gilmore(2001) suggested that both, intellectual and(perceptual-) motor skills, are acquired in similar ways. Until today, there are only few studies investigating the relationship of the development of motor and cognitive skills. In a study conducted by Jenni, Chaouch, Calfish, and Rousson(2013), children between 7 and 18 years were tested using the Zurich Neuromotor Assessment(ZNA) and standardized intelligence tests. The correlations found between the performances in both parts were generally weak, which led the authors to suggest that motor and intellectual domains are largely independent. Another recent study conducted by Gonzalez and colleagues(2014) examined children between five and 10 years, using the Behavioral Rating Inventory of Executive Function(BRIEF) and two motor tasks with a focus on grasping. Results revealed significant correlations between the strength of right hand preference for grasping and numerous elements of the BRIEF, showing an interconnectedness of lateralization and EF. Moreover, Jansen(2014) conducted a review study, where she summarized results on studies on the relationship of motor activity and cognitive 28

General introduction functions. She concluded, that there is a positive effect of motor activity on the development of EF, and that specific physical activity can help to enhance specific cognitive functions in children. Altogether, motor(or better: physical) activity can play an important role in the development of EF and therefore, can act as a mediator on the relationship of ESC and EF. Therefore, results on the development of motor planning and cognitive functions, as well as their relationship at different ages, are inconclusive. Further research is needed on this topic. Below, the motivation of the current research project is constituted, which will be addressed in the following Chapters. 1.5. Motivation of the current research endeavor When researching on the developmental pattern of anticipatory planning, especially in terms of the ESC effect, it becomes obvious, that all studies find a positive developmental trajectory of anticipatory planning abilities in childhood. Nevertheless, results are quite different, regarding the slope of this development. It is hard to identify a specific pattern out from the existing literature. Moreover, all studies are considerably different, regarding the tasks and procedures applied, making it even more difficult to interpret the results on a common level. Some of the findings are controversial, as a number of studies were not able to replicate the ESC effect in children. Therefore, the first aim of the research project is to provide a systematic review of the research on ESC planning in children. The systematic review of the literature will be provided in Chapter 2 of this dissertation. Differences in grasp selection between adults and children have been construed as indicating a deficit in childr en's planning abilities(see Hughes, 1996; Smyth & Mason, 1997), with the presence of ESC as an indicator for"thinking ahead", refe rring somewhat ambiguously to some kind of cognitive planning skills. This led to an important debate, initially raised by van Swieten et al.(2010), about whether perfor29

General introduction mance in grasp selection tasks is driven by the development of executive planning (i.e. actively planning ahead to solve actions correctly or to avoid mistakes, for example) or the development of motor planning(i.e. planning motor actions in advance in order to solve them correctly or most economically), or both(Scharoun& Bryden, 2013; Stöckel et al., 2012). In this regard, theories of child development assumed for a long time that the developmental origins of perceptual-motor skills and intellectual skills are closely related, for example by Piaget(1936) and Vygotski(1977), who based the development of intelligence upon the emergence of skilled action. Such a potential relationship in the development can be expected from the similarity of the developmental trajectories of EF and the ESC effect(see Anderson, 2002, for the projected developmental trajectory of the different EF, and Chapter 2, for an overview of the developmental trajectories for different ESC tasks). Hence, the second aim of the research project is to investigate the possible influence of executive functions on the developmental trajectory of the ESC effect. To this end, an experimental study examining the development of motor planning abilities and cognitive skills in children from 3- to 10-years of age was conducted. The experimental study will be reported in Chapter 3. 30

Chapter 2 Chapter 2 2. The end-state comfort effect in childhood a systematic review This chapter is based on the publication of Wunsch, K., Henning, A., Aschersleben, A.,& Weigelt, M. (2013). A systematic review of the end-state comfort effect in normally developing children and in children with developmental disorders. Journal of Motor Learning and Development, 1(3), 59-76. However, all data presented in this chapter was checked again and updated on the 15 th of October, 2014. 31

The end-state comfort effect in childhood- a systematic review 2.1. Introduction into the systematic review Adults reach for objects in an anticipatory manner, taking the future states of their body during the sequence of motor actions into account, while planning the intended maneuver(Lashley, 1951; Rosenbaum, Cohen, Jax, Weiss,& van der Wel, 2007). Remember the person, who is picking up an overturned cup out of the kitchen cabinet in the morning to pour coffee into it. Most likely, he/she will reach for the cup in a way that ensures a comfortable posture of the hand at the end of the maneuver (see Chapter 1.2.1.). To this end, he/she will use an initial grip with the thumb pointing downward, in order to hold the glass comfortably with the thumb pointing upward when the rotation is completed. He/she will thus tolerate an uncomfortable posture of the hand at the beginning of the grasping action in anticipation of the final posture that ensures a comfortable position for the following action of pouring coffee into the cup. This sensitivity to avoid uncomfortable postures at the end of object manipulations has been first reported by Rosenbaum and colleagues(Rosenbaum et al., 1990) and has been termed the end-state comfort(ESC) effect. Ever since its first discovery, this effect has been taken to signify the presence of anticipatory motor planning skills. While there is a large body of research showing the ESC effect in adults, surprisingly few studies have assessed the ESC effect in children. In fact, the developmental pattern of the ESC effect and its relation to the maturation of higher cognitive abilities are largely unknown. The findings of studies on tool use and object manipulation in infants and young children are inconclusive. While McCarthy, Clifton, and Collard(1999) found 19-month old children to reliably alternate their hands in a spoon-handling task, which suggests efficient effector selection to be present in children of this age, a number of studies showed that many other action features(like the arm's end posture) are not planned in advance at this age(Barrett, Traupman,& 32

The end-state comfort effect in childhood- a systematic review Needham, 2008; Claxton, McCarty,& Keen, 2009; Lockman, 2000; Manoel& Conolly, 1997). For example, Connolly and Dalgleish(1989) examined grasp behavior in a spoon handling task in children aged between 11 and 18 months and observed that these children did not select their grasps in a way that favors a more comfortable posture in the end of the action. Instead, they most often selected a comfortable initial grip. The fact that young children appear to be biased towards minimizing initial discomfort, however, cannot be taken as evidence for a planning deficit per se, because these children were generally able to solve the particular task. It is therefore of interest, when and at what rate more sophisticated motor planning skills develop. As mentioned in Chapter 1.3.2.2., research on the development of ESC is meager and results seem to be inconclusive. Therefore, light will be shed on the topic of ESC development in the following sections of Chapter 2. A first impression on the studies suggest an increase of ESC planning in preschool children(Weigelt& Schack, 2010) and a continuation of this trend in primary school children until the age of ten(Crajé, Aarts et al., 2010; Stöckel et al., 2012; Thibaut& Toussaint, 2010). On the other hand, a number of studies were not able to replicate the ESC effect in children of the same age(Adalbjornsson et al., 2008; Manoel& Moreira, 2005). Therefore, the general aim of Chapter 2 is to provide a systematic review of the research on ESC planning in children. More specific goals of the present review are:(1) to identify and provide a detailed overview of the existing studies on the ESC effect in children,(2) to compare the various tasks and procedures employed in order to highlight differences and similarities between these studies,(3) to examine the different developmental patterns of ESC planning in normally developing children and in children with developmental disorders, 4) to discuss possible factors influencing the development of ESC planning, and 5) to provide new directions for future research. 33

The end-state comfort effect in childhood- a systematic review 2.2. Method In order to identify studies assessing the ESC effect in children, a search procedure(based on other systematic reviews like the one of Blauw-Hospers& Hadders-Algra, 2005 or Greaves, Imms, Dodd,& Krumlinde-Sundholm, 2010, for example) containing eight distinct steps was applied on three databases to acquire and further process the data. These eight steps and the resulting data set at each step are depicted in Figure 5. First, a keyword search was performed for journal articles and book chapters on the ESC effect in the databases of Medline 1 , PubMed 2 and Scopus 3 by using the following keywords, which were carried over from known papers on ESC planning: end-state comfort, anticipatory planning, end-posture comfort, end-posture planning, grip selection, bar-transport-task, handle-rotation-task, start-state comfort and startposture comfort. As a result, three separate data sets were generated, with the results of all keyword searches summarized; one set for Medline(2592 articles and book chapters), one for PubMed(971 articles and book chapters), and one for Scopus(1617 articles and book chapters). Second, to reduce each of the three data sets to studies conducted with children, a search with the original keywords and three additional keywords, c hildren, childhood and development, was conducted. The new data sets of Medline included 519, of PubMed 249, and of Scopus 497 journal articles and book chapters. Because there were still articles and book chapters in the reference bases without any relevance to motor development in children(such as articles named Endof-life discussions and advance care planning for children on long-term assisted ven1 https://portal.dimdi.de/websearch/, last access on October, 15 th , 2014 2 http://www.ncbi.nlm.nih.gov/pubmed/, last access on October, 15 th , 2014 3 http://www.scopus.com/home.url, last access on October, 15 th , 2014 34

The end-state comfort effect in childhood- a systematic review tilation with life-limiting conditions, for example, which contain single words of the keywords, but do belong in another field of interest), the content of each database was further scrutinized in a third step and manually separated into studies on motor development and studies on other topics. Other topics included medicine(e.g. end-of life discussions, well-child care, advance-care planning), psychology(e.g. cognitivebehavioral therapy, gaze behavior, skill acquisition in music), neuroscience(e.g. occulomotor abnormalities, attention shifts), nursing(e.g. nutrition intake, depressive distress), and/or health professions(e.g. syllable planning, social participation acquirement). The latter ones were then excluded from the databases. This resulted in 91 studies in Medline, 78 in PubMed, and 89 in Scopus. Fourth, all databases were further scrutinized to ensure that all studies on adults and on animals were excluded from the following analysis. This resulted in 48 studies on children in Medline, 57 in PubMed, and 74 in Scopus. Fifth, the remaining data sets of the three databases were then combined into a single data set, consisting of 179 studies on children. As the studies on ESC in children appeared in every data set, 57 studies can be subtracted from this amount. The remaining 122 studies almost included studies on anticipation in infants and toddlers, whereas ESC was not explicitly measured. As most of these studies appeared in all three databases, the break-down from 179 to 19 studies can be explained. While making sure that every study only appeared once, a preliminary set of 19 studies was formed. Each study of this preliminary data set was scrutinized for tasks employing grasping actions and focusin g on children's grip selection at the beginning and/or the end of the movement. One study, which did not fulfill this criterion(i.e. Cowie, Smyth,& Braddick, 2010, who examined anticipatory planning in a locomotor task) was also excluded in this step. This resulted in a final set of 18 studies on manual ESC planning in childhood(see Figure 5). 35

The end-state comfort effect in childhood- a systematic review Sixth, the reference lists of all 18 articles retrieved on children were checked one by one for further related references, which were potentially missing from the previous search. No additional publications were identified. Seventh, the combined data set was further partitioned into studies on children with a focus on normal development(n= 12) and on developmental disorders(n= 6) 4 , which included children with cerebral palsy(CP), developmental coordination disorder(DCD), mild learning deficits(MLD), and autism. Eighth, all studies were evaluated for the type(s) of ESC tasks used in combination with the child population tested. Three different tasks were identified: the bartransport-task, the handle-rotation-task, and the overturned-glass-task. 4 Please note that studies on children with different developmental disorders(n= 6) usually contained also of a group of normally developing children. The focus of the studies, however, was on the developmental disorders and the normally developing children were only included as a control group. 36

The end-state comfort effect in childhood- a systematic review Figure 5. Schematic overview of the distinct steps to analyze the reviewed studies. 37

The end-state comfort effect in childhood- a systematic review 2.3. Results As a result of the search procedure described above, a total of 18 studies investigating the ESC effect in children in manual actions were identified. Twelve of these studies assessed ESC planning in normally developing children(Adalbjornsson et al., 2008; Jongbloed-Pereboom, Nijhuis-van der Sanden, Saraber-Schiphorst, Crajé,& Steenbergen, 2013; Jovanovic& Schwarzer, 2011; Knudsen, Henning, Wunsch, Weigelt,& Aschersleben, 2012; Manoel& Moreira, 2005; Robinson& Fischman, 2013; Scharoun& Bryden, 2013; Stöckel et al., 2012; Thibaut & Toussaint, 2010; Toussaint, Tahej, Thibaut, Possamai,& Badets, 2013; Weigelt & Schack, 2010; Wilmut& Byrne, 2014b). The remaining six studies focused on the ESC effect in children with different developmental disorders, including children with autism(Hughes, 1996; van Swieten et al., 2010), children with developmental coordination disorder(Smyth& Mason, 1997; van Swieten et al., 2010; Wilmut& Byrne, 2014a), with mild learning deficits(Hughes, 1996), and with cerebral palsy(Crajé, Aarts, et al., 2010; Kirpatrick, Pearse, Eyre,& Basu, 2013). Across these studies, three different kinds of basic tasks(or versions thereof) were employed: the bartransport-task, the handle-rotation-task, and the overturned-glass-task. In the following section, the original version of each of these tasks is described in more detail. Then, all 18 studies on the development of planning skills in children are reviewed, with a focus on the particular modifications of the task used and the procedure employed, as well as on the results of each study. Studies on normally developing children are presented first, followed by studies on children with developmental disorders. Finally, the results for the two child populations are reviewed in relation to the tasks employed. For an overview of all studies and their results, see Table 2. 38

The end-state comfort effect in childhood- a systematic review Table 2. Overview of the mean percentage of ESC behavior found in the different studies. Task Author(s) Dev. Disorders/ Condition Hughes(1996) Jovanovic and Schwarzer(2011) Normally developing MLD Autism Normally developing Knudsen et al.(2012) Normally developing BTT Manoel and Moreira(2005) Normally developing Smyth and Mason(1997) Normally developing DCD Stöckel et al.(2011) Normally developing Thibaut and Toussaint(2010) Normally developing Toussaint et al.(2013) Normally developing Weigelt and Schack(2010) Normally developing Crajé et al.(2010) Normally developing CP Jongbloed-Pereboom et al.(2013) Normally developing Kirpatrick et al.(2013) CP Smyth and Mason(1997) HRT Normally developing DCD Normally developing Van Swieten et al.(2010) DCD Autism Wilmut& Byrne(2014a) Normally developing DCD Wilmut& Byrne(2014b) Normally developing Adalbjornsson et al.(2008) Normally developing Knudsen et al.(2012) OGT Robinson& Fischman(2013) Normally developing Normally developing Sharoun and Bryden(2013) Normally developing AE No-AE LC HC AE No-AE unimanual bimanual Age 12 3 4 5 6 7 8 9 10 11-12 13-14 14 71 42 70 28 6 8 0 60 0 38 75 75 86 88 25 38 88 75 89 100 40 50 50 50 50 50 50 25 38 29 50 29 50 49 60 65 70 65 60 65 65 60 50 67 92 45 65 50 80 55 75 20 45 70 10 30 65 75 10 30 35 24 21 35 41 50 42 55 50 62 61 20 35 40 35 20 35 45 40 20 35 10 25 35 65 55 62 71 82 20 35 75 69 88 88 86 100 57 69 94 75 93 94 65 10 35 50 90 95 30 62 81 90 95 39

The end-state comfort effect in childhood- a systematic review Note. AE= action effect, BTT= bar-transport-task, CP= cerebral palsy, DCD= developmental coordination disorder, HRT= handle-rotation-task, MLD= mild learning deficits, OGT= overturned-glasstask, LC= low precision requirements, HC= high precision requirements. Note further: The ages declared in the table for the study of Hughes(1996) for the MLD- and the autistic children do not match their chronological ages. They are the nonverbal mental ages(NVMA), assessed by the Matrices task of the British Abilities Scale(Elliot et al., 1983). The NVMA of the MLD group correspond to a chronological age of 10 years for those with a low NVMA of 4.7 and to 11 years of age for those with a high NVMA of 6.9. In the autistic group, the children with low NVMA(4.7) correspond to the chronological age of 12 years, those with high NVMA(9.8) to 13 years of age. Furthermore, in the Scharoun and Bryden(2013) study, data is collapsed over critical and uncritical trials. In the studies of Wilmut and Byrne(2014a, 2014b), only data for one-step movements are provided, not for two-step and three-step movements. In the study of Kirpatrick et al.(2013), data was collapsed across left- and right-handers and across clockwise and counterclockwise rotations. In the study conducted by Robinson and Fischman(2013), data shows the percentage of participants, who showed ESC in at least one of the two similar tasks. 2.3.1. Tasks used to investigate the ESC effect in children In the next section, the tasks used and the procedures employed to investigate the development of motor planning skills in children are described for their original versions, respectively. In the sections to follow, this should help to better highlight the modifications to the original tasks and to better understand the possible implications this may have had on some of the results on children. 2.3.1.1. Bar-transport-task This task was originally developed by Rosenbaum et al.(1990)(see Figure 6A). They instructed adult participants to take hold of a bar with one grey and one black end, which was suspended horizontally on two supports, and to place it in a vertical position with either its white or black end onto either a blue or red target circle to the left or right side of the supports on the table surface 5 . Importantly, participants were not instructed how to grasp the handle and therefore, it was of interest if they chose an overhand or an underhand grasping posture at the beginning of the action, 5 The use of different colors for bar ends or target locations is reported for the sake of completeness. However, there is no evidence whether or not the colors used in a particular experiment(additionally) affec ted children's task performance. 40

The end-state comfort effect in childhood- a systematic review resulting in a comfortable thumb-up vs. an uncomfortable thumb-down end position. Also note, that participants only held the bar's end briefly on the target circle(no i nserting action), before returning it. There were no familiarization trials and participants performed the task standing throughout the experiment. 2.3.1.2. Handle-rotation-task This task was introduced by Rosenbaum et al.(1993). Participants were presented with a disk with eight target numbers(1-8), which were distributed equidistantly around the disk(similar to the numbers of a clock). A handle was fixed at the center of the disk and could be rotated 360 degrees like the arm of a clock(see Figure 6B). One end of the handle was marked by a cardboard tab. The apparatus was positioned on the wall in a 45 degree angle and participants were instructed to reach out and rotate the handle with their right hand to a particular target position, i.e. to cover the designated target number with the marked end of the handle. Participants were not instructed how to grasp the handle and therefore, it was of interest, whether participants grasped the handle with the thumb towards the mark or away from it, enabling them to finish the rotation in a comfortable or in an uncomfortable posture of the arm and hand. The task was performed standing. 2.3.1.3. Overturned-glass-task The overturned-glass-task was already conceptualized by Rosenbaum et al. (1990), after the first author observed a waiter manipulating glasses in a restaurant, just as described in the introduction to this dissertation. Fishman(1997) was the first to bring this observation into the laboratory. The task was to pick up a drinking cup that was placed upside down on a table, to turn it over, and to pour water from a pitcher into it(see Figure 6C). In half of the trials, the glass needed to be held while pouring water into it, and in the other half, it had to be put on a coaster. Participants 41

The end-state comfort effect in childhood- a systematic review were not instructed how to grasp the cup and thus, were free to select any grasping posture. It was of interest, if participants grasped the glass with an uncomfortable thumb-down grasp, before turning it over to pour water into it, to ensure a comfortable end-posture with the thumb pointing upwards. Participants performed the task sitting. Figure 6. Schematic illustration of the original versions of the three tasks used to examine the ESC effect in children. A) Bar-transport-task(Rosenbaum et al., 1990) with initial position(1) and four possible end positions(2, 3). B) Handle-rotation-task(Rosenbaum et al., 1990). C) Overturned-glass-task (Fischman, 1997) with first turning action(1) and subsequent pouring action(2). 2.3.2. The ESC effect in children of various populations After the tasks used to study ESC planning in children have been described, the focus in the following section will now be on the results of studies with normally developing children and of studies with children of different developmental disorders. Please note that the focus will only be on children' s motor performance, while the findings regarding other intellectual competencies are not reviewed here. As opera42

The end-state comfort effect in childhood- a systematic review tional definitions greatly varied across studies within and across tasks, the presence of ESC planning in a particular group of children(e.g. children of a certain age or with a particular developmental disorder) is supported when more than half of these children show the ESC effect in the majority of trials(i.e. 50% of participants showed in  50% of all trials anticipatory planning behavior ) under critical conditions(i.e. when an initial uncomfortable grip is tolerated in order to finish the action comfortably). 2.3.2.1. Studies on normally developing children In a most recent study, Toussaint et al.(2013) examined the link between action planning and motor imagery in 6- to 8-year old children. Therefore, they correlated the results of these tasks, finding that the ability to engage sensorimotor mechanisms when solving a motor imagery task was concomitant with action planning efficiency. To test ESC, they used the original bar-transport-task conducted by Rosenbaum and colleagues(1990). Here, the bar rested on two cradles, with one blue and one red end. On the left and the right of the supports, a black and a white target disk was placed. A noticeable detail of the procedures was that all children performed the task sitting(and not standing as in the original study by Rosenbaum et al., 1990). Before each trial, children were asked to put their hands on their knees. Then, each child performed five blocks of four randomly presented trials, resulting in a total of 20 trials. There were no familiarization trials and the task was not demonstrated by the experimenter. Sixty-four children were divided by half into two age groups, resulting in one group of 6-year old children, and one group of 8-year old children. Almost all children used an overhand grip if this resulted in ESC. In the critical conditions, however, 55% of the 6-year-olds and 75% of the 8-year-olds showed sensitivity for ESC planning. This difference was found to be statistically significant. 43

The end-state comfort effect in childhood- a systematic review Manoel and Moreira(2005) examined whether young children between 2½ and 6 years of age plan motor actions on the basis of the ESC effect and whether an increasing demand for precision would make the effect more apparent. To this end, they used a modified version of the original bar-transport-task by Rosenbaum et al. (1990). Specifically, Manoel and Moreira(2005) presented only a single target hole on top of a squared box, which was positioned behind the supports. In addition to the cylindrical bar(low precision condition, LC) used in the original version, a second bar with semi-cylindrical distal parts(high precision condition, HC) was introduced to increase precision requirements. Importantly, the bar end had to be inserted into the target disk. The authors tested six conditions in their study: In the two LC conditions, participants had to insert the green or red end of the cylindrical bar into the target hole. In the four HC conditions, the bar was initially presented either with the flat sides facing upwards or downwards on the supports and then, either the green or the red end of the semi cylindrical bar had to be inserted into the target hole. ESC depended on initial grip selection(overhand vs. und erhand) relative to the bar's end orientation. Again, children performed the task sitting. Each child completed five trials in each of the conditions, resulting in a total of 30 trials. All trials were randomly presented. There were no familiarization trials and the task was not demonstrated by the experimenter. Forty children were divided into seven age groups, with mean ages of 2 years 8 months(n= 5), 3 years 4 months(n= 6), 3 years 11 months(n= 6), 4 years 2 months(n= 5), 4 years 10 months(n= 6), 5 years 2 months(n= 6), and 5 years 10 months(n= 6). Results did not show the ESC effect to be present in the children of all age groups, irrespective of whether they performed in the low or high precision condition. Instead, all children preferred an initial-state comfort, i.e. to grasp the bar with an overhand grip across all conditions. According to these results by Manoel& 44

The end-state comfort effect in childhood- a systematic review Moreira(2005), children of up to 6 years of age do not plan their object manipulations with regard to ESC. A modified version of the original bar-transport-task with only a single target location was also used by Weigelt and Schack(2010). The target hole was on top of a squared box, which was placed on the table in front of the support. The bar was marked with one black and one white end, and was presented in a horizontal orientation. All children performed the task standing in front of the apparatus. They completed a total of six trials, in which each bar end(black vs. white) had to be inserted into the target hole three times. The initial orientation of the bar was kept constant for each child across all trials(e.g. black end always pointing to the right), while the color to be inserted was randomized. Thus, depending on the bar's final orientation, with either the left end or right end inserted, ESC could be reached with either an initial overhand grip in the uncritical trials or with an initial underhand grip in the critical trials. Fifty-one right-handed children were tested by Weigelt and Schack(2010), being 3, 4, or 5 years of age(n= 17 in each age group). The ESC effect was considered to be present if a child showed ESC in at least two out of the three trials in each condition. There were no familiarization trials. All children reached for the bar with an overhand grip if this resulted in ESC in the uncritical trials. In the critical trials, 18% of the 3-year-olds, 47% of the 4-year-olds, and 70% of the 5-year-olds selected an underhand grip and finished the object manipulation in a manner consistent with ESC. Interestingly, 11(65%) of the 3-year-olds never used an underhand grip and therefore, inserted the bar always with an awkward hand posture. The same was true for 7 (41%) of the 4-year-olds and 4(24%) of the 5-year-olds. Also, none of the 3-yearolds selected an underhand grip when the critical trial was presented for the first time. By the third trial, however, 5(29%) of these children used an underhand grip and thus, had changed their grip strategy. Such a systematic pattern of grip adjust45

The end-state comfort effect in childhood- a systematic review ment was not present in the 4- and 5-year-old children. Different to the results of Manoel and Moreira(2005), Weigelt and Schack(2010) showed the presence of ESC planning in preschool children, as well as an increase of these planning abilities in these children from 3 to 5 years of age. In a more recent study, Stöckel et al.(2012) used a modified version of the bar-transport-task that was similar to the one by Weigelt and Schack(2010). A difference, however, was that participants had to hold the bar in the final position for five seconds. Thirty-six primary-school children in three age groups(7-year-olds, 8-yearolds and 9-year-olds), with 12 children in each age group, were tested. They performed two trials in each of the four conditions(i.e. each end had to be inserted twice for each start orientation of the bar), resulting in a total of eight trials, which were presented in a randomized order. There were no familiarization trials before testing. In the uncritical condition(i.e. an initial overhand grip resulted in a comfortable finial posture), 33(92%) of the children of all three age groups showed the ESC effect. However, in the critical conditions(i.e. an initial underhand grip was required to end in a comfortable finial position), 6(50%) of the 7-year-olds, 8(67%) of the 8-year-olds and 11(92%) of the 9-year-olds showed the ESC effect. These differences in grip behavior between the age groups in the latter condition were significant and thus, demonstrate a steady increase of the ESC effect in primary school children. Thibaut and Toussaint(2010) tested children's plann ing skills in two different experiments. In Experiment 1, they used the original bar-transport-task by Rosenbaum et al.(1990), with the only difference being that their participants were sitting. A horizontal bar was always presented with the red end facing to the left and the blue end facing to the right. This resulted in four types of trials: blue end onto white disk, blue end onto black disk, red end onto white disk and red end onto black disk. A block consisted of these four trial types, which were presented in random order. Five blocks resulted in a total of 20 trials. There were no familiarization trials. 46

The end-state comfort effect in childhood- a systematic review They tested 120 right-handed children, aged 4, 6, 8, and 10 years(n= 30 in each age group), as well as twenty adults(mean age= 33 years). Almost all participants grasped the bar with an overhand grip when this resulted in ESC in the uncritical trials. In the critical trials, in which an underhand grip had to be adopted, all adults showed the ESC effect. A different pattern was found for children, which suggested that the ESC effect was present in 12(40%) of the 4-year-old, in 21(70%) of the 6year-old, in 15(50%) of the 8-year-old, and in 24(80%) of the 10-year-old children. This pattern of results supports the notion of motor reorganization, which takes place around the age of 8 years, as suggested by other authors(Bard, Hay,& Fleury, 1990; Meulenbroek& van Galen, 1988). In Experiment 2, Thibaut and Toussaint(2010) varied the original bartransport-task and replaced the bar with a two-tailed pencil, with one end red and the other end blue. Participants were asked to take the pencil and make a dot(pointingwith-pencil-task) or to draw a line without constraints(tracing-with-pencil-task), or to draw a line within a given alley(pencil-alley-task) in a particular color(red vs. blue) on a sheet of paper. For all three tasks, the blue end of the pencil always faced to the right. There were two types of trials, either tracing with the blue end or tracing with the red end. In each task, each trial was presented two times in random order. They tested 80 right-handed children between 4 and 10 years of age(n= 20 in each group). In the pointing-with-pencil-task, 45% of the 4-year-old, 75% of the 6-year-old, 50% of the 8-year-old, and 95% of the 10-year-old children performed in a manner consistent with ESC. Results were similar in the tracing-with-pencil-task. Here, 50% of the 4-year-old, 75% of the 6-year-old, 55% of the 8-year-old, and 90% of the 10year-old children grasped the pencil ending comfortably. Therefore, the pointing-withpencil-task and the tracing-with-pencil-task replicated the results of Experiment 1, with the performance of 8-year-old children being lower than those of 6- and 10-yearolds. In contrast, in the pencil-alley-task 45% of the 4-year-old, 80% of the 6-year-old, 47

The end-state comfort effect in childhood- a systematic review 80% of the 8-year-old, and 95% of the 10-year-old children behaved in an anticipatory manner. Hence, motor reorganization seemed to affect task performance of the 8-year-old participants to a smaller degree in the pencil-alley-task task. Jovanovic and Schwarzer(2011) employed a largely modified version of the original bar-transport-task. They presented children with a vertically oriented bar whose two ends differed in diameter, such that only the narrow end of the bar, but not the large end, fitted into an empty cylinder(single target object). Children's task was always to fit the narrow end into the cylinder. Also, the correct insertion of the bar into the target cylinder resulted in a light effect, which could be readily noticed by the child. Children sat in front of the apparatus on the lap of their parents and completed a total of six trials. In the three uncritical trials, no rotation was required and the bar could be grasped with the thumb pointing upwards to place it into the target hole by a simple translation movement. In the three critical trials, however, a 180 degree rotation(instead of the 90 degree rotation in the original task version) had to be completed, before inserting the bar's end into the targe t hole. To finish the object manipulation comfortably in the critical trials, the bar had to be grasped with the thumb pointing down. Also, two types of grip adaptations with respect to grasp height were helpful: In the uncritical condition without rotation, it was better to grasp the bar high, whereas, in the critical condition, it was better to grasp the bar low(i.e. close to the narrow end or close to the large end, respectively). Both trial conditions(critical vs. uncritical) were presented in alternation. Thus, children were allowed a familiarization trial in a way that the experiment always started with an uncritical trial, in which the bar could be transferred"easily" and inserted without a rotation. The experimenter grasped the bar, lifted it, and put it into the cylinder, which caused the lights to switch on. As this demonstration always took place in the uncritical condition, the experimenter used a comfortable grip when transferring the bar into the target hole. Furthermore, the task was demonstrated twice, before 48

The end-state comfort effect in childhood- a systematic review the child was encouraged to perform the same action. In the subsequent trials, the child was then only encouraged to``switch on the lights again''. The study included 81 children, which were divided into one of three age groups: 1½ years(n= 36), 2 years(n= 25), and 3½ years(n= 20). Results showed that 3(8%) of the 1½-yearolds, none of the 2-year-olds, but 12(60%) of the 3½-year-olds demonstrated the ESC effect. These results suggest that the ESC effect is not present in the majority of children below the age of three years. Knudsen et al.(2012) used a similar version to Jovanovic and Schwarzer (2011). At their body midline, participants were presented a box with an insertion hole on its top and a smiley configuration of LED' s inserted at its front facing the participant. To the left and to the right of this box, a bar holder was placed, which held the bar in an upright position when placed into it with the narrow side down. Only this side was able to be inserted into the hole; the other side had a platform at its end. Participants were asked to insert the bar into the opening of the box with their preferred hand. Half of the six trials were critical trials, in which the bar rested on its platform at the beginning of the trials and then had to be rotated by 180 degrees to fit with its narrow end into the target hole. Here, an initial thumb-down grip was required to end the action comfortably. Every testing session started with a demonstration of the starting state and the desired end state by the experimenter. However, the child never saw the experimenter grasped the bar, as the set-up and all actions of the experimenter were covered. Participants performed the task standing, starting approximately 70 cm away from the apparatus. The starting position of the bar was always opposite to side of the hand, which the participant was about to use in the upcoming trials. Knudsen et al.(2012) tested 116 participants between 3 and 8 years, divided into nine age groups(3-year-olds, 4-year-olds, 5-year-olds, 6-year-olds, 7-year-olds, 49

The end-state comfort effect in childhood- a systematic review 8-year-olds[all n= 16], and adults[n= 20]). Each group was divided by half; an AEgroup(receiving action effects, e.g. turning on the lights of the smiley on the box when inserting the bar) and a no-AE group(receiving no action effect). Results revealed no differences between the two groups in each age group. In the uncritical trials, where the bar could be translated without any rotation, almost all participants in all age groups grasped the bar with a thumb-up grip. In the critical trials, however, where a 180 degree rotation was required, 13% of the 3-year-olds, 38% of the 4year-olds, 81% of the 5-year-olds, 75% of the 6-year-olds, 88% of the 7-year-olds, 94% of the 8-year-olds, and 100% of the adults used an initial awkward thumb-down grip to end the action comfortably(averaged over both AE- and no-AE conditions; for detailed results please see Table 1). In another most recent study conducted by Jongbloed-Pereboom et al.(2013), a modulated version of the handle-rotation-task was used, adapted from the Crajé, Aarts, et al.(2010) study(see below). In their study, the bar was replaced by a wooden sword, which was presented on a table in front of the participants, in one of six possible orientations. Two orientations represented the critical condition in which only an uncomfortable initial grip resulted in ESC. The remaining four orientations resembled the uncritical condition in which a comfortable initial grip resulted in ESC. Participants had to grasp the handle of the sword and to stick it into a tight hole in a wooden block, which was positioned behind the sword on the table. Thus, this task additionally incorporated a transport phase and an insertion action. To familiarize themselves with the task, each child was asked to demonstrate the insertion of the sword once from the 12 o'clock position. Then, 18 test trials(3 trials for each of the 6 initial positions) were performed in random order. Participants performed the task sitting. 351 participants were tested in one of eight age groups: 3-year-olds(n= 27), 4-year-olds(n= 36), 5-year-olds(n= 44), 6-year-olds(n= 50), 7-year-olds(n= 52), 50

The end-state comfort effect in childhood- a systematic review 8-year-olds(n= 46), 9-year-olds(n= 51), and 10-year-olds(n= 45). Whereas almost all participants used a comfortable grip in the uncritical conditions, 21% of the 3-yearolds, 35% of the 4-year-olds, 41% of the 5-year-olds, 50% of the 6-year-olds, 42% of the 7-year-olds, 55% of the 8-year-olds, 50% of the 9-year-olds, and 62% of the 10year-olds used an awkward initial grip in the critical trials, which resulted in ESC. Wilmut& Byrne(2014b) also examined ESC planning in normally developing children using the handle-rotation-task. A modified version of the task introduced by Rosenbaum and colleagues(1993) was used. Participants were seated in front of a wooden octagon, which was surrounded by eight different colors. The size of the octagon was adjusted to hand size. It could be rotated about its center, so that an arrow could be turned to one of the eight colors on the board. Then, a color, or a sequence of colors, was named and participants were instructed to grasp hold of the octagon and rotate it so that the arrow pointed to the color(s) in the order they were listed. Sequences of one, two, and three colors were used, beginning with the shortest for all participants. Sequences were presented in a blocked rather than in a randomized fashion, making the procedure exactly the same for all participants. The eight color positions were rated by an independent sample using different participants as either comfortable or uncomfortable, similar as done in the study of Rosenbaum and colleagues(1993). Based on these ratings, ESC could be detected for every start- and end- posture of the grasping hand. They tested sixty children within three age groups(each n= 20): 4- to 6- yearolds, 7- to 9-year-olds, and 10- to 12-year-olds. Results showed that ESC increased with age, with 62% of the youngest children, 71% of the 7- to 9-year-olds and 82% of the oldest children showing ESC. The data demonstrated that 10- to 12-year-olds ended movements with the same degree of comfort as adults, but younger children showed movements ending in less comfortable position. 51

The end-state comfort effect in childhood- a systematic review Adalbjornsson et al.(2008) employed the overturned-glass-task to test children`s planning skills. They sat in front of the table with a plastic drinking cup and a pitcher on it. The task was to pick up the cup that was sitting upside down on the table, turn the cup over, and pour water from the pitcher into it. Each child performed three trials, without any familiarization trials before the testing. The task was performed while children sat at the table. Altogether, 40 children(n= 20 per group) were tested. The younger group aged from 2 to 3½ years, the older group from 5 to 6 years. Only 4(20%) of the children in the younger group and 7(35%) of the older group used a thumb-down grip to grasp the overturned cup. According to these results, the majority of 5- and 6-year-old children do not show the ESC effect in the overturned-glass-task. Besides the bar-transport-task, Knudsen et al.(2012) also examined anticipatory planning skills in the overturned-glass-task. At their body midline, participants were presented a pod coaster, which lid up when a cup was placed onto it. To the left and to the right of the coaster, white cardboard circles were placed on the table, with a plastic cup standing on one of the two circles. Participants were asked to put the glass right side up on the coaster. Half of the six trials were preferred-hand trials, the other half was performed with the non-preferred hand. To reach ESC, a thumb-down grip was required followed by a 180 degree rotation of the glass. The same participants were tested as for the bar-transport-task described above. Again, the influence of action effects was examined(when placing the glass upright, the light of the coaster turned on in one group of subjects) but again no difference between groups was obtained. Moreover results revealed no difference between preferred-hand and non-preferred-hand trials. In critical trials, 63% of the 3year-olds, 69% of the 4-year-olds, 82% of the 5-year-olds, 82% of the 6-year-olds, 75% of the 7-year-olds, 100% of the 8-year-olds, and 100% of the adults used an 52

The end-state comfort effect in childhood- a systematic review initial awkward thumb-down grip to end the action comfortably(averaged over both AE- and no-AE conditions; for detailed results please see Table 2). In a most recent study, Scharoun and Bryden(2013) asked participants to complete two different tasks, each involving the manipulation of a cup: The one task required to pick up the cup and pour water from a pitcher into it(bimanual task). The other entailed to pick up the cup and pass it to the experimenter(unimanual task), who then poured water into it. The cup's placement was altered between upright and inverted, and it was 15cm away from the participant(the pitcher was 25 cm away). Neither the pitcher nor the cup had a handle. Each participant completed 12 trials (each trial type three times, e.g. upright or upside down, unimanual or bimanual). Participants were free to use either hand and performed the task sitting. Also, hand preference consistencies were examined due to a second testing session after 24 hours. 112 participants were tested, divided into 6 age groups: 3- to 4-year-olds(n= 13), 5- to 6-year-olds(n= 19), 7- to 8-year-olds(n= 23), 9- to 10-year-olds(n= 17), 11-to 12-year-olds(n= 20), and adults(n= 20). Results revealed a significant main effect of age group in both the upright- and the inverted-cup condition. Collapsed over these two conditions, 10% of the 3- to 4-year-olds, 35% of the 5- to 6-year-olds, 50% of the7- to 8-year-olds, 90% of the 9- to 10-year-olds, 95% of the 11- to 12-yearolds, and 95% of the adults showed sensitivity for ESC planning in the unimanual trials. In the bimanual trials, however, 30% of the 3- to 4-year-olds, 62% of the 5- to 6-year-olds, 81% of the 7- to 8-year-olds, 90% of the 9- to 10-year olds, 95% of the 11- to 12-year-olds, and 98% of the adults showed sensitivity for ESC planning 6 . 6 Please note that no data are provided for the critical trials only, therefore the comparison with other studies is limited. 53

The end-state comfort effect in childhood- a systematic review Recently, Robinson and Fischman(2013) conducted another study using the overturned-glass-task. Participants were instructed to use one hand to pick up a plastic drinking cup that was placed upside-down on a table, turn the cup over, and pour water from a measuring cup into it. Children were tested in two conditions: a novisual cue condition, which was as described above, and a visual-cue condition, where yellow cardboard was glued to the bottom of the drinking cup as well as on the board in front of the cup. Each child performed three trials per condition, and ESC was declared as being present when shown in at least two out of these three trials. Participants were seated in this experiment. They tested 17 preschool children with a mean age of 4.4 years(SD= 4.75 months). The task was similar to the ones used by Adalbjornsson et al.(2008) and Fischman(1997). Out of the 17 children, 8(47.1%) used a thumb-down grip in both conditions, 3(17.6%) used it in only one condition, and 6 children(35.3%) never used it. The visual cue had no effect on planning in this sample. To sum up the results on studies with normally developing children, 9 of the 12 studies provided evidence for the presence of ESC planning in the majority of children beginning from 3 years(Jovanovic& Schwarzer, 2011; Knudsen et al., 2012) up to 12 years of age(Jongbloed-Pereboom et al., 2013; Knudsen et al., 2012; Robinson& Fischman, 2013; Scharoun& Bryden, 2013; Stöckel et al., 2012; Thibaut & Toussaint, 2010; Toussaint et al., 2013; Weigelt& Schack, 2010; Wilmut& Byrne (2014b). The results further suggest an increase of motor planning skills within this age range. Two studies, however, did not find the ESC effect to be present in children of these ages(Adalbjornsson et al., 2008; Manoel& Moreira, 2005). Figure 7 shows a scatter diagram of the results separated for the different tasks used for normally developing children. It can be said, that the developmental trajectories for children tested in the bar-transport-task and in the overturned-glass task are similar, with the ESC effect to be fully present at about 8 years of age. Results of children tested 54

The end-state comfort effect in childhood- a systematic review with the handle-rotation-task, however, show a protracted development of the ESC effect, with the effect being present in only half of the children at 12 years of age. Figure 7. Scatter diagram of the results of the three tasks used to assess ESC planning in normally developing children across all studies reviewed here. 2.3.2.2. Studies comparing normally developing children and children with developmental disorders Hughes(1996) was the first to examine motor planning skills for children with autism(n= 36) and with mild learning deficits(MLD; n= 24), as well as for a control group of normally developing children(n= 28). All groups were further subdivided with regard to their verbal and nonverbal mental age(NVMA), assessed by the matrices task of the British Abilities Scale(BAS; Elliot, Murray,& Pearson, 1983), but not by their chronological age. This resulted in a total of six subgroups:(1) 18 autistic children of 5 years with low NVMA's(score= 4.86);(2) 18 autistic children of 10 years with high NVMA's(score= 9.78);(3) 12 MLD -children of 5 years with low 55

The end-state comfort effect in childhood- a systematic review NVMA's(score= 4.72);(4) 12 MLD children of 7 years with high NVMA's(score= 6.92),(5) 14 normally developing children of 3 years(NVMA score= 3.28); and 14 normally developing children of 4 years(NVMA score= 4.02), according to the BAS. Hughes(1996) used the original version of the bar-transport-task, without any modifications. Before testing, children were given up to five practice trials, or until they appeared confident. Then, all children received a total of eight test trials, half of which were uncritical trials(i.e. an overhand grip resulted in ESC) and the other half were critical trials(i.e. requiring an initial underhand grip for ESC). All trials were presented in random order. Results for the uncritical trials showed that all children used an overhand grip in at least three out of the four trials. For the critical trials, there was an increase in the ESC effect from two children(14%) of the 3-year-olds to ten children(71%) in 4-year-olds of the normally developing group. In the MLD-children, the ESC effect rose from five children(42%) in the group with a low NVMA to six children (50%) in the group with a high NVMA. Similarly, in the children with autism, the ESC effect increased from only one child(6%) in the group with a low NVMA to five children(28%) in the group with a high NVMA(10 years). Thus, Hughes(1996) provided evidence for motor planning deficits in children with developmental disorders. Smyth and Mason(1997) tested 96 children(4 to 8 years old) with developmental coordination disorder(DCD) and compared their performance with 91 normally developing children(4 to 8 years old). All of these primary school children were distributed into four age groups. Motor planning skills were assessed for two tasks, the bar-transport-task and the handle-rotation-task. In the bar-transport-task, Smyth and Mason(1997) used red and blue disks, which were lying behind the supports, and not besides them, as in the original task version. No demonstration was given by the experimenter, but participants received four practice trials. In the 16 experimental trials, each colored end was placed on each disc four times. All conditions were presented in random order. In the handle-rotation-task, the authors modified the original 56

The end-state comfort effect in childhood- a systematic review task version a little, as they replaced the numbers for the target positions with little pictures(depicting a horse, a penny, a clock, an apple, a tree, a clown, a flower, and a car) and mounted the apparatus on a nursery chair. Children were asked to grasp and turn the handle until it covered a specific picture on one of the eight positions. Trials could start from each of the eight positions and involved a handle rotation of 180 degrees. Four blocks of eight trials, each presented in random order, resulted in a total of 32 trials. Children were allowed two practice trials in this task. Unfortunately, the mean values for children`s performance in both tasks were only provided graphically by Smyth and Mason(1997). Therefore, it is not possible to tell the exact number of children(or the proportion thereof), who finished the action comfortable, neither for the bar-transport-task, nor for the handle-rotation-task. For the bar-transport-task, however, the graphic depiction(cf. Figure 4 in Smyth and Mason, 1997) indicates that more than half of the children in each group demonstrated the ESC effect. Any differences between the DCD group and the normally developing group, as well as between the different age groups, were not statistically significant. For the handle-rotation-task, the graphic depiction does also not provide a hint for any differences in planning skills between the DCD group and the group of normally developing children. However, children seem to improve with age. Both observations are supported by the statistical analyses. Most recently, Wilmut and Byrne(2014a) investigated ESC abilities in children with DCD, as well as in adults with DCD. Therefore, a version of the handle-rotationtask was applied, the same as the one describe above(Wilmut& Byrne, 2014b). They tested 20 children with DCD(mean age= 9 years) and 20 typically developing children with the same age. Moreover, 17 adults with DCD with a mean age of 25 years, and 17 age-matched normally developing adults(mean age= 25 years) were tested. Results showed, that in both, children and adult groups, participants with DCD showed significantly less grasps ending in ESC than their typically developing 57

The end-state comfort effect in childhood- a systematic review counterparts. 65% of the normally developing children showed ESC, whereas this amount is decreased to 55% in the DCD group. This result was independent of the length of the movement(i.e. one-step, two-step, or three-step movements, see Chapter 2.3.2.1 in the other study conducted by Wilmut and Byrne, 2014b). Kirpatrick and colleagues(2013) used a version of the handle-rotation-task, but with some differences. The authors used a version adapted from Mutsaarts, and colleagues(2006), which was modified for the use of children and adapted to children's hand size. A plastic disc was placed in fro nt and sloping upwards in the direction away from the participants at an angle of 10 degrees. The disc was surrounded by six target pictures. Participants were seated in front of the apparatus. Then, a target picture was presented at the center of a computer screen, surrounded by moving arrows. Participants were instructed to turn the handle in the direction of the arrows so that the wooden stick pointed to the target picture. Children were given 5 familiarization trials prior to testing. Only 180 degree turns were analyzed in terms of grip selection. Altogether, 76 children with HCP(hemiparetic cerebral palsy) with a mean age of 8.72 years were tested. Results revealed that 61% of the children showed a sensitivity for ESC planning. The side of lesion had no influence on anticipatory planning skills. Van Swieten et al.(2010) assessed motor planning skills in 27 children with DCD(6 to 13 years old) and in 20 children with autism(9 to 14 years old), as well as in 70 normally developing children(5 to 14 years old) and 40 adults(19 to 32 years old). A largely modified version of the handle-rotation-task was used. Specifically, they presented the disk lying flat on the table in front of the participants and explicitly asked them to always grasp the bar with their thumb pointing towards the marked end. The bar could be grasped in two ways: Either with a pronated or a supinated grip. Much different to the original version was that participants were sitting in front of a computer screen. On the screen, a white arrow indicated the direction in which the 58

The end-state comfort effect in childhood- a systematic review bar needed to be rotated, until the thumb was aligned with a red dot indicating the end position. Each rotation movement covered 180 degrees and could be performed clockwise or counterclockwise. Participants were given eight practice trials in which the experimenter made sure that they followed the instructions(e.g. grasp the object with the thumb at the red end and turn it following the white arrow). They were further told that there were always two ways in which they could grasp the bar and were asked to think about which of the two ways they were going to use. Then, participants performed 8 trials for each of the 4 conditions, resulting in a total of 32 test trials. Results for the normally developing children showed that 20% of 5- to 8-yearolds and 50% of the 9- to 14-year-olds demonstrated the ESC effect. In the DCD group, only 10% of the 5- to 8-year-olds and 25% of the 9- to 13-year-olds showed the ESC effect in at least one of the conditions. Further statistical analyses confirmed that the difference in grip selection between younger children with DCD and younger normally developing children, as well as between older children with DCD and older normally developing children was statistically significant. Interestingly, 9- to 14-yearold children with autism showed the same grip selection pattern as normally developing children in this age group(50% ESC grips) and thus, performed much better than the DCD population in this respect. Crajé, Aarts, et al.(2010) tested the ESC effect in 24 children(3 to 6 years old) with cerebral palsy(CP) and 24 normally developing children(3 to 6 years old). These children were equally distributed into four age groups. Crajé, Aarts, et al. (2010) introduced a new task version, similar to the one used by JongbloedPereboom et al.(2013), as described above. In their study, the bar was replaced by a wooden sword, which was presented on the table in front of the participants in one of six possible orientations. According to the authors, two orientations represented critical condition in which only an uncomfortable initial grip resulted in ESC. The remain59

The end-state comfort effect in childhood- a systematic review ing four orientations resembled uncritical condition in which a comfortable initial grip also resulted in ESC. Participants had to grasp the handle of the sword and to stick it into a tight hole in a wooden block that was positioned behind the sword on the table. Thus, this task additionally incorporated a transport phase and an insertion action. To familiarize themselves with the task, each child was asked to demonstrate the insertion of the sword once from the 12 o'clock position. Then, 18 test trials(3 trials for each of the 6 initial positions) were performed in random order. Almost all children in the normally developing group(90% of the 3-year-olds, 95% of the 4-year-olds, and 100% of the 5- and 6-year-olds) inserted the sword with a comfortable posture in the uncritical trials. In the critical trials, however, the ESC effect increased from 10% in 3-year-old, to 30% in 4-year-old, to 65% in 5-year-old, and to 75% in 6-year-old normally developing children. The differences between critical and uncritical trials were significant for 3- and 4-year-old children, but not for the 5- and 6-year-olds. Children with CP used comfortable grip postures in 75% of the 3and 4-year-olds, in 90% of the 5-year-olds, and in 95% of the 6-year-olds in the uncritical trials. In the critical trials, however, only 10% of the 3-year-olds, 20% of the 4year-olds, 30% of the 5-year-olds, and 25% of the 6-year-olds showed the ESC effect. Thus, different from the control group, there was no evidence for an increase of motor planning skills with age in the CP group. Together, the studies on children with developmental disorders provide mixed results and must be further examined for the specific impairment. In children with developmental coordination disorder(DCD), it was shown that they demonstrated the ESC effect in the bar-transport-task(Smyth& Mason, 1997). But, in the handlerotation-task, results were inconclusive. Whereas Smyth and Mason(1997) and van Swieten et al.(2010) were not able to find any sensitivity for ESC planning in their samples, Wilmut and Byrne(2014a) were able to detect ESC planning in children with DCD at the age of 9 years. Children with cerebral palsy(CP) did not show the 60

The end-state comfort effect in childhood- a systematic review ESC effect in the study of Crajé, Aarts, et al.(2010), but Kirpatrick et al.(2013) were able to find ESC sensitivity in children with HCP at the age of 9 years. Only about half of the children with motor learning deficits(MLD) and a high nonverbal mental age(of 7 years) showed ESC planning in the bar-transport-task, whereas numbers were lower for children of a low nonverbal mental age(Hughes, 1996). Finally, in both studies including children with autism, the majority of children did not show the ESC effect, neither in the bar-transport-task(Hughes, 1996), nor in the handlerotation-task(van Swieten et al., 2010). For all studies, it should be also noticed that normally developing children, who served as control groups, did not always outperform those children with developmental disorders. For an overview of all studies and their results, please refer to Table 2. 2.4. Discussion The aim of the present work was to give a systematic overview of research on the development of the ESC effect(as an indication of the presence of motor planning skills) in children. Three databases Medline, PubMed, and Scopus were scrutinized, using a predetermined set of keywords and an 8-step search procedure. A total of 18 published studies were identified, 12 of which assessed ESC planning in normally developing children(Adalbjornsson et al., 2008; Jongbloed-Pereboom et al., 2013; Jovanovic& Schwarzer, 2011; Knudsen et al., 2012; Manoel& Moreira, 2005; Robinson& Fischman, 2013; Scharoun& Bryden, 2013; Stöckel et al., 2012; Thibaut & Toussaint, 2010; Toussaint et al., 2013; Weigelt& Schack, 2010; Wilmut& Byrne, 2014b). The remaining six investigations focused on ESC in atypically developing children, including children with autism(Hughes, 1996; van Swieten et al., 2010), developmental coordination disorder(Smyth& Mason, 1997; van Swieten et al., 2010; Wilmut& Byrne, 2014a), mild learning deficits(Hughes, 1996), and/or cerebral palsy (Crajé, Aarts, et al., 2010; Kirpatrick et al., 2013). Across all of these studies, three different ESC tasks were used: the bar-transport-task(Hughes, 1996; Jovanovic 61

The end-state comfort effect in childhood- a systematic review & Schwarzer, 2011; Knudsen et al., 2012; Manoel& Moreira, 2005; Smyth& Mason, 1997; Thibaut& Toussaint, 2010; Toussaint et al., 2013; Weigelt& Schack, 2010), the handle-rotation-task(Crajé, Aarts, et al., 2010; Jongbloed-Pereboom et al., 2013; Kirpatrick et al., 2013; Smyth& Mason, 1997; van Swieten et al., 2010; Wilmut& Byrne, 2014a; Wilmut& Byrne, 2014b), and the overturned-glass-task(Adalbjornsson et al., 2008; Knudsen et al., 2012; Robinson& Fischman, 2013; Scharoun& Bryden, 2013). Except for Smyth and Mason(1997) and Knudsen and colleagues (2012), who employed both the bar-transport-task and the handle-rotation-task or the bar-transport-task and the overturned-glass-task, respectively, all other studies tested for the ESC effect using only a single task. 2.4.1. Age effects of ESC planning in different tasks Across the studies reviewed here, children's age groups ranged from 1½ years to 14 years of age. The 14 studies that found ESC planning abilities in early childhood observed children ranging between 1½ years to 12 years of age. Given that adults do show ESC planning, it is still an open question why in some studies younger children showed ESC planning(Jovanovic& Schwarzer, 2011; Knudsen et al., 2012), whereas in other studies older children up to 14 years of age did not(van Swieten et al., 2010). These results also contrast with findings of those studies reporting the ESC effect in childhood suggesting a further development in motor planning abilities with increasing age. It is possible that this absence of the ESC effect in older school-aged children is due to the choice of task. These children were assessed with the handle-rotation-task(van Swieten et al., 2010), in which results are inconclusive for younger children, as they showed ESC planning in the more childoriented sword version(Crajé, Aarts, et al., 2010; Jongbloed-Pereboom et al., 2013), but not in the more adult-oriented version(Smyth& Mason, 1997). In another variation of this task, Wilmut and Byrne(2014a) found typically developing children as soon as they reached nine years of age to behave in terms of ESC, but on a very low 62

The end-state comfort effect in childhood- a systematic review level. In another study, however, they found children as young as 4 years of age to behave in terms of ESC in the same task(Wilmut& Byrne, 2014b). Most importantly, the observation of Jovanovic and Schwarzer(2011), suggesting that the ESC effect is not present in children of 1½ years and in children aged 2 years, as well as those of Weigelt and Schack(2010) and Scharoun and Bryden(2013), with the ESC effect absent in children of 3 and 4 years, confirms that such anticipatory planning skills are not an inborn trait, but rather develop over the course of sensory-motor maturation. The span of development ranges from 3 to 12 years of age, with the most noticeable growth spurt around 5 to 8 years in the majority of children. Thus, the development of ESC planning is not reflected by a linear increase of the planning skill, but rather by a positively accelerated function, which exhibits a fast developmental spurt around 6 years of age and then approaches an asymptote somewhere beyond 10 years of age. Deviations from this function may arise from motor reorganization. It is likely that different sensory-motor maturation processes, which support the development of cognitive control during early childhood(Piaget& Cook, 1952), also enable the development of ESC planning(Fischer, 1980). Interestingly, 8-year-old children provided less evidence for anticipatory planning than 6-year-old children in the study of Thibaut and Toussaint(2010), 7-year-old less than 6-year-old children in the study of Jongbloed-Pereboom et al.(2013), and 6-year-old less than 5-year-old in the bartransport-task and 7-year-old less than 6-year-old children in the overturned-glasstask by Knudsen et al.(2012), even if not statistically significant in the latter study. Thibaut and Toussaint(2010) refer to the motor reorganization hypothesis(Bard, Hay,& Fleury, 1990) to explain these contra-intuitive results. According to this hypothesis, motor structures reorganize in children around the age of 8, resulting in a momentary instability of previously acquired abilities. However, motor reorganization did not seem to affect the developmental pattern revealed by Stöckel and colleagues 63

The end-state comfort effect in childhood- a systematic review (2012). Hence, more studies with children of this age group are needed to shed light on the role that motor reorganization plays on the development of anticipatory planning skills. To sum up, the findings of the studies reviewed here differed with regard to the age at which most of the normally developing children show ESC planning. In studies employing various versions of the bar-transport-task, at least half of the children showed the ESC effect at the age of 3½ years(Jovanovic& Schwarzer, 2011), 4 years(Hughes, 1996; Smyth& Mason, 1997), 5 years(Knudsen et al., 2012; Weigelt & Schack, 2010), 6 years(Thibaut& Toussaint, 2010; Toussaint et al., 2013) and/or 7 years(Stöckel et al., 2012). In contrast to these findings, in Manoel and Moreira (2005) only 39% of the oldest age group tested at 6 years of age showed ESC planning. In the three studies using the overturned-glass-task, at least half of the children showed the ESC effect at the age of 3 years(Knudsen et al., 2012), 5 to 6 years for bimanual manipulation or 7 to 8 years for unimanual manipulation(Scharoun& Bryden, 2013). Contrary to these findings, less than half of the children in the oldest age group with 5- to 6 years in the study conducted by Adalbjornsson et al.(2008) showed the ESC effect. Finally, in studies using various versions of the handlerotation-task, Crajé, Aarts, et al.(2010) reported that the majority of 5-year-olds showed evidence for ESC planning, Jongbloed-Pereboom et al.(2013) found 6-yearold children to behave at a rate of 50% in terms of ESC, with the 7-year-olds showing less, the 8-year-olds again more ESC sensitivity, whereas Wilmut and Byrne(2014b) found normally developing children in a modified version of this task to behave to 62% in terms of ESC at the age of 4 years already. In contrast, Smyth and Mason (1997) and van Swieten et al.(2010) did not find evidence for the ESC effect in the majority of their children at 8 years(Smyth& Mason, 1997) and between 9 and 14 years(van Swieten et al., 2010). 64

The end-state comfort effect in childhood- a systematic review In clinical populations, there is evidence for ESC planning in the bar-transporttask in children with MLD at a nonverbal mental age of about 7 years(Hughes, 1996). For children with DCD, the majority showed ESC planning in the bartransport-task already at 4 years of age(Smyth& Mason, 1997). Whereas Wilmut and Byrne(2014a) found 9-year old children with DCD to behave in terms of ESC to an amount of 55%, the majority of children with DCD did not show the ESC effect in the handle-rotation-task by 8 years(Smyth& Mason, 1997) or 9 to 14 years of age (van Swieten et al., 2010). Similarly, the majority of preschool children with CP did not show ESC planning in the handle-rotation-task by the age of 6 or 7 years(Crajé, Aarts, et al., 2010; Kirpatrick et al., 2013). Also, the majority of children with autism did not show ESC planning, neither in the bar-transport-task by a nonverbal mental age of 10 years(Hughes, 1996), nor in the handle-rotation-task by 14 years of age (van Swieten et al., 2010). Overall, the findings of those studies that observed the ESC effect to be present in preschool children suggest a systematic increase in the number of normally developing children across the preschool years. 2.4.2. Other factors constraining the ESC effect in children Given the inconsistencies in findings between the studies reviewed here, both for normally developing children and children with developmental disorders, the potential factors influencing the presence of the ESC effect will be discussed in the remainder of the discussion section. One possible explanation for the different results may relate to differences in task demands between the three tasks employed. More specifically, differences in the number of action steps, in the object rotations, in the precision requirements, the familiarity with the task, the procedures, and in motivation might explain the ambiguous findings, as well as differences in sample characteris65

The end-state comfort effect in childhood- a systematic review tics. Finally, developmental factors will be discussed in relation to the different child populations assessed in the studies reviewed. 2.4.2.1. Number of action steps A possible explanation for the different results concerning the ESC effect relates to the issue of task complexity, which relates to how many actions steps had to be performed to solve a particular action. It may be that more complex tasks reduce children's cognitive resources for ESC planning. With regard to the number of action steps to-be-performed, the bar-transport-task and the handle-rotation-task only require two action steps(grasp and place onto/into target; grasp and turn/insert), as compared to three steps in the overturned-glass-task(grasp, turn, pour). For the task requiring two action steps, the ESC effect was reported to be present in(at least) 5year-old children in a number of studies(Hughes, 1996; Jovanovic& Schwarzer, 2011; Knudsen et al., 2012; Robinson& Fischman, 2013; Smyth& Mason, 1997; Stöckel et al., 2012; Thibaut& Toussaint, 2010; Weigelt& Schack, 2010; Wilmut& Byrne, 2014b). Unfortunately, findings are inconclusive with regard to the more complex three-step overturned-glass-task. Adalbjornsson et al.(2008) were not able to find any sensitivity for ESC planning in the overturned-glass-task. However, Scharoun and Bryden(2013) found the effect to be present in 5-6-year-olds and 7-8year-olds, respectively; Knudsen and colleagues(2012) as soon as 3 years of age, and Robinson and Fischman(2013) with 4 years of age. Given these inconsistent findings, it is still an open question whether any differences in the amount of actions steps needed to solve the object manipulation may explain overall inconsistencies in the reviewed findings. In this regard, it is most interesting that two recent studies conducted by Wilmut and Byrne(2014a, 2014b) were able to show that the ability to plan for ESC decreased with increasing number of actions steps required to complete the task. 66

The end-state comfort effect in childhood- a systematic review 2.4.2.2. Required degree of object rotation The tasks used differed with regard to the required degree of object rotation: (1) The bar-transport-task typically involved a manual rotation over 90 degrees(e.g. Hughes, 1996; Smyth& Mason, 1997; Stöckel et al., 2012; Thibaut& Toussaint, 2010; Toussaint et a., 2013; Weigelt& Schack, 2010), but some task versions were also performed with a rotation of 180 degrees(Jovanovic& Schwarzer, 2011; Knudsen et al., 2012);(2) the handle-rotation-task, or versions thereof, included manual rotations of up to 180 degrees(e.g. Crajé, Aarts, et al., 2010; Jongbloed-Pereboom et al., 2013; Wilmut& Byrne, 2014a, 2014b); and(3) the overturned-glass-task required manual rotations of 180 degrees(Adalbjornsson et al., 2008; Knudsen et al., 2012; Robinson& Fischman, 2013; Scharoun& Bryden, 2013). In this regard, results of studies on adults suggest that the ESC effect found for tasks employing a 90 degree object rotation(e.g., Rosenbaum et al., 1990; Weigelt et al., 2006) is reduced when the task requires an higher amount of object rotation, both for unimanual(Rosenbaum et al., 1993; Hughes, Seegelke,& Schack, 2012) and bimanual object manipulation(e.g., Hughes& Franz, 2008; Hughes et al., 2011; Janssen et al. 2010). Thus, the task demands relative to the degree of manual rotation appear to influence the presence of the ESC effect. The required degree of object rotation may have also affected the performance of children in a number of studies reviewed here. For example, it may explain the absence of the ESC effect for the overturned-glass task in the study conducted by Adalbjornsson et al.(2008), although other studies found the effect to be present in older children(Knudsen et al., 2012; Robinson& Fischman, 2013; Scharoun& Bryden, 2013). Also, the majority of children in the study by Jovanovic and Schwarzer(2011) did not show the effect when the bar had to be rotated over 180 degrees. However, in their study, the absence of the effect can also be explained with the young age of the children(i.e., the effect was not present in children below 67

The end-state comfort effect in childhood- a systematic review the age of 3 years). Using a similar apparatus, Knudsen et al.(2012) were able to find ESC planning in children of 5 years. In the largely modified version of the handlerotation-task used by Crajé, Aarts, et al.(2010), Jongbloed-Pereboom et al.(2013), and Kirpatrick et al.(2013) the ESC effect seemed to develop with age and was only present for older, normally developing children, whereas children as young as 4 years showed anticipatory planning skills in the version used by Wilmut and Byrne (2014b). Together, the degree of manual rotation required to solve a particular task may influence the children`s motor performance. 2.4.2.3. Precision requirements Another constraint to the ESC effect may relate to the precision requirements of the task( precision hypothesis, Rosenbaum et al., 1993). The assumption is that people`s anticipatory planning strategies are related to the amount of precision required to end the task(Rosenbaum et al., 2006). Here, precision refers to how accurate a certain movement(e.g. an object placing action) must be completed in the final part. For example, inserting a particular object into a small target hole may require more precision than simply placing the object on a target circle of similar size. In fact, when the task does not require much precision, ESC is significantly reduced(Rosenbaum et al., 1996). Similar to this latter finding, it has been shown that children need less planning when throwing a ball into a bucket(low precision), as compared to fitting the ball into a small tube(high precision)(Claxton, Keen,& McCarthy, 2003). Nevertheless, these authors demonstrated differences in the kinematic profile of the grasping action, depending on future task demands(high or low precision demands), and revealed that even 10-month old infants show some anticipatory planning skills when considering systematic patterns in these reach-to-grasp profiles. Also, it has been assumed, that finishing manipulations at or near the middle of the pronation/supination range should be easier than at or near the extremes 68

The end-state comfort effect in childhood- a systematic review (Rosenbaum et al., 1996). In essence, people`s precision increases while performing in comfortable postures and movements can be made more quickly within that range of comfort near the neutral-zero point of motion(Rosenbaum et al., 1996; Short& Cauraugh, 1997). In addition, movements are executed more quickly within a comfortable range of motion, known as the middle-is-faster effect(Rosenbaum et al., 1996). For the studies on children reviewed in the current work, the precision requirements varied(especially) for the bar-transport-task. While four studies showed an increase in the ESC effect with age when the task required inserting the bar(high precision) into a target hole at the end of the action(Jovanovic& Schwarzer, 2011; Knudsen et al., 2012; Stöckel et al., 2012; Weigelt& Schack, 2010), Manoel and Moreira(2005) did not find an influence of this factor(i.e. inserting action). However, Hughes(1996) and Smyth and Mason(1997) found the effect to be present in the majority of their 4-year-old normally developing children and Toussaint et al.(2013) in 6-year old children, even though the object only had to be placed on a target(low precision). Also, when precision requirements were higher in the pencil alley task in Thibaut and Toussaint(2010, Experiment 2), older children showed an increase in the ESC effect. Together, findings are inconsistent with regard to the role of precision requirements and more research is needed to examine the influence of this factor on the planning skills of children. 2.4.2.4. Familiarity with the task Previous studies differed also with regard to whether children were allowed to practice the task in familiarization trials before the actual testing began. Whereas Crajé, Aarts, et al.(2010), Hughes(1996), Knudsen et al.,(2012), Jovanovic and Schwarzer(2011), Smyth and Mason(1997), van Swieten et al.(2010), and Wilmut and Byrne(2014b) gave the children the opportunity to get familiar with the task, 69

The end-state comfort effect in childhood- a systematic review Adalbjornsson et al.(2008), Jongbloed-Pereboom et al.(2013), Kirpatrick et al., 2013; Manoel and Moreira(2005), Robinson& Fischman, 2013; Scharoun and Bryden(2013), Stöckel et al.(2012), Thibaut and Toussaint(2010), Toussaint et al., 2013; Weigelt and Schack(2010), and Wilmut and Byrne(2014b) did not give any familiarization trials. In the 11 studies without any familiarization trials 9 studies found the ESC effect to be present. In the 7 studies, where children were allowed to get familiar with the task, 6 studies were able to provide evidence for ESC planning in children. Therefore, children's prior experience with a particular task(i.e. their famil iarity) cannot fully explain the differences in the findings between studies. Also, older children may show ESC planning only because of a higher level of expertise in similar tasks. As many toys for infants include actions, such as inserting objects into holes, but not pouring real liquids into real containers, it may be that young children are not as familiar with pouring water into a glass as older children and adults. Older children are even more familiar with different task requirements the more they are accustomed to different kinds of toys and to how to deal with them. However, though this may hold true for the youngest children in the study of Adalbjornsson et al.(2008), the oldest children in this study should have gained a fair amount of experience with pouring liquids into glasses. In the study by Knudsen et al.(2012), children were not forced to pour any liquids, but just to turn over the cup. Besides the action effect, which appeared when the cup was placed on the coaster(i.e. a light switching on), there was no further maneuver(i.e. a filling action) intended. This may explain the finding, that even 3year-olds showed the ESC effect in their study. These results by Knudsen et al. (2012) are in contrast to those of Scharoun and Bryden(2013), however. They observed significantly more children of 3 to 8 years demonstrating the ESC effect in the bimanual pick-up and pour task than in the unimanual pick-up and pass task. The results of the unimanual condition thus suggest that younger children may not adopt 70

The end-state comfort effect in childhood- a systematic review awkward grips to end an action comfortably, if the task does not require it. Robinson and Fischman(2013) examined only 4-year olds, who showed the ESC effect to an amount of 65%. In any case, it may be that preschool children are more used to be handed a glass in the right orientation(especially if glasses are kept in places children cannot reach for them on their own), than to have to turn it first, such that specifically the turning action may have been less familiar to them. Realizing a pouring action may further influence the presence of ESC. In addition, offering familiarization trials in the beginning of the experiment may enhance the execution of ESC planning strategies, at least when children are younger. 2.4.2.5. Test procedure Whether participants were seated or standing during the testing appears to influence the results. In 13 out of the 18 studies(Adalbjornsson et al., 2008, Crajé, Aarts, et al., 2010; Jongbloed-Pereboom, 2013; Jovanovic& Schwarzer, 2011; Kirpatrick et al., 2013; Manoel& Moreira, 2005; Robinson& Fischman, 2013; Scharoun& Bryden, 2013; Thibaut& Toussaint, 2010; Toussaint et al., 2013; van Swieten et al., 2010; Wilmut& Byrne, 2014a, 2014b) participants were seated. Only four tested them while standing(Knudsen et al., 2012; Smyth& Mason, 1997; Stöckel et al., 2012; Weigelt& Schack, 2010). It is not clear if participants in the study of Hughes(1996) were sitting or standing. ESC was found in all four studies in which children were standing, whereas ESC was observed in 11 of the studies in which children were sitting. It seems reasonable to assume that participants' grasping b ehavior is less constrained as long as they start from a neutral-zero-position(standing upright, with their arms hanging straight down their body, palms touching the legs). However, when they are sitting on a chair in front of the table, the range of arm motion is restricted by the table's surface, which serves as a ba rrier within the reach. It 71

The end-state comfort effect in childhood- a systematic review is possible that younger children try to avoid bumping with their elbows into the table. This may prevent children to use the underhand grip more often. 2.4.2.6. Motivation In relation to the issue of familiarity mentioned above, any differences in the findings may be explained with reference to overall attractiveness of the task. A study conducted by Gollwitzer and Brandstätter(1997) revealed, that those implementation intensions, which are backed up by strong goal intentions, are more effective than implementation intentions without such motivational support. On the other hand, strong goal intentions do not necessarily lead to effective goal pursuit per se, but goal intentions(of the same motivational strength) followed by implementation intentions are more often realized than those without implementation intentions. This means, that choosing an action plan(i.e. acting in terms of ESC or not) seems to be modulated by intentional goals. Therefore, it could be that children have higher implementation intensions and choose an ESC grip to finish comfortably, when an action effect occurs. The actions in the manual tasks were likely less fun to perform than playing to be a pirate and sticking the wooden sword into the tight hole. Motivational factors may also explain why Jovanovic and Schwarzer(2011) found ESC planning in children as young as 3½ years, which is a remarkable finding, whereas other studies did not report ESC planning in much older children. In their modification of the bartransport-task, a salient action effect was added at the end of the action sequence. Action effects, such as lights(Paulus, Hunnius, Vissers,& Bekkering, 2011), play an important role in how children control their actions. It has been shown that children plan and select their actions by anticipating these corresponding action effects(e.g., Elsner& Aschersleben, 2003). In the study by Jovanovic and Schwarzer(2011), children were only able to produce this presumably interesting light effect when they cor72

The end-state comfort effect in childhood- a systematic review rectly inserted the bar into the cylinder(independent of initial grip). This issue was addressed in the study of Knudsen et al.(2012), who wanted to find out if this action effect was the source of children's enhanced performance in the bar -transport-task. Inserting bars into cylinders might be more interesting when the insertion is instrumental for producing an interesting effect. This might have affected the child's overall motivation to perform the task and thereby focusing her attention on the task. They tested different age groups in task versions with and without action effects and did not find any differences between these conditions, suggesting that the action effect may not be the primary source of enhanced performance in the bar-transport-task. It is still possible, however, that the absence of an action effect in other studies using different tasks may explain why children only showed ESC planning at a later age. Further research is needed to address this question. 2.4.2.7. Influence of habitual and goal directed factors. To select actions to solve everyday tasks, a habitual, stimulus-driven and an intentional, goal-directed system are employed by the central nervous system (Herbort& Butz, 2011). The habitual system associates stimuli with responses that were rewarding in the past. The goal-directed system, on the other hand, selects actions dependent on the match of anticipated action outcomes and current needs. For the ESC effect, it has been assumed that grip selection for object manipulation is determined by the goal-directed action selection system, e.g. the posture-based motion planning account(Rosenbaum et al., 2007a). In contrast to this assumption, recent studies, using response-compatibility paradigms, have also provided support for an involvement of the habitual action selection system(e.g. Masson, Bub,& Breuer, 2011; Tucker& Ellis, 1998). Accordingly, the perception of an object automatically evokes actions that are habitually applied to grasp these objects. The presentation of everyday objects may therefore automatically activate those actions that are habitual73

The end-state comfort effect in childhood- a systematic review ly used to grasp them, independent of the current intention of the participant. A recent study by Herbort and Butz(2011) demonstrated that the ESC effect may be compromised for such habitual grasping behavior, when participants were presented with objects of everyday live. It may be that such habitual grasping effects also influenced the planning strategies in some of the children, especially, when the objects used were everyday life objects or objects that children are familiar with(Adalbjornsson et al., 2008; Crajé, Aarts, et al., 2010; Jongbloed-Pereboom et al., 2013; Knudsen et al., 2012; Robinson& Fischman, 2013; Scharoun& Bryden, 2013). Interestingly, five of these six studies using everyday life objects found ESC planning in children. The interaction of the habitual vs. goal-directed grasping systems and its influence of the ESC effect in the development of children`s planning abilities is another interesting field for future studies. 2.4.2.8. Sample size The number of children assessed in the reviewed studies ranged from 17 to 351 children. However, in a number of studies, the sample size of each group tested was quite small(e.g. five to six children in each age group; Crajé, Aarts, et al., 2010; Manoel& Moreira, 2005), such that potentially present age effects were less likely to be detected by most standard statistical procedures. Another constraint could be the different sample sizes for the age groups tested within the studies. For example, van Swieten et al.(2010) tested 27 children with DCD in two age groups, 20 children with autism in only one age group, 70 normally developing children in two age groups, as well as 40 adults in another group. Analyzing the performance of participants in groups of different sample sizes may be problematic as comparisons can be inconclusive. Thus, sample size can be seen as a critical factor in some of the studies reviewed here. 74

The end-state comfort effect in childhood- a systematic review 2.4.3. Parallels of cognitive and motor development Despite these inconsistencies, the studies reviewed above point to a pattern of a gradual improvement for motor planning skills across preschool and school children. It is likely that this developmental pattern is linked to a major growth spurt in the development of more general cognitive functions, occurring in children at about 5 to 6 years of age. This would point to the interdependency between the development of higher cognitive control processes and the maturation of sensory-motor functions(as indicated by advanced motor planning abilities). A number of cognitive abilities are impaired in children with various developmental disorders. Therefore, a comparison between the performance of normally developing children and children with developmental disorders may inform us more about the influence of cognitive planning skills on the ESC effect. The findings of the studies reviewed above, however, provide only mixed evidence. Although prior research demonstrated deficits in children with autism for tasks drawing on executive functions(Geurts, Verté, Oosterlan, Roeyers,& Sergeant, 2004) and complex planning skills(Prior& Hoffmann, 1990). Van Swieten and colleagues(2010) found autistic children to behave in the same manner as normally developing children in the handle-rotation-task. Also, Hughes(1996) showed an increase of the ESC effect from 5% in 3-year-old to 30% in 4-year-old autistic children(but this increase was not as large as in normally developing children). Compared to the controls, autistic children were of similar age or older in terms of their nonverbal mental age in this study (Hughes, 1996). In contrast, the children in van Swieten et al.(2010) had the same chronological age as controls, and thus, might have been even younger than the controls in terms of mental age, respectively. Finally, in contrast to the findings in normally developing children, children with autism were more likely to show the ESC effect in the handle-rotation-task compared to the bar-transport-task, thus showing a pattern opposite to normally developing children. Obviously, these considerations must 75

The end-state comfort effect in childhood- a systematic review be taken with caution due to a lack of studies on children with autism. Given a deficit in executive functioning in autism, it is still an open question whether cognitive planning skills play a crucial role in the development of the ESC effect. Children with MLD(Hughes, 1996) showed an increase from 40% at 3 years to 50% at 4 years in the bar-transport-task. Though, these children showed less ESC planning than normal controls, the 4-yearolds' performance(in terms of percentage of children showing ESC in the bar-transport-task) was comparable(if not even better) to the performance of normally developing children of the same age(Manoel & Moreira, 2005; Smyth& Mason, 1997; Thibaut& Toussaint, 2010; Weigelt& Schack, 2010). With regard to more specific motor deficits, three studies focused on children with DCD(Smyth& Mason, 1997; van Swieten et al., 2010; Wilmut& Byrne, 2014a) and two on children with CP(Crajé, Aarts, et al., 2010; Kirpatrick et al., 2013). Whereas children with autism are well known to have difficulties in cognitive planning, children with DCD may have specific difficulties in motor planning, due to their abnormal development of motor skills, such as poor eye-hand coordination and prehension difficulties. Interestingly, children with DCD did not show anticipatory planning skills in the handle-rotation-task(Smyth& Mason, 1997; van Swieten et al., 2010), but they were able to show this sensitivity in the bar-transport-task(Smyth & Mason, 1997), and in the handle-rotation-task(Kirpatrick et al., 2013). Also, there were no significant differences in performance between children with DCD and normally developing children in Smyth and Mason(Smyth& Mason, 1997). Though, children with DCD showed significantly less ESC planning than controls in van Swieten et al.(2010). But obviously, the percentage of children showing ESC increased in both groups in both studies with higher age. These findings suggest that children with DCD are delayed, but not impaired, in the development of ESC planning. In addition, these findings support the hypothesis that overall task demands are lowest in the bar-transport-task. 76

The end-state comfort effect in childhood- a systematic review Different from DCD, motor disorders in CP are secondary to brain lesions and may therefore affect one side of the body more than the other(unilateral CP). Crajé, Aarts, et al.(2010) reported less ESC planning in the handle-rotation-task for children with CP compared to controls. However, there was an increase in both groups in percentage of children showing the ESC effect in the preschool years. Thus, similar to v an Swieten et al.'s(2010) findings on DCD, ESC planning may be delayed in children with CP whereas performance of normal controls continued to increase after 5 years of age, there was a decrease or stagnation of ESC planning in children with CP between 5 and 6 years of age. Given that the development of ESC planning might not be linear(Thibaut& Toussaint, 2010) and given the findings for children with DCD, it is still an open question whether the majority of children with CP would show ESC planning at an older age. Unfortunately, the study conducted by Kirpatrick et al.(2013) did not include any control group. Overall, the findings of the few studies on anticipatory planning in children with motor impairments are largely inconclusive. Both conditions assessed in the studies reviewed cover heterogeneous disorders and may show comorbidities also with cognitive deficits. The motor impairment in DCD may occur in isolation or with other cognitive deficits, such as learning disabilities or attention deficit disorders. Similarly, the motor disorders of CP are often accompanied by disturbances of sensation, cognition, and perception, among others(Rosenbaum, Paneth, Leviton, Goldstein,& Bax, 2007). Future research is therefore needed to investigate the relative roles of cognitive processes and motor skills in the development ESC planning. 2.5. Conclusion Twelve out of the 18 studies reviewed found the ESC effect to be present in normally developing children across different ages(although to different extends). The emergence of such anticipatory planning skills is consistent with the findings of 77

The end-state comfort effect in childhood- a systematic review studies employing the same tasks with adults(Rosenbaum et al., 1990; Rosenbaum & Jorgensen, 1992). It is still an open question why evidence for anticipatory planning in childhood was found in some studies(Crajé, Aarts, et al., 2010; JongbloedPereboom et al., 2013; Jovanovic& Schwarzer, 2011; Kirpatrick et al., 2013; Knudsen et al., 2012; Robinson& Fischman, 2013; Scharoun& Bryden, 2013; Smyth& Mason, 1997; Stöckel et al., 2012; Thibaut& Toussaint, 2010; Toussaint et al., 2013; Weigelt& Schack, 2010; Wilmut& Byrne, 2014a, 2014b), but not by others (Adalbjornsson et al., 2008; Hughes, 1996; Manoel& Moreira, 2005; van Swieten et al., 2010). Possible reasons for inconsistencies in the findings may relate to differences in task demands, such as precision requirements, number of action steps to be performed, familiarity, and motivation. Despite the different findings regarding the age of emergence and the developmental trajectory, the studies reviewed suggest a marked increase in ESC planning around the end of the preschool years, as well as a further increase throughout childhood. Further research is needed to assess the development of ESC planning and to determine the relative influence of motor skills and cognitive factors on its developmental course. 2.6. Implementations Based on the findings on the ESC development in the studies reviewed above, it becomes clear that the developmental trajectory of the ESC effect differs between the populations tested. This may be due to the tasks employed to test children's se nsitivity for ESC. The tasks used were variants of the original version of the bartransport task(Rosenbaum et al., 1990), the handle-rotation task(Rosenbaum, Vaughan, Jorgensen, Barnes and Stewart(1993), and the overturned-glass task (Fischman, 1997). Most relevant for the present project: Some studies found an early 78

The end-state comfort effect in childhood- a systematic review onset 7 of ESC development in normally children(as early as 3 years of age, e.g. Jovanovic& Schwarzer, 2011; Knudsen et al., 2012), other studies found the effect to be present at a later stadium of child development(as late as 10 years of age, e.g. Jongbloed-Pereboom et al., 2013; Thibaut& Toussaint, 2010; Scharoun& Bryden, 2013), or even were not able to find any sensitivity for ESC planning in the examined ages(e.g. until the age of 5 year, Manoel& Moreira, 2005; until the age of 7 years, Smith& Mason, 1997; or even until the age of 14 years, van Swieten et al., 2010). For children with developmental disorders, the onset may be even further delayed, e.g. in MLD(Hughes, 1996), autism(Hughes, 1996; van Swieten et al., 2010), in DCD(Smyth& Mason, 1997; van Swieten et al., 2010; Wilmut& Byrne, 2014a), and in CP(Crajé, Aarts, et al., 2010; Kirpatrick et al., 2013). Above, many factors potentially modulating the development of the ESC effect, like age effects, the number of actions steps needed to accomplish the task, the required degree of object rotation, precision requirements, the familiarity with the task, task procedures, motivational aspects, and sample size were discussed. These factors may have influenced the time of onset of anticipatory planning skills. This concludes the literature review and thereby the first research aim has been fulfilled. The second research aim is to examine the possible influence of executive functions(EF) on the developmental trajectory of the ESC effect. Both, EF and the ESC effect rely on planning abilities, which develop in children over the course of their cognitive and sensorimotor maturation. This raises the question whether the development of EF interacts with the development of motor planning abilities, as signified by the ESC effect. The existence of such a relationship has been proposed recently(Scharoun& Bryden, 2013; Stöckel et al., 2012; van Swieten, et al. 2010; 7 ESC is assumed to be present, if more than 50% of the participants in one age-group showed the effect reliably, i.e. in more than 50% of the trials. If, for example, more than 50% of 5-year-olds showed the effect, but less than 50% of the 6-year-olds, planning for ESC is declared as being present from 7 years on, as the number of participants then stays consistently over 50%. 79

The end-state comfort effect in childhood- a systematic review Wilmut& Byrne, 2014b), which is further supported by the similarity of the developmental trajectories of EF and the ESC effect(see Figure 4 for the projected developmental trajectory of the different EF, and the Table 2 for an overview of the developmental trajectories for the different ESC tasks). Therefore, the second aim of the research project is to investigate the possible influence of EF on the developmental trajectory of the ESC effect. To this end, an experimental study examining the development of motor planning abilities and cognitive skills in children from 3- to 10-years of age was conducted. The experimental study will be reported in the following Chapter. 80

Chapter 3 Chapter 3 3. Motor planning and executive functions across young ages This chapter is based on the publication of Wunsch, K., Pfister, R., Henning, A., Aschersleben, G.,& Weigelt, M.(under review). The development of motor planning skills and executive functions across young ages. In prep. 81

Motor planning and executive functions across young ages 3.1. Introduction Differences in grasp selection between adults and children have been construed as indicating a deficit in children's planning(see Hughes, 1996; Smyth& Mason, 1997), with the presence of ESC as an indicator for"thinking ahead", referring somewhat ambiguously to some kind of planning skills. This led to an important debate, initially raised by van Swieten et al.(2010), about whether performance in grasp selection tasks is driven by executive planning(i.e. actively planning ahead to solve actions correctly or to avoid mistakes, for example) or motor planning(i.e. planning motor actions in advance in order to solve them correctly or most economically), or both(Scharoun& Bryden, 2013; Stöckel et al., 2012). Van Swieten and colleagues(2010) argued that the ESC effect cannot fully rely on executive planning, because adults do not consistently select grasps, which end in comfortable positions. Thus, if executive planning was the driving force, executive functions would fail on some trials, but not on others, which is unlikely to be the case. Instead, van Swieten and colleagues(2010) proposed that grasp selection relies predominantly on pure motor planning processes and that the most efficient movement is selected for each grasp. On the other hand, it is commonly assumed that motor skills are very similar to intellectual skills in terms of acquisition and representation(Rosenbaum et al., 2001), which suggests the exact opposite of the aforementioned argument. In this regard, it has been assumed for a long time, that perceptual-motor skills and intellectual skills have closely related developmental origins, as already noted by Piaget and Cook (1952), who based the development of intelligence upon the emergence of skilled action. Both, EF and the ESC effect are associated with planning. EF is an umbrella term that incorporates a collection of inter-related processes underlying purposeful, goal-directed behavior(Gioia et al., 2001). These executive processes are essential for the formation and maintenance of goals and strategies, preparation for action, 82

Motor planning and executive functions across young ages and verification that plans and actions have been implemented appropriately(Luria, 1973). Processes associated with EF are numerous and commonly include anticipation, goal selection, planning, initiation of activity, self-regulation, mental flexibility, deployment of attention, and utilization of feedback(Anderson, 2002). Executive processes develop with the biological maturation of the frontal cortex throughout childhood and adolescence in a multi-stage process(Stuss, 1992), that is, with different developmental trajectories and rates for different functions(Passler et al., 1985). Therefore, EF play an important role in a child's cognitive functioning, behavior, em otional control, and social interaction, but develop at different rates and at different times, differently for children's individual development. Thi s raises the question whether EF development may predict or influence the development of motor planning, as signified by the ESC effect(see also Scharoun& Bryden, 2013; Stöckel et al., 2012). The similarity of the developmental trajectories of EF and the ESC effect suggests a potential relationship in development(see Anderson, 2002, for the projected developmental trajectory of the different EF(depicted in Figure 4), and Chapter 2, for an overview of the developmental trajectories for the different ESC tasks (see Table 2)). Therefore, the following study investigates the possible role of EF on the developmental trajectory of the ESC effect. To this end, eight groups of children and a group of adults were with examined within three ESC tasks, in order to assess their motor planning skills and in three EF tasks to measure their cognitive planning skills. The following predictions were made:(1) an increase of task performance in each subtest as children get older,(2) inter-correlations between the ESC tasks, as well as (3) positive correlations in each age group between participants' performance on ESC and EF tasks. 83

Motor planning and executive functions across young ages 3.2. Method 3.2.1. Participants Nine age groups with a total of 217 participants were recruited. For a detailed overview of p articipant's demographics please see Table 3. All children were recruited from local daycare centers, elementary schools or via announcements in a local newspaper in Paderborn; all adults were students at the University of Paderborn. Handedness was determined using a short test, detecting how they threw a ball, used a spoon, and wrote/drew with a pencil. For all groups of children, parents provided their informed consent for participation and for video recording their child during the experiment. Participation was voluntary, without any financial compensation. Children received a personal certificate of participation and some sweets. 3.2.2. Tasks and Procedures A test-battery was designed to assess ESC planning and EF, consisting of three different tasks each: the bar-transport-task, the sword-rotation-task, and the grasp-height-task for ESC, and the Tower-of-Hanoi-task(TOH), the Mosaic-task, and the D2-attention-endurance-test for EF. For each task, the respective materials were placed on a table in front of the participant, except for the grasp-height-task, in which the shelf was placed next to the table. Participants stood in front of the table for the bar-transport-task and for the sword-rotation-task, and in front of the shelf for the grasp-height-task. In all EF tasks, they sat at the table, with seat height adjusted for each participant. Table height was 55 cm for kindergarten children and 75 cm for school children and adults. To adjust for differences in body height, preschoolers smaller than 110 cm stood on a 10 cm high podium, and school-aged children smaller than 120 cm or 130 cm stood on a 20 or 10 cm high podium. Dividing these heights(body height plus podium height) by table height revealed comparable heightcoefficients between preschoolers, school-aged children, and adults(please 84

Motor planning and executive functions across young ages Table 3. Demographic overview of the sample. 85

Motor planning and executive functions across young ages see Table 3). A camera was positioned 150 cm besides the participant, at a height of 160 cm, and recorded the whole experiment for later coding. Participants were tested individually by one experimenter. For several preschoolers, also a teacher or a parent was present in order to make the child feel com fortable. Prior to testing, adult participants or children's parents completed a short questionnaire on handedness, on how they completed their way to kindergarten/school/university, and on leisure time activities, sport participation, and spoken languages. Afterwards, children completed a short test to determine handedness as the hand that was used in at least two out of the three tasks. This hand was marked with an ink stamp. In the ESC tasks, children were instructed to always use the "stamp hand". The ESC tasks were run without familiarization trials. The order of all six tasks was randomized across participants to prevent effects of serial order. In the ESC tasks, participants stood in front of the table or the shelf, 10 cm away from the edges, respectively, or sat in front of the table with the respective apparatus on it for the EF tasks, with materials 10 cm away from the edge of the table. On average, the entire session lasted between 90 and 100 minutes. The duration differed according to participant's time needed to complete the different tasks and according to requests for breaks between the subtests. In general, most of the adults were able to complete the testing session in about 75 minutes, whereas some children needed up to 150 minutes to complete the whole session. 3.2.2.1. Motor tasks used to assess ESC In all three motor tasks, participants stood in front of the table or the shelf, 10 cm away from the edge behind a starting line, the apparatus or the middle platform at body midline, with their arms hanging on each side of their body, the palm facing their upper legs(hereafter: starting position). Participants were instructed to always 86

Motor planning and executive functions across young ages use their dominant hand("stamp hand", see above) for all grasps made in the ESC tasks. Bar-transport-task: In this study, a modified version of the original bartransport-task(Rosenbaum et al., 1990) was used, which was similar to the one employed by Weigelt and Schack(2010). Different to the original task version, there was only one target at midline in front of the apparatus(and not two targets on either side as in the original task). A wooden bar, 20 cm long, with one black and one white end rested horizontally on two cradles, 15 cm above the table. A 5 cm high, black cylindrical container served as the movement target and was placed 10 cm in front of the support. To k eep precision requirements comparable across age groups, the bar's diameter measured 1.5, 2.0, or 2.5 cm for preschoolers, school-children and adults, respectively, and the target hole diameter measured 2.0, 2.5, or 3.0 cm for preschoolers, school-children and adults, respectively(see Figure 8). Figure 8. Apparatus used for the bar-transport-task. Left the standard start position(orientation of the bar was counterbalanced), right the different sizes of bars and target holes used for different ages. The start orientation of the bar(i.e. black or white end on the right side) was counterbalanced across participants and remained the same throughout the experiment. Participants were instructed to take up the starting position, to then grasp the bar firmly with their"stamp hand", and to insert the black or white end of the bar into 87

Motor planning and executive functions across young ages the target hole, as indicated by the experimenter. After the insertion action was completed, they were instructed to get back to the starting position. To prevent observational learning, the experimenter used a pincer grip at one end of the bar to reposition the bar back on the two cradles. Participants completed six trials in randomized order, three trials for each end. They could use either an overhand or an underhand grip to grasp the bar. This resulted in either an upright(i.e. thumb-up) or an inverted (i.e., thumb-down) hand position at the end of the movement(see Figure 9), depending on the start-orientation of the bar. Figure 9. Both pictures show a critical trial, where the white end of the bar had to be inserted into the target hole, and the possible outcomes. Upper part: An initially comfortable overhand posture results in a finally uncomfortable thumb-down posture. The child therefore does not behave in a manner consistent with ESC. Lower part: An initial grip adaption to a more uncomfortable underhand position results in a finally comfortable end position with the thumb pointing upwards. In the three uncritical trials, an overhand grip automatically resulted in a comfortable thumb-up position; in the three critical trials, however, an underhand grip was necessary to end in the comfortable thumb-up position and therefore, in ESC. Grasp 88

Motor planning and executive functions across young ages choice was coded from the video. Following recent studies(Adalbjornsson et al., 2008; Stöckel et al., 2012; Weigelt& Schack, 2010), we considered the ESC effect to be present if in a given condition(critical and uncritical trials) at least two out of the three trials ended in the comfortable position. Sword-rotation-task: A variation of the original rotation task(Rosenbaum et al., 1993) was used, which was similar to the task versions created by Crajé, Aarts, et al. (2010) and Jongbloed-Pereboom et al.(2013). A wooden sword(30 cm in length, 3.2 cm in width, and 0.8 cm in height; handle length= 10 cm) was horizontally placed on a platform(47 x 47 cm) in front of a target box in one of six start positions(Position 1 = 0°(12 o'clock position), Position 2= 90°, Position 3= 135°, Position 4= 180°, Pos ition 5= 225°, and Position 6= 270°). The sword's blade had to be inserted into a tight fitting hole in a wooden block(47 cm in length, 16 cm in width, and 16 cm in height; hole 3.5 x 1 cm)(see Figure 10). The same apparatus was used for all age groups. Figure 10. Apparatus for the sword-rotation-task. The sword is in the start position(Position 1). Positions are numbered clockwise. 89

Motor planning and executive functions across young ages Again, participants were instructed to take up the starting-position and were told that they were a pirate(to adults this task was explained without the cover story) and that they had to insert the sword into the box by firmly grasping the handle in exactly the position it laid on the table, that is, without turning the sword before grasping it. Each session started with Position 1 to make sure participants understood the task. The experimenter retrieved the sword from the box and repositioned it on the platform always by grasping it at the cross guard to avoid observational learning. The task consisted of three blocks of 6 trials(one for each position), resulting in a total of 18 trials. Trial positions were randomized in each block. Participants could choose grips that resulted in either a comfortable end position(ESC; with the thumb pointing towards the blade) or an uncomfortable end position(no ESC; with the thumb pointing away from the blade)(see Figure 11). Figure 11. Both pictures show a critical trial(Position 2) and the possible outcomes. Upper part: An initially comfortable overhand posture results in a finally uncomfortable posture with the thumb pointing away from the blade. The child therefore does not behave in a manner consistent with ESC. Lower part: An initial grip adaption into a more uncomfortable underhand position results in a comfortable end position with the thumb pointing towards the blade. 90

Motor planning and executive functions across young ages Within each block, two trials were critical. Here, grasping the sword in a more uncomfortable hand position(at Positions 2 and 3) resulted in a comfortable end position. Grasp choice was coded from the video. We considered the ESC effect to be present if at least 4 out of the 6 critical trials ended in a comfortable position. Grasp-height-task: An adaptation of the original grasp-height-task(Cohen& Rosenbaum, 2004) was used, which was similar to the version employed by Weigelt and colleagues(2007). Participants had to transport a standard toilet plunger from a home platform to a lower or higher target platform, and back to the home platform. Board heights were individually adjusted to participants' body height by taking the board heights and average height of adult participants reported in Rosenbaum and colleagues(2006) as reference. For example, for an 87 cm tall child, shelf heights were 25.4, 43.2, and 61.0 cm for the low, middle and high shelf, respectively. On each shelf board, a wooden platform was attached in such a way that it protruded 15 cm from the shelf. The home platform was attached to the horizontal center of the middle shelf board. The two target platforms were attached to the participant's side of handedness, one on the low and the other on the high shelf board. The toilet plunger stood on the home platform: a circular rubber base(10 cm in diameter and 5 cm high) supported the cylindrical wooden shaft(2.5 cm in diameter and 33 cm or 44 cm in length for child or adult participants, respectively, see Figure 12). 91

Motor planning and executive functions across young ages Figure 12. Apparatus used for the grasp-height-task. The plunger is in the start position. Shelf heights were scaled to participants body height. Participants were told to take up the starting position. Child participants were told that their performance was videotaped in this task in order to program a robot afterwards that could then perform the same actions as they did and children therefore needed to closely follow the instructions. Participants' task was to stand in the start position, to grasp the plunger firmly on the shaft and to transport it to the platform indicated by the experimenter(home-to-target moves). Afterwards, they had to resume the starting position and then bring the plunger back to the home platform (target-to-home moves). Participants were told that home and target platform could differ across trials. Participants completed a total of six trials: the plunger was transported from the home platform to the high target platform and back to the home platform in three trials, and from the home platform to the low target platform and back to the home platform in another three trials. Conditions were always blocked, with the start platform for the first three trials counterbalanced across participants. Grasp height was coded from the video. Participants showed planning for ESC if they 92

Motor planning and executive functions across young ages grasped the plunger lower for the home-to-target moves than for the target-to-home moves for the high platform, and higher for the home-to-target moves than for the target-to-home moves for the low platform(see Figure 13). Figure 13. End-State Comfort in the grasp-height-task. ESC was considered to be present if participants showed this pattern in at least two out of the three trials for each platform. Also, mean differences in grasp heights were computed for later analyses. 3.2.2.2. Cognitive tasks used to assess EF In all three cognitive tasks, participants sat at a table, on a height-adjustable chair. Materials were placed in front of them, according to the particular test manual. Tower-of-Hanoi-task: We used a slightly modified version of the original task (Simon, 1975), which was similar to Welsh(1991), but used only one apparatus as in the original version. It consisted of 3 pegs in a row(height: 23 cm, diameter: 4 cm, 93

Motor planning and executive functions across young ages distance between pegs: 15 cm) that were attached on a bottom plate. On these pegs, up to five discs of varying size and color could be located: black(diameter: 13 cm), blue(diameter: 11 cm), green(diameter: 9 cm), red(diameter: 7 cm), and yellow(diameter: 5 cm). The target position was indicated by a picture, displayed at a 75 degree angle on a music stand, 20 cm behind the apparatus(see Figure 14). The peg on the very right was the target peg, and was marked with a black duct tape at the top. Participants were instructed to build a tower of discs as shown by the target position on the picture, starting from the arrangement presented. Each participant was to solve up to ten different tower problems, with increasing degree of difficulty. Depending on the number of discs and on the starting arrangement, trials differed in the minimal number of moves necessary to complete the tower. Three test versions were created with difficulty adjusted to age: a version for 3- and 4-year-olds, for 5and 6-year-olds, and for 7-year-olds and older children and adults. Every test started with a familiarization trial with three discs(2, 3 or 6 moves according to the test version), in which rules were explained and questions could be clarified. Participants had to follow three rules:(1) move only one disc at a time,(2) a disc may only be in your hand or on a peg, but not on the table or somewhere else, and(3) a smaller disc can be placed on top of a bigger disc, but a bigger disc cannot be placed on top of a smaller disc. In test trials, each starting position was initially covered by placing a cardboard in front of the apparatus to assess latency. The task was terminated whenever participants were not able to solve a tower problem in up to twice the minimal number of moves necessary to solve the problem, or if participants did not move any disc within 90 seconds. The single tower problems included 3 or 4 discs in the two easier test versions, and 4 or 5 discs in the most difficult version. Minimal number of moves necessary ranged between 2 and 15 for 3- and 4-year-olds, between 4 94

Motor planning and executive functions across young ages and 15 for 5- and 6-year-olds, and between 7 and 31 for children aged 7 and older, and adults 8 . Figure 14. Apparatus used for the Tower-of-Hanoi-task. The Tower shown on the picture needed to be re-built on the peg marked with the black tape. Start- and end-positions of the discs were checked from the video and the number of steps to complete the tower was counted. Number of tower problems completed correctly served as dependent measure. Mosaic-task: This task is a subtest of the Wechsler Scale of Intelligence 9 . In the Mosaic-task, a given mosaic pattern has to be re-created with a set of building 8 For each age group, problem difficulties were as follows, whereby the first number always indicates the number of discs used, the second number always indicates the number of moves for optimally solving the problem: 3-2, 3-3, 3-4, 3-5, 3-6, 4-7, 4-8, 4-9, 4-10, 4-11, and 4-15 for 3- and 4-year olds; 3-4, 3-5, 3-6, 4-7, 4-8, 4-9, 4-10, 4-11, 4-13, 4-15 for 5- and 6-year olds; and 4-7, 4-9, 4-10, 4-11, 413, 4-15, 5-20, 5-24, 5-27, and 5-31 for children older than 7 years and adults. 9 The Mosaic-task is a subtest of the HAWIVA®-III(Hannover Wechsler Intelligenztest für das Vorschulalter; German translation and adaption of the WPPSI®-III(Wechsler Primary and Preschool Scale of Intelligence) of David Wechsler(2002); Ricken, Fritz, Schuck,& Preuß, 2007; for preschool children), the HAWIK®-IV(Hamburg Wechsler Intelligenztest für Kinder; German translation and adaption of the WISC®-IV(Wechsler Intelligence Scale for Children) of David Wechsler(2003); Petermann& Petermann, 2008; for school children) and the WIE®(Hamburg Wechsler Intelligenztest für Erwachsene; German translation and adaption of the WAIS®-III(Wechsler Adult Intelligence Scale) of David Wechsler(1997); von Aster, Neubauer,& Horn, 2006; for adults) to assess visuospatial and motor skills. 95

Motor planning and executive functions across young ages blocks. The target picture or the 3D model was positioned 18 cm away from the edge of the table, 10 cm to the side of participants' body midline on the oppos ite side of handedness. Participants had to arrange up to nine cubes(side length: 2.5 cm) of different colors(all red sides, all white sides or red and white sides; see Figure 15). Figure 15. Apparatus used for the Mosaic-task. The pattern displayed on the picture should be rebuild with the several cubes. The test was administered in accordance with the respective test manual and age group. Completion time was coded from the video and performance was checked for accuracy. The percentage of scored points was calculated as indicated in the respective test manual, and served as dependent measure for this task. D2-attention-endurance-task: Three versions of this speeded test of selective attention(Brickenkamp, 1962) were used: the analogous subtests in the Intelligence and Development Scales for preschool children(IDS-P; Grob, Reimann, Gut,& Frischknecht, 2013) and for school children(IDS, Grob, Meyer,& Hagmann-von Arx, 2009), and the D2-R for adults(Brickenkamp, Schmidt-Atzert,& Liepmann, 2010). Preschool children`s task was to sort cardboard cards(6 x 6 cm) showing a duck, according to the presence or absence of a distinct characteristic. A pencil, lying 25 96

Motor planning and executive functions across young ages cm away from the child at the side of their handedness indicated where to stack the cards with the given characteristic(see Figure 16). Figure 16. Set-up of the D2-task according to the IDS-P for children aged 3 to 5 years. Cards had to be sorted due to a specific pattern. In the paper-pencil version for school children, participants were presented with a DIN A3 sheet of paper showing rows of ducks with or without distinct characteristics. Children had to mark the ducks with the target characteristics(see Figure 17). 97

Motor planning and executive functions across young ages Figure 17. Set-up of the D2-task according to the IDS for children aged 5 to 10 years. Ducks had to be sketched out due to a specific pattern. In the paper-pencil version for adults, rows of the letters p and d were presented on a DIN A4 sheet of paper, and letters with distinct characteristics had to be marked(see Figure 18). Videos of preschool children were checked to verify the number of properly sorted cards. Performance was scored in accordance with the respective test manual. According to the test manual, the total number of scored points served as dependent measure. Figure 18. Set-up of the D2-task according to the D2R adults. All" p' s" had to be sketched out due to a specific pattern. 98

Motor planning and executive functions across young ages 3.2.3. Data Analysis/ Scoring Chi-Square tests were used to examine group differences in the ESC tasks due to the dichotomous nature of the dependent variables. For the EF tasks 10 , oneway ANOVAS were used to examine these differences. To test for age effects, regression analyses were applied. Bivariate Pearson's correlations were computed in order to find possible relationships within the ESC tasks and the EF tasks, and between these motor and cognitive tasks. 3.3. Results This section will be divided into three major parts, according to the specific research goals:(1) the development of ESC,(2) the relationship within the tasks measuring ESC, and(3) the correlation between ESC and EF. 3.3.1. The development of ESC In the following section, the results of the three tasks used to measure anticipatory planning are depicted. 3.3.1.1. Bar-transport-task Figure 19 illustrates the mean percentage of participants in each group who showed ESC planning in at least two out of three trials in each condition. In the uncritical trials(e.g. trials where the right side of the bar had to be inserted into the tar10 We measured different dependent variables for the Mosaic and the Tower of Hanoi tasks, e.g. the total amount of correctly solved items, the total score and the score with bonus-points for fastest resolve in the Mosaic task, and the amount of incorrect moves, the number of correctly solved items or the number of the item where the first mistake was made before doing the first move in the Tower of Hanoi task. Due to their correlations(Mosaic-task: r=.229 for correctly solved items and total score; r =.194 for correctly solved items and bonus-score, and r=.950 for total score and bonus-score; all p< .01; Tower-of-Hanoi-task: r=.261 for the amount of incorrect moves and the item with the first mistake, r=.149 for the amount of incorrect moves and the number of correctly solved items, and r=.218 for the item with the first mistake and the number of correctly solved items; all p<.05), however, we chose the most superordinate variable for further analysis. 99

Motor planning and executive functions across young ages get hole for right-handers), all participants in all age groups adopted an overhand grasp in at least two out of the three uncritical trials and therefore ended in a comfortable end position. In the critical trials, only 24% of the 3-year old children showed sensitivity for ESC planning, this amount increased to 48% in the 4-year-olds, to 62% of the 5-year-olds. Here, a stagnation of the developmental trajectory can be seen in 5-to-8-year old children: 64% of the 6-year-olds, 62% of the 7-year-olds, and 63% of the 8-year-olds showed ESC behavior. Then, the amount of participants showing ESC increased again from 78% in the 9-year-olds, to 95% in the 10-year-olds, up to 96% in adults. A chi-square analysis showed these differences in the proportion of children showing ESC in the critical trials to be significant, ²(7)= 34.93, p<.001. Figure 19. Percentage of the participants in each group showing end-state comfort in at least two out of three trials in the uncritical(initial overhand grip) and critical(initial underhand grip) conditions in the bar-transport-task. A regression analysis of the percentage of participants showing ESC in at least two out of three critical trials across the child groups revealed the developmental trajectory to be statistically significant with a slope of 7.8% per year, t(6)= 5.622, p 100

Motor planning and executive functions across young ages =.001. The entire regression model including the intercept yielded an adjusted R²= .814, F(1,7)= 31.608, p<.001. Single chi-square four-field tests revealed that all children groups up to the age of 9 years behaved significantly less often in terms of ESC than adults did; ²(1)= 27.93, p<.001 for the 3-yearolds; ²(1)= 15.71, p< .001 for the 4-yearolds; ²(1)= 10.12, p<.01 for the 5-yearolds; ²(1)= 8.98, p< .01 for the 6-yearolds; ²(1)= 10.12, p<.01 for the 7-yearolds; ²(1)= 9.62, p<.01 for the 8-yearolds; ²(1)= 4.02, p<.05 for the 9-year-olds). There was no difference in behavior between 10-yearolds and adults(²(1)= 0.04, p>.05). Considering all children, a Kruskal-Wallis Test revealed the differences between the age groups to be statistically significant across all three trials(Trial 1 ² (7) = 28.62, p<.01; Trial 2 ² (7)= 27.71, p<.001; Trial 3 ² (7)= 26.76, p<.001). This shows the distinct developmental trajectory for ESC planning. We also investigated whether children change their grip behavior across the trial repetitions and thus, exhibit short-term learning effects over the course of the three critical trials. Figure 20 depicts the percentage of children in each age group performing in a manner consistent with ESC in each of the three critical trials. 101

Motor planning and executive functions across young ages Figure 20. Percentage of participants showing end-state comfort in each of the three critical trials in the bar-transport-task. Whereas 3-year-olds, 7-year-olds, and 10-year-olds showed an increase in ESC grasps with increasing trial number(from 14% to 29% to 33% in 3-year-olds; from 58% to 62% to 65% in 7-year-olds; and from 86% to 95% to 100% in 10-yearolds), 4-year-olds showed the same amount of ESC grasps in all trials(48%). Fiveyear-olds increased in ESC grasp performance from Trial 1 to Trial 2 and Trial 3 (from 42% to 62%, respectively), in the same way as 9-year-olds did(from 70% to 83%, respectively). Six-year-olds and 8-year-olds showed ESC planning more often in Trial 2 than in Trial 1, but then the percentage of ESC conform grasps decreased again in Trial 3(from 55% to 64% to 59% in the 6-year-olds, and from 59% to 74% to 63% in the 8-year-olds). Thus, there is no systematic pattern detectable, so that short-term learning effects may not have occurred across these three trials. 102

Motor planning and executive functions across young ages 3.3.1.2. Sword-rotation-task Figure 21 illustrates the mean percentage of participants in each group who showed ESC planning in the uncritical conditions(i.e. Positions 1, 4, 5, and 6) and in the critical conditions(i.e. Positions 2 and 3). In the uncritical trials, all participants adopted a grip with the thumb being oriented towards the blade, and thus, ended in a comfortable position when inserting the sword into the box. In the critical trials, however, only 43% of the 3- and 4-year old children showed sensitivity for ESC planning, this amount increased to 58% of the 5-year-olds, to 64% in the 6-year-olds. Again, similar to the bar-transport-task, even if more delayed, a stagnation of the developmental trajectory can be detected in 7-to-10-year old children. Here, 73% of the 7year-olds, 74% of the 8-year-olds, 78% of the 9-year-olds, and 76% of the 10-yearolds showed ESC behavior. All adults used an ESC conform grasp in at least four out of the six critical trials. A chi-square analysis showed these effects in the proportion of children showing ESC in the critical trials to be marginally significant, ²(7)= 12.89, p=.075. Figure 21. Percentage of participants in each group showing end-state comfort in the sword-rotation task in at least seven out of the twelve uncritical trials and in at least four out of the six critical trials. 103

Motor planning and executive functions across young ages A regression analysis of the percentage of participants showing ESC in at least four out of six critical trials across the child groups revealed the developmental trajectory to be statistically significant with a slope of 5.5% per year, t(6)= 7.156, p< .001, adjusted R²=.878, F(1,7)= 51.209, p<.001. Single chi-square four-field tests revealed that all children groups behaved significantly less often in terms of ESC than adults did; ²(1)= 21.19, p<.001 for the 3-yearolds; ²(1)= 21.24, p<.001 for the 4-yearolds; ²(1)= 14.88, p<.001 for the 5-yearolds; ²(1)= 12.12, p<.001 for the 6-yearolds; ²(1)= 8.66, p<.01 for the 7-yearolds; ²(1)= 4.24, p<.05 for the 8-yearolds; ²(1)= 6.75, p<.01 for the 9-yearolds; and ²(1)= 7.42, p<.01 for the 10-year-olds). Considering all children, a Kruskal-Wallis Test did not consistently reveal the differences between the age groups to be statistically significant across all three trials in Position 2(Trial 1 ² (7)= 7.22, p>.05; Trial 2 ² (7)= 10.47, p>.05; Trial 3 ² (7) = 15.58, p<.05) and in Position 3(Trial 1 ² (7)= 29.97, p<.001; Trial 2 ² (7)= 11.44, p>.05; Trial 3 ² (7)= 16.95, p<.05). We also investigated whether children change their grip behavior across the trial repetitions and thus, exhibit short-term learning effects over the course of the six critical trials. Figure 22 depicts the percentage of children in each age group performing in a manner consistent with second-order motor planning in each of the three critical trials, in the upper graph for Position 2, in the lower graph for Position 3. A closer examination of the data did not reveal a systematic pattern. It is not the case that children started to plan for ESC with increasing trial number. Therefore, like in the bar-transport-task, short-term learning effects may not have improved participants performance in terms of ESC. 104

Motor planning and executive functions across young ages Figure 22. Percentage of the participants showing end-state comfort in each of the three critical trials for Position 2(upper picture) and Position 3(lower picture), respectively. 105

Motor planning and executive functions across young ages 3.3.1.3. Grasp-height-task Figure 23 depicts the percentage of participants who demonstrated ESC in the grasp-height-task. Results showed, that 19% of the 3-year-olds, 35% of the 4-yearolds, 19% of the 5-year-olds, 32% of the 6-year-olds, 38% of the 7-year-olds, 33% of the 8-year-olds, 39% of the 9-year-olds, 45% of the 10-year-olds, and 93% of the adults showed ESC planning for the high target moves. In the low target moves, however, 5% of the 3-year-olds, 4% of the 4-year-olds, 8% of the 5-year-olds, none of the 6-year-olds, 4% of the 7-year-olds, 22% of the 8-year-olds, 26% of the 9-yearolds, 14% of the 10-year-olds, and 50% of the adults showed ESC planning. Thus, ESC planning seems to be more frequent when bringing the object to high positions than to low positions. But, even 10-year-old children showed only half as many grasp behaviors in terms of ESC as adults do. This hints to a rather protracted development of ESC planning for tasks exploiting a continuous task space. Figure 23. Percentage of the participants showing end-state comfort in the grasp-height-task for the high target position(light grey bars) and the low target position(dark grey bars). The light grey line indicates means of grasp height in both conditions; the dashed line shows the regression line. 106

Motor planning and executive functions across young ages A chi-square analysis revealed that there are significant differences between the groups in the amount of participants showing ESC averaged across both positions(²(7)= 120.59, p<.001). A regression analysis of the mean differences in grasp height across the child groups revealed the developmental trajectory to be statistically significant with a slope of 1.81 cm per year, t(6)= 12.986, p<.001, adjusted R²=.878, F(1,7)= 166.011, p<.001. Single chi-square four-field tests revealed that all children groups behaved significantly less often in terms of ESC than adults did (²(1)= 34.24 for the 3 -yearolds; ²(1)= 27.21 for the 4 -yearolds; ²(1)= 36.86 for the 5-yearolds; ²(1)= 3 0.49 for the 6-yearolds; ²(1)= 27.34 for the 7 -year-olds; ²(1)= 20.95 for the 8 -yearolds; ²(1)= 15.32 for the 9 -yearolds; and ²(1)= 17.70 for the 10-year-olds, all p<.001). For further analysis, a score was computed for the mean differences in grasp height across both target positions(see Figure 24). The mean grasp heights for home-to-target and for target-to-home moves were computed for both, high target and low target trials. For both, the differences of the mean grasp heights were computed. The means from these two differences were multiplied by(-1). This resulted in the following distribution: Mean grasp-height difference for 3-year-olds was-14.55, for 4-year-olds-14.71, for 5-year-olds-12.92, for 6-year-olds-10.77, for 7-year-olds 8.61, for 8-year-olds-5.35, for 9-year-olds-4.66, for 10-year-olds-3.51, and 4.68 for adults. These values are the dependent variable for all further correlational analyses. 107

Motor planning and executive functions across young ages Figure 24. Mean differences in grasp height averaged for the high-target-moves and the low-targetmoves. Error bars represent confidence intervals for the distribution of grasp height. Considering all children, a Kruskal-Wallis Test revealed the differences between the age groups to be statistically significant in trials one and three for the high target platform(Trial 1 ² (7)= 17.07, p<.05; Trial 2 (7) ²= 11.37, p>.05; Trial 3 ² (7)= 22.09, p<.01) and in all trials for the low target platform(Trial 1 ² (7)= 21.78, p<.01; Trial 2 ² (7)= 25.48, p<.001; Trial 3 ² (7)= 28.98, p<.001). Again, it was investigated whether participants change their grip behavior across the trial repetitions and thus, exhibit short-term learning effects over the course of the six trials. Figure 25 depicts the percentage of children in each age group performing in a manner consistent with second-order motor planning in each of the three trials, in the upper graph for the high target platform, in the lower graph for the low target platform. The results show a similar pattern as observed in the above mentioned tests. There can be no systematic pattern in grasp choice found. Therefore, like in the bar108

Motor planning and executive functions across young ages transport-task and in the sword-rotation task, short-term learning effects may not have improved participants performance in terms of ESC. Figure 25. Percentage of participants showing end-state comfort in each of the three trials for the high (upper picture) and the low target platform(lower picture). 109

Motor planning and executive functions across young ages 3.3.1.4. Intercorrelations of the ESC tasks The results above show that there are similar developmental trends for ESC performance within each single motor task. When plotting the results of the three ESC tasks similar as is done in Figure 7 in the systematic review(Chapter 2.3.2.1.), it is clearly observable that the development of ESC planning is delayed for the graspheight-task. Whereas ESC seems to be fully present in children of ten years in the bar-transport-task, the development of ESC planning in the sword-rotation-task seems to be protracted into adolescence, as ten year old children show ESC only to an amount of about 80%. Figure 26. Scatter diagram of the results of the three tasks used to assess ESC planning in normally developing children in the experimental study. 110

Motor planning and executive functions across young ages Consequently, the question arises whether the different measure of ESC performance are interrelated. For this reason, we computed Pearson's pairwise bivariate correlations between the three motor tasks(see Table 4). Table 4. Pairwise Pearson's correlations between the three motor tasks. Note that two cor relations for the adult sample could not be computed due to ceiling effects in the sword-rotation-task. Small to medium correlations emerged when considering the entire sample (last row of Table 4), even though these correlations are mostly driven by the between-group differences described above. By contrast, there was no sign of intercorrelations of the different ESC measures within the groups(except for the correlation of the bar-transport-task and the sword-rotation-task for the 4-year-olds). To further assess these within-group differences, we Z-transformed each correlation coefficient for each of the non-adult groups, averaged these transformed values and retransformed the resulting values to correlation coefficients. This procedure yielded mean correlations of r=.14 for the bar-transport-task and the sword-rotation-task, r= .07 for the bar-transport-task and the grasp-height-task, and r=.06 for the sword111

Motor planning and executive functions across young ages rotation-task and the grasp-height-task. Testing the corresponding mean Z-values against zero did not yield any significant differences, p s>.193. 3.3.1.5. Summary The results in the motor task support previous findings indicating that children's ability for antic ipatory planning increases with age. Whereas only 24% of the youngest children showed ESC in the bar-transport-task, this amount increased up to adult-like level of 95% in 10-year old children. In the sword-rotation-task, a similar pattern emerged. Here, as many as 43% of the youngest children showed ESC, whereas the development until an adult-like amount progresses more slowly, as only 76% of the 10-year-olds showed ESC planning. Therefore, the development seems to have an earlier onset in this task, but may be delayed somewhat further into adolescence. In the grasp-height-task, anticipatory planning seems not to be seen as an efficiency constraint, as children as well as adults did not show the grasp-heighteffect reliably. 3.3.2. The development of EF In the following section, the results of the three EF tasks are reported. As the comparison across age groups is difficult due to different variable outcomes(see Chapter 3.3.3.), only descriptive data will be presented. 3.3.2.1. The Tower-of-Hanoi-task Figure 27 illustrates the mean percentage of correctly solved items in each group in the Tower-of-Hanoi-task. Results showed that 3-year-olds solved 13% of all items correctly, 4-year-olds solved 24%, 5-year-olds solved 46%, 6-year-olds solved 54%, 7-year-olds solved 42%, 8-year-olds solved 48%, 9-year-olds solved 49%, 10year-olds solved 57%, and adults solved 66% out of the ten items. Please note that task difficulty increased with increasing item number, and that 3- to 4-year-olds(light 112

Motor planning and executive functions across young ages grey bars), 5- to 6-year-olds(dark grey bars), and participants 7 years and older (black bars) received different test protocols. Figure 27. Percentage of correctly solved items in the Tower-of-Hanoi-task. The shading of the bars indicates different versions of the test protocol used according to the age of participants. 3.3.2.2. The Mosaic-task Figure 28 shows the mean percentage of points(relative to the respective total score) in the Mosaic-task for each group. Results showed that 3-year-olds reached 38% of the possibly total score, 4-year-olds reached 49%, 5-year-olds reached 71%, and 6-year-olds reached 80%. These groups accomplished the easier version of the task(see light grey bars in Figure 28 ). In the school children's version, 7 -year-olds reached 51%, 8-year-olds reached 69%, 9-year-olds reached 75%, and 10-year-olds reached 86%(see dark grey bars in Figure 28). In the adult version, participants reached 68% of obtainable points. Even if preschool children aged between 3 and 6 years and 11 months had used the same protocol, the total reachable score would have differed, as children aged 3 to 4 years started with Item 1, whereas 4-year-olds started with Item 7, and 6 year-olds started with Item 10. Test protocol changed at age 6, depending on whether children were still in kindergarten or already in school. Preschool children used the same protocol as younger children, whereas 6-year old 113

Motor planning and executive functions across young ages school children received the more difficult version, in which they started with Item 1, whereas children aged 8 years and older started with Item 3. The two different bars in Figure 28 for 6-year-olds correspond to whether children were attending kindergarten or school. Figure 28. Percentage of points reached depending on the total score in the Mosaic-task. The shading of the bars indicates different versions of the test protocol used according to the age of participants. 3.3.2.3. The D2-attention-endurance-task For the D2-attention-endurance-task, Figure 29 illustrates the mean percentage of points reached depending on the total score in each group. Results showed that 3-year-olds reached 36% and 4-year-olds reached 48% of the possibly total score in the card-sort D2 from the IDS-P(see light grey bars in Figure 29). Five-yearolds reached 28%, 6-year-olds reached 37%, 7-year-olds reached 45%, 8-year-olds reached 53%, 9-year-olds reached 64%, and 10-year-olds reached 71%(see dark grey bars in Figure 29) in the D2-task from the IDS. In the adult version, participants reached 43% of obtainable points . 114

Motor planning and executive functions across young ages Figure 29. Percentage of points reached depending on the total score in the D2-task. The shading of the bars indicates different versions of the test protocol used according to the age of participants. 3.3.2.4. Intercorrelations of the EF tasks As for the ESC tasks we executed Pearson's pairwise bivariate correlations in order to assess intercorrelations between EF tasks(see Table 5). Table 5. Pairwise Pearsons's correlations between the three cognitive tasks. Note: Overall correlations could not be computed as different test protocols were used for age groups. 115

Motor planning and executive functions across young ages There were only a few significant correlations within the age groups, specifically for the D2- and the Tower-of-Hanoi-task in 3-year-olds and the D2- and the Mosaic-task in 6- and 8-year-olds and adults. The large majority of the correlations, however, was not significant. 3.3.2.5. Summary The results in the cognitive task support previous, findings indicating that children tend to increase their performance with increasing age(Anderson, 2002). Between-group comparisons were impeded by the different measures obtained in the different test versions used for the different age groups. In the Tower-of-Hanoi-task only 10-year-olds consistently solved more than half out of the ten items correctly. These results are in line with previous work(Welsh, 1991), showing an increasing ability in solving more difficult problems with increasing age. In the Mosaic-task, children as young as 5 years were able to gain more than half of the total points. Therefore, visual perception seems to develop at a similar stage as ESC did in the bar-transport-task and in the sword-rotation-task. Unfortunately, there are no standard-values given in the manuals of the three different test batteries for the Mosaic-task. To the best of our knowledge, previous studies used the entire test batteries to assess IQ, and therefore did not provide any data of the mosaic-test per se. In the D2-attention-endurance-test children from the age of 8 years reached 53%(43%) of obtainable points(please find the standard-values from the testmanuals in parentheses behind the result). Compared to the standard-values, participants' attention levels were above average in the current study. 116

Motor planning and executive functions across young ages 3.3.3. The relationship between ESC and EF Prior to the examination of possible relationship between motor planning and executive functioning, Table 6 provides an overview over the results from the different motor and cognitive tests, separated by age groups. 117

Motor planning and executive functions across young ages Table 6. Detailed results of all age groups in the six different tasks. 118

Motor planning and executive functions across young ages The main purpose of this examination was to assess a possible relationship between motor planning and executive functions. Therefore, Pearson 's correlations were computed between all motor and cognitive tasks(see Table 7). Table 5. Pairwise Pearson's correlations between the cognitive and the motor tasks. The analyses yielded mostly small and non-significant correlations across the participants of each individual group, except for the following: performance in the D2attention-endurance-task correlated significantly with the grasp-height-task in the 5year-olds and 9-year-olds, the Mosaic-task correlated with the bar-transport-task in 10-year-olds, with the sword-rotation-task in the 8-year-olds, and with the graspheight-task in 5-year-olds, whereas the Tower-of-Hanoi-task was only correlated with the bar-transport-task in 4-year-olds. In contrast to the ESC tasks, overall correlations across the participants of different groups are not possible for the EF tasks, because we opted to use different versions of the tasks for different age groups. In other words: As cognitive functions develop during childhood, it was not possible to use only one test for every single 119

Motor planning and executive functions across young ages cognitive function. In all tasks, two or three different versions of each test were used to examine executive functions as described in the methods section. As for the intercorrelations of the ESC tasks, however, we computed mean correlations across all non-adult groups and tested the resulting mean Z-score against zero. This procedure yielded a significant correlation between the mosaic task and the bar-transport-task, r =.218, p=.018, whereas the remaining correlations were not significant p s>.082. 3.4. Discussion The goal of the present study was three-fold: to examine(1) the developmental trajectories of the different tasks,(2) possible relationships between all motor tasks used, and(3) potential relations between the performance in both, the motor and the cognitive task, specifically whether EF predict ESC planning. To this end, a specific test battery examined the development of motor planning skills and EF in children and adults. To examine motor planning skills, the bar-transport-task, the sword-rotation-task, and the grasp-height-task were conducted. To test EF, different, age-related versions of the Tower-of-Hanoi-task, of the Mosaic-task, and of the D2attention-endurance-test were used. This test battery was employed to assess the performance of eight groups of children, aged 3 to 10 years, and one group of adults. With regard to the developmental trajectories observed for the presence of motor planning skills(as indicated by the ESC effect), the results support previous studies using the bar-transport-task(Hughes, 1996; Jovanovic& Schwarzer, 2011; Knudsen et al., 2012; Smyth& Mason, 1997; Stöckel et al., 2012; Thibaut& Toussaint, 2010; Toussaint et al., 2013; Weigelt& Schack, 2010) and the sword-rotationtask(Crajé, Aarts et al., 2010; Jongbloed-Pereboom et al., 2013). Specifically, the propensity to perform a certain motor action in a manner consistent with the ESC effect(steadily) increased from young kindergarten children to school children, and adults. Thus, children's motor planning skills develop as they get older. For the bar 120

Motor planning and executive functions across young ages transport task, adult-like performance was reached around the age of ten years(see also Knudsen et al., 2012; Stöckel et al., 2012; Thibaut& Toussaint, 2010). For the sword-rotation task, this development seems to be somewhat delayed, as even 10year old children showed the ESC effect less often than adults(see also JongbloedPereboom et al., 2013). In addition, the results of the grasp-height-task suggest that this basic developmental trend across young ages generalizes from a rather dichotomous grip selection(underhand vs. overhand) to grasp choices in a continuous task space(here, along the vertical axis of the object). However, the age at which participants display adult-like performance in these kinds of tasks appears to be much further protracted, as only about half of the 10-year old children showed the ESC effect in the grasp-height task. Participants' performance in the cognitive tasks is also in line with previous findings(Anderson, 2002). In general, EF matured with age, as indicated by the increase of children's task performance as they get older. For the Tower -of-Hanoi task, only the oldest children of the test-protocols for 5-6-year-olds, and participants aged 10 and older in the most difficult task version were able to solve more than half of the disc-transfer problems correctly in the given difficulty(6-year-olds solved 54%, 10year-olds solved 57%). For the Mosaic-task, 5- and 6-year-olds reached more than half of the attainable total points for the easier task version(71% for the 5-year-olds, 80% for the 6-year-olds), and all of the older children were able to accomplish more than half of the test as well(51% of the 7-year-olds, 69% of the 8-year-olds, 75% of the 9-year-olds, and 86% of the 10-year-olds). Moreover, in the D2-task younger children aged 3 and 4 years were not able to reach more than half of all total points as in the Tower-of-Hanoi task mentioned above. Here, children as old as 8 years solved more than 50% of the items correctly(8-year-olds reached 53%, 9-year-olds reached 64%, and 10-year-olds reached 71%). 121

Motor planning and executive functions across young ages As for the second goal of the study, the correlation analyses did not provide much support for intercorrelations between motor tasks. Although, correlations emerged when analyzing the data across the entire sample, no such relationship was found within any of the different age groups(except for the correlation between the bar-transport-task and the sword-rotation-task in 4-year-olds). This pattern of results is similar to that found in two previous studies 11 (Knudsen et al., 2012; Smyth& Mason, 1997), which tested the same children in two different motor tasks within one single experiment. In all other studies conducted so far, only a single ESC task was used. The current work therefore extends this previous line of research and is the first to investigate the development of motor planning skills by using three different tasks in a within-subjects design. There are several possible reasons why performance in the motor tasks was not correlated. This outcome may be due to differences in the number of required action steps, in the precision requirements, in children's perception of comfort, the required degree of object rota tion, in children's familiarity with the task, and/or in m otivational aspects(for a more detailed discussion of these influencing factors see Chapter 2.4.). To examine possible interdependencies between the development of motor planning and EF was the third goal of the present study. Mean correlations in child participants were between r=-.013 and r=.218 and therefore, motor performance 11 As the first author of this paper was a co-author in the study by Knudsen and colleagues(2012), data was checked post-hoc for relations between the two ESC tasks used in either study(the graspheight-task was excluded due to the continuous nature of the dependent variable). Phi-coefficient analyses revealed similar results: In the study by Knudsen and colleagues(2012), the bar-transporttasks and the overturned-glass-task was only correlated in 4-year-old children( =.522, p<.05), In our study, however, we found both tests to be correlated in the 4- and 5-year-olds( =.476, p<.01 and =.440, p<.01), but not in the other participant groups. This means, that the two tasks used in our study and the two tasks used in the study by Knudsen and colleagues(2012) possibly measure different aspects of motor planning. Hence, the developmental trajectories of ESC planning may not be related between these different tasks, possibly due to different task constraints, as already discussed in Chapter 2. 122

Motor planning and executive functions across young ages and EF were not related for this set of tasks. The notion that motor skill development may not be as closely related to the maturation of EF has been previously assumed and put forward by van Swieten et al.(2010). These authors argued that if adults do not always perform in a manner consistent with ESC, then EF must fail in these cases. However, it is also plausible that EF may sometimes fail under certain circumstances, even in adults, as has been shown, for example, by Blakemore and Choudhury(2006), de Luca and colleagues(2003) or Salthouse, Atkinson and Berish (2003). An alternative explanation for the null-findings in the present correlational analyses may relate to the specific EF and motor tasks used. It may be that the cognitive processes necessary to succeed in the Tower-of-Hanoi task, the Mosaic task, and the D2-attention-endurance-task are not or only in part required in the bartransport-task, the sword-rotation-task, and the grasp-height-task. It is still an open question whether the performance in other EF tasks may be stronger related to the ESC task used here, and whether performance in other motor tasks may be predicted by the EF tasks used here. In fact, a recent study by Gonzalez and colleagues (2014) reported a correlation of performance between hand and space use(in a grasp-and-place and in a grasp-and-built-task) and EF for 5- to 10-year old children. A more elaborated research program across a larger number of cognitive and motor tasks is certainly warranted to further exploit the links between the cognitive and the motor development of young children. 3.5. Conclusion In summary, the present study examine d children's performance in three o bject manipulation tasks and compared their performance with three cognitive tasks that measured EF. There was a clear developmental trajectory for all abilities examined here(motor planning skills, as well as problem solving skills, visuo-spatial abili123

Motor planning and executive functions across young ages ties, and attention). Contrary to the predictions made, findings showed only weak and unreliable intercorrelations between the different motor tasks. In addition, the performance in the cognitive tasks used to test EF did n ot reliably predict participant's pe rformance in the different ESC tasks. Future research is needed to further assess potential interdependencies between motor skill development and the maturation of cognitive abilities. Specifically, the findings of the current work suggest to carefully constructing test batteries with respect to the abilities underlying different cognitive and motor tasks. 124

Chapter 4 Chapter 4 4. General Discussion 125

General Discussion In the foregoing chapters, an introduction was given to motor behavior, cognitive control, and the development of both. Then, in the review study(Chapter 2) and in the experimental study(Chapter 3), the two aims of the research project were considered. The first aim was addressed in the review study in Chapter 2. The literature overview provided detailed insights into the existing studies on the ESC effect in children. The various tasks and procedures employed were compared to highlight differences and similarities between these studies and the different developmental patterns of ESC planning in normally developing children and in children with developmental disorders. Moreover, possible factors influencing the development of ESC planning were discussed and new directions for future research were provided. Altogether, 18 studies investigating the presence of the ESC effect in children have been identified. Twelve of these studies assessed ESC planning in normally developing children(Adalbjornsson et al., 2008; Jongbloed-Pereboom et al., 2013; Jovanovic& Schwarzer, 2011; Knudsen et al., 2012; Manoel& Moreira, 2005; Robinson& Fischman, 2013; Scharoun& Bryden, 2013; Stöckel et al., 2012; Thibaut & Toussaint, 2010; Toussaint et al., 2013; Weigelt& Schack, 2010; Wilmut& Byrne, 2014b). The remaining six studies focused on the ESC effect in children with different developmental disorders, including children with autism(Hughes, 1996; van Swieten et al., 2010), children with developmental coordination disorder(Smyth& Mason, 1997; van Swieten et al., 2010; Wilmut& Byrne, 2014a), with mild learning deficits (Hughes, 1996), and with cerebral palsy(Crajé, Aarts, et al, 2010; Kirpatrick et al., 2013). Across these studies, three different kinds of basic tasks(or versions thereof) were employed: the bar-transport-task, the handle-rotation-task, and the overturnedglass-task. Results were inconsistent. Some studies found ESC planning in children as young as 3 years(Jovanovic& Schwarzer, 2011), others did not find ESC planning to be present until the age of 10(Thibaut& Toussaint, 2010), while some studies were 126

General Discussion not able to find the effect at all in the children tested(e.g. Manoel& Moreira, 2005). The findings were independent of developmental consistency(i.e. if children were normally developing or with developmental disorders). Possible reasons for the differences in the findings may relate to the particular task demands, such as precision requirements, the number of action steps to be performed, familiarity, and motivation. Despite the different findings regarding the age at which the effect emerged and the developmental trajectory, the studies presented in the review study in Chapter 2 suggest a marked increase in ESC planning around the end of the preschool years, as well as a further increase throughout childhood. The experimental study reported in Chapter 3 was conducted to shed more light on this topic. It was examined how the ability for ESC planning develops during childhood and how this development relies on the development of EF. A test battery consisting of six sub-tests, three of which tested motor planning and three tested executive functions in children aged 3 to 10 years, was employed. A total sample of 217 participants was examined. Results revealed that the majority of 5-year-olds showed ESC planning in the bar-transport-task and in the sword-rotation-task(more than 50% of all children showed ESC in more than 50% of the trials). In the grasp-heighttask, only the majority of adults showed anticipatory planning skills. Here, the developmental trajectory seems to be distinctly protracted in children, compared to the two other motor tasks. At the same time, only small to medium correlations were found between the three motor tasks, considering the whole sample, but there was no sign of intercorrelations of the different ESC measures within the age groups. The correlational analyses between the behavior in the ESC tasks and in the EF tasks yielded mostly small and non-significant correlations across the participants of each individual group, with few exceptions: Performance in the D2-attention-endurance-task correlated significantly with the grasp-height-task in the 5-year-olds and 9-year-olds, the Mosaic-task correlated with the bar-transport-task in 10-year-olds, with the swordrotation-task in the 8-year-olds, and with the grasp-height-task in 5-year-olds, and 127

General Discussion the Tower-of-Hanoi-task was only correlated with the bar-transport-task in 4-yearolds. Therefore, on a first sight, this study did not find any relationship of between the development of motor and executive planning abilities in children. Taken together, results showed(1) a positive developmental trajectory for each of the sub-tests, with better task performance as children get older,(2) that the performance in the separate tasks was not correlated across participants, and(3) that there was no relationship between performance in the motor tasks and in the cognitive tasks used in the present study. It is obvious, that the ability to plan for ESC develops fastest in the bartransport-task, followed by the sword-rotation-task(see Figure 26). These results are in line with what was observed in the review study(Chapter 2). In the grasp-heighttask, which was never examined in children before, a very protracted development of ESC planning is observable. In the following section, possible reasons for the different developmental trajectories and the lack of a relationship between the tasks will be discussed. 4.1. Constraints on the development of anticipatory planning abilities In the experimental study, the results of the systematic review can be confirmed, concerning the age effects in the examined ESC tasks. A positive developmental trajectory was examined in all three tasks, although with different slopes. In the following paragraphs, possible constraints will be discussed, which may have influenced the developmental trajectories in the different tasks. 4.1.1. Motor reorganization The argument of motor reorganization was confirmed, as children aged 7 and 8 years showed a stagnation(or even a decrease in the grasp-height-task) of ESC 128

General Discussion planning development in these ages. These findings are supported by other studies on ESC(e.g. Thibaut& Toussaint, 2010), who found 6-year-olds to perform better in the bar-transport-task than 8-year-olds, and studies on other tasks measuring goaldirected movements, like in visual aiming(e.g. Bard et al., 1990). These authors found that 8-year-olds moved a lever more slowly and with less accuracy than 6year-olds. They assumed, that this demonstrates, that 8-year-olds need more feedback control than other age groups. In another study, Meulenbroek and van Galen (1988) described reorganization processes in writing, as they observed a decrease in writing speed and an increase in writing dysfluency between 8 and 9 years of age. They suggest that older children try to improve writing quality using visual feedback. Other reorganization processes take place at this age in drawing(e.g. Lange-Küttner, 2009). However, the assumption that this stagnation may be due to cognitive maturation processes can be denied, as there is no major spurt in the development of EF measured in the developmental study. The review study suggested to find ESC planning in the majority of children in the bar-transport-task first at ages between 3 ½ and 7 years, which was confirmed by the data of the experimental study for children of 5 years of age(in line with the findings of Knudsen et al., 2012 and Weigelt& Schack, 2010). The handle-rotation-task seemed to have a protracted development based on the review study, which can partly be confirmed in our study. We found the majority of children showing the ESC effect at the same age as in the bar-transport-task, but with a shallower slope. Whereas planning strategies seem to be almost fully present in 10-year-olds, only 76% of the children in this age group showed the effect in the sword-rotation-task. Therefore, the development of ESC planning seems to be protracted into adolescence, or at least after the age of 10 years in the sword-rotation-task. For the graspheight-task, one cannot compare the present results to another study, because anticipatory planning skills in a continuous task with object translation have not been in129

General Discussion vestigated in children before. The only study, which examined grasp height in children so far, was conducted by Jovanovic and Schwarzer(2011). In their version of the bar-transport-task, with the bar standing in an upright position at the beginning of each trials, they measured grasp heights for baseline(translation) and reverse(rotation) trials. They found, that children grasped the bar at a higher point in the baseline condition than in the reverse condition. The results of the experimental study in Chapter 3 suggest, however, that ESC planning in a continuous task space seems not to be that predictable as it is in the other tasks, where the presence of ESC continuously increased with age. Even adults did not show the effect reliably. 4.1.2. Degree of object rotation Results of studies on adults suggest that the ESC effect found for tasks employing a 90 degree object rotation(e.g. Rosenbaum et al., 1990; Weigelt et al., 2006) is reduced, when the task requires an higher amount of object rotation, both for unimanual(e.g. Rosenbaum et al., 1993; Hughes et al., 2012) and bimanual object manipulations(e.g. Hughes& Franz, 2008; Hughes et al., 2011; Janssen, Crajé, Weigelt,& Steenbergen, 2010). The motor tasks used in the experimental study (Chapter 3) differed regarding the required degree of object rotation: there was no rotation needed in the grasp-height-task, a 90 degree rotation in the bar-transporttask, and up to 180 degree rotations in the sword-rotation-task. The aforementioned assumption can be confirmed for the two tasks which required rotation, as the regression slope for the bar-transport-task is steeper than for the sword-rotation-task (see Figure 19 and 21 in Chapter 3.3.1.). The developmental trajectories of the bar-transport-task and the swordrotation-task are in line with previous research(e.g. Crajé, Aarts, et al., 2010; Weigelt & Schack, 2010). It is still an open question, however, why the behavior in both tasks is not related in participants of the different age groups. One possible explanation is that the underlying mechanisms for 90 degree rotations and higher degree rotations 130

General Discussion are different. It may be, that planning for bigger rotations may need more cognitive capacity, which may not be available in all children younger than 10 years of age. Cognitive development occurs in spurts, as suggested by cognitive development theorists like Piaget(1936) for example. The possible underlying abilities develop at different times, at different rates, and interindividually different. If the underlying mechanisms for both, low rotation tasks and higher rotation tasks are not the same, it may be that they develop at different times in maturation. In some children, one skill develops before the other, and vice versa in other children. 4.1.3. Precision requirements Another constraint for the ESC effect to show up may relate to the precision requirements of the task( precision hypothesis, Rosenbaum et al., 1993). The assumption is that people`s anticipatory planning strategies are related to the amount of precision required to end the task(Rosenbaum et al., 2006; Rosenbaum et al., 1996; Short& Cauraugh, 1999). Rosenbaum and colleagues(1996) confirmed the prediction that the ESC effect relies on precision requirements, whereas other studies were not able to support this assumption(e.g. Hughes et al., 2012). In their study, Hughes and colleagues(2012) examined this assumption in a version of the bar-transporttask, requiring 180 degrees of object rotation by adding precision requirements to the start and end of the object manipulation. According to the precision hypothesis, they hypothesized that people plan their movements to afford comfortable start postures if precision requirements are high at the start of the movement, and that they should plan their movement to afford comfortable end postures, if precision requirements are high at the end of the action. Unfortunately, they found half of the participants follow their hypothesis, the other half planned for ESC, independent of precision requirements at the start and/or end of the action. Therefore, the influence of precision requirements still remains to be clarified. Nevertheless, our results gained in the experimental study provided partial support for the precision hypothesis. The bar-transport131

General Discussion task and the sword-rotation-task required precision when inserting the bar/sword into the target hole, whereas the grasp-height-task did not require that much precision, as the platforms were big enough to easily place the plunger onto them. This may be another reason why ESC is only little observed in this task. Therefore, it can be reasoned that if precision requirements were higher, then more children would have shown ESC in the grasp-height-task. 4.1.4. Perceived comfort It can be assumed, that children do not perceive an underhand grip or a thumb-down grasp as awkward as adults do and therefore, do not see the necessity of adjusting their grasp strategy in terms of ESC. This may be due to another view of comfort in children, compared to adults. Adults rate a rather supinated position of 90 degrees of their hand as quite awkward. As childrens' joint flexibility is rather higher than in adults, it may be that they do not feel uncomfortable in this end position. Until today, only one study examined perceived comfort in children(Wilmut& Byrne, 2014b). Unfortunately, these authors did not provide the data to analyze the difference in children and adults on the different grasp orientations. In any case, comfort ratings of children at different ages and in different ESC tasks should be examined in future studies. 4.1.5. Different kinds of learning effects The studies reviewed in Chapter 2 as well as the tasks used to asses anticipatory planning skills also differed regarding constraints, which may have provoked any learning effects. These can be originated through differences in the experience with the task(i.e. if children were already familiar with the kind of task through toys or deliberate play), differences in the exposure to the task(i.e. if children were given any familiarization trials, and how many trials were comprised in the testing session), and 132

General Discussion differences in test procedures regarding observational learning(i.e. if children were shown the correct solution of the task by the experimenter prior to testing). One assumption regarding experience with the task is that children may have a higher level of expertise in some tasks than in others. As many toys for infants include actions, such as inserting objects into holes, but not putting objects into shelves, it may be that children are not as familiar with these kinds of actions and thus, did not show the grasp-height-effect until the age of 10 years. This may be due to the existence of two independent systems, which are required to solve goaldirected tasks: a habitual, stimulus-driven and an intentional, goal-directed system, which are employed by the CNS(Herbort& Butz, 2011, see Chapter 2.4.1.7.). The perception of an object automatically evokes actions that are habitually applied to grasp these objects. The presentation of everyday objects may therefore automatically activate those actions that are habitually used to grasp them, independent of the current intention of the participant. It may be that such habitual grasping effects also influenced the planning strategies in some of the children, especially, when the objects used were everyday life objects. This might be the case for the grasp-heighttask, because children are familiar with wooden blocks or sticks of trees, but they use these objects rather for playing than for simply placing them. Therefore, they may have stored a habitual grasp posture somewhere in the middle of the bar, making it easier for them to use it for deliberate play. Also, the exposure to the task differed across the studies reviewed in Chapter 2. The first difference in test procedure was whether or not participants were given the opportunity to practice the task before the actual testing began. Some studies gave children an opportunity to practice prior to testing(e.g. Crajé, Aarts, et al., 2010; Knudsen et al., 2012; Jovanovic& Schwarzer, 2011), whereas others did not(e.g. Adalbjornsson et al., 2008; Jongbloed-Pereboom et al., 2013; Weigelt& Schack, 2010). Results showed, that the presence of ESC does not seem to depend on the 133

General Discussion chance to get familiar with the task, as some of the studies found evidence for ESC planning in children, others did not. These differences in findings were independent of the presence of familiarization trials. The second constraint regarding exposure relies on the amount of trials participants were given during the testing session. Whereas children had to complete only six trials in the bar-transport-task(three critical trials) and in the grasp-height-task, but 18 trials(six critical trials) in the sword-rotation-task, one might think(based on learning theories), that more trials could lead to better performance. Upon today, there is hardly any empirical evidence for this notion. On exception is provided by McCarthy and Keen(2005), who presented infants a spoon, lying in different orientations, for 12 consecutive trials. Results revealed, that 12-month old infants used the appropriate radial grip more often when the spoon was presented in the same orientation all 12 times, than when the orientation of the spoon was alternated. But, even when the spoon was presented in a blocked fashion(i.e. six times with the bowl pointing to the left, six times with the bowl pointing to the right), there were no learning effects observable. These observations are in line with the results provided in Chapter 3(see Figure 20 for the performance in the three trial repetitions in the bar-transport-task, Figure 22 for the six trials repetitions in the sword-rotation-task, and Figure 25 for the six trials in the grasp-height-task). Results showed that experiencing nonefficient grips in earlier trials had no influence on grasp choice in the following trials. However, it is still an open question, whether the developmental onset of ESC would have been found earlier in the present tasks if children were provided with much more trials to practice Besides the findings, that children improved their strategies after repeated trials of spoon use(McCarthy& Keen, 2005), there is also much other evidence that many aspects of motor development involves learning by experience(for a review see Berger& Adolph, 2007). Also, learning theories assume that observational learn134

General Discussion ing and instructions play an important role in cognitive(and motor) development(e.g. Vygotski, 1977). Hence, children might optimize their grasp strategies through observational learning. Only one of the studies reviewed in Chapter 2 did explicitly examine the role of these effects in children(Jovanovic& Schwarzer, 2011). In their study, Jovanovic and Schwarzer(2011) showed that demonstrating the task by the experimenter led to a significant improvement in grasp strategies applied by the children, suggesting that children indeed benefited from observation. One quite plausible interpretation of this result is that the acquisition of efficient motor patterns is the result of learning, which would be in line with other findings(Berger& Adolph, 2007; Boncoddo, Dixon,& Kelly, 2010; Oztop, Bradley,& Arbib, 2004). As the tasks were not demonstrated in the experimental study, no assertion can be made about observational learning effects in general, but a possible influence can be denied for the tasks used in the experimental study(Chapter 3). However, it is still an open question, whether observational learning or verbal instructions can modify children's b ehavior, so that ESC behavior emerges much earlier in the course of motor(and cognitive) development. 4.1.6. Directedness of the tasks In a study on toddlers, Claxton and colleagues(2009) found that self-directed actions le d to an earlier onset of ESC planning, which is in line with Vygotski's theory (1977). The child learns best from interaction with the environment and from the interaction with other social agents. The results of the review study(Chapter 2) suggest, that self-directed actions lead to an earlier onset of development of ESC planning. Therefore, self-directed tasks should be taken into account in future investigations on the development of ESC planning. 4.2. Theory-driven assumptions Regarding the DOF problem mentioned in Chapter 1.1(Bernstein, 1967), it can be assumed from the data of ESC planning in adults that people tend to choose 135

General Discussion their grasp strategy due to the most economic behavior(Rosenbaum et al., 2012). When grasping the cup, which is standing upside down in the cabinet, adults will choose an uncomfortable thumb-down posture when initially grasping the cup in order to end the action comfortably, with their thumb pointing upwards, in a manner consistent with ESC. The DOF are limited to those outcomes, which enable the most economical behavior and therefore results in an almost automatical response to the stimulus, enabling the person to drink from the cup after its rotation. Children, however, have a greater variability in selecting their grasping posture; the control of the DOF is lower than in adults. With regard, Bernstein(1967) postulated three learning stages. In the first stage, the control problem is accounted for by reducing the number of degrees of freedom by simply freezing them(Vereijken, van Emmerik, Whiting, & Newell, 1992). Whether or not a learner freezes the DOF is influenced by the need for control. Children's initial lack of control over the great number of DOF l eads to high variability in performance(Haehl, Vardaxis,& Ulrich, 2000). When DOF are eliminated, the variability of performance is reduced. Therefore, children younger than five years(according to the results of the experimental study) are not able to vary their grasping posture, i.e. to adapt an uncomfortable thumb-down posture at the beginning of the action of turning the cup over. They basically reduce the DOF as they freeze them, meaning that always the same default grip is used. In the second stage, as learning progresses, children can explore additional degrees of freedom and increasingly incorporate them into task solution, thereby adding flexibility to the performance(Vereijken, van Emmerik, Bongaardt, Beek,& Newell, 1997). The exploration of DOF leads to an increase in performance variability, whereas successful exploitation leads to increased flexibility in action outcomes. This means, that children aged five years and older start not to freeze the DOF, but rather explore different outcomes when using a higher amount of DOF. Therefore, they try out how the inclusion of other DOF affects their behavior. They begin to use thumb-down grasps for grasping the cup, taking more DOF into account. Then, in the 136

General Discussion third stage, expertise has progressed to a level where reactive forces are no longer fought or resisted, but actively exploited Considering the results of the ESC tasks in the experimental study, this stage model can be confirmed. Younger children seem to freeze the DOF, resulting in default grasps across the different tasks. As they grew older, children start to explore more and more DOF, enabling them to adapt grip choice prior to grasping. Based on spatial features of DOF, it can be assumed, that the amount of DOF in the graspheight-task(i.e. grasp height) is higher in continuous tasks compared to tasks, where only two different grips are possible, like in the bar-transport task. Concerning grip choice, one can chose only an overhand- or an underhand grip in the bar-transport-task. Moreover, in the continuous grasp-height-task, no joint flexibility is necessary in a way as in other tasks, which need rotation of the object. This may be another concern of not finding ESC in the grasp-height-task in children as old as 10 years of age. As there is no need to explore new DOF as suggested in S tage 2 of Bernstein's model, the possible outcomes were not much different d epending on grasp height, children tend to retain their default grip without any adjustment. Even in adults, the amount of ESC is comparably low, which may be due to the same constraints. Our results, however, are in line with the suggestion of a most recent study(Hermens, Kral,& Rosenbaum, 2014), in which the authors argued that ESC may be restricted to relatively simple grasping movements. According to the ideomotor approach(as outlined in Chapter 1.2.1.), it can be assumed that people experience their motor abilities only indirectly through the perception of achievable sensory effects. This is the case in ESC planning. Referring again to the example of the Introduction: When a person wants to take out a cup from the kitchen shelf in the morning, which is oriented upside-down, he/she will anticipate the sensory consequences when grasping the cup either with a thumb-up or with a thumb-down gasp. As ESC is about control(e.g. Rosenbaum et al., 2012), the per137

General Discussion son will chose the initially uncomfortable thumb-down grasp, in order to achieve the posture, which allows for the most movement control at the end of the action, when pouring coffee into the cup. As children younger than five years did not show evidence for ESC planning in the tasks included in the experimental study, it can be assumed that they are not as sufficient to anticipate, evaluate, and correct sensory consequences(and to correct mistakes). Only by the time they are able to do so, the motor action could be adjusted in terms of ESC. Imagine a young child going to the kitchen shelf to get a cup out of it for a hot chocolate in the morning, he/she will grasp it with a default grip(i.e. with the thumb pointing upwards), resulting in an uncomfortable posture at the end of the movement. Under the age of(minimal) 5 years, this consequence does not get evaluated with regard to control, and therefore, no grip adaption(error correction) is implemented. This may be because people first have to learn about the sensory effects, which can be produced through motor actions(Elsner& Hommel, 2001). Not until then, sensorimotor units are built, which are necessary for the anticipation of movements(or showing the ESC effect). These functional units, which are built with the integration of motor programs and action effects, represent actions and provide a person cognitive access to voluntary actions: the motor program can be activated through the anticipation of the desired consequences(Elsner& Hommel, 2001). Relying on the ABC theory(Chapter 1.3.2.), which takes the primacy of actioneffect learning, as well as the conditionalization of action-effect relations into account (Hoffmann 1993, 2003; Hoffmann et al., 2004, 2007), it can be assumed, that behavioral competence(in the introduction example: grasp the cup, pour coffee into it, and to drink coffee from the cup) emerges by the acquisition of action-effects. Thereby, one needs to distinguish between body-intern and body-extern action effects. The latter are any effects, which appear as a result of an action somewhere in the environment. When thinking about making coffee in the morning, a person goes into the 138

General Discussion kitchen and presses the button on the coffee machine. The subsequent effect of that action(i.e. pressing the button), is that the machine starts brewing coffee. It might be, that these kind of body-external action-effects have an influence on the presence on ESC in children. This assumption was examined in a study by Jovanovic and Schwarzer(2011), who found children as young as three years to show the ESC effect, when the task provided a body-extern action effect(i.e. a light turned on when the child inserted the bar into the target hole). Knudsen and colleagues(2012) investigated their role more profoundly. They used a similar version of the bar-transporttask as Jovanovic and Schwarzer(2011), as well as the overturned-glass-task. Action-effects were added to both tasks, meaning that the correct execution of the action resulted in a shed-on of lights. However, in this study body-external actioneffects did not seem to influe nce children's behavior in both tasks. A body-internal action effect is the subsequently perceived posture at the end of a movement. In the cup example, with the cup standing upside down in the cabinet, there are two possible ways to grasp it: with a thumb-up or with a thumb-down grasp. Both versions result in different outcomes after the rotation of the cup. Whereas grasping with a thumb-down grasp will result in a comfortable end state, an initial thumb-up grasp will result in a very uncomfortable end position, with the arm pronated or supinated by 180 degrees. This perceived outcome can be denoted as a bodyinternal action effect, which can either be positive(comfortable) or negative(uncomfortable). Next, the ABC theory by Hoffmann(1993, 2003) and the conceptualization about the DOF by Bernstein(1967) will be blended to provide a new look at anticipatory behavioral control in children. 4.3. A new look at anticipatory behavioral control in children Even young children can anticipate action effects(for review see Daum& Aschersleben, 2014; Elsner, 2007). However, based on the results of Chapter 2, 139

General Discussion young children do not show the ESC effect. This is surprising, because the literature suggests that even infants are able to anticipate action effects, but most of these anticipated effects were body-external effects. For example, a seminal study conducted by Rovee and Rovee(1969) showed 2- to 5-month old infants to learn the relation between leg kicks and the following contingent movements of a mobile above their head. Such body-external effects were also examined in studies using ESC tasks by Jovanovic and Schwarzer(2011), as well as by Knudsen et al.(2012), in which motor actions were aided by visual effects(i.e. switching on lights). Jovanovic and Schwarzer(2011) found the ESC effect to be present in 3-year old children in a version of the bar-transport-task, after children produced light effects at the end of the actions. However, as Knudsen and colleagues(2012) demonstrated, such visual effects may not be critical for the acquisition of anticipatory planning abilities in the bartransport-task and the overturned-glass-task. Hence, the ability to anticipate body-external effects may not be(fully) decisive for the ESC effect to be present in children. Rather, it is most plausible to assume, that instead of body-external action effects, the ability to anticipate body-internal action effects(i.e. ending an action in a comfortable versus in an uncomfortable end position) may be critical for the development of ESC planning. This assumption will be further grounded within a threestage model based on Bernstein's conceptualization of the DOF problem(1967, see Chapter 4.3. ) and Hoffmann's ABC theory(2009, see Chapter 1.3.2. and 4.3.). The model depicted in Figure 30 shows a three-phase learning model of anticipatory behavioral control, using the coffee-cup example from the beginning of this thesis(Chapter 1). Phase 1 of the model refers to children younger than(approximately) 4-5 years, the age, at which children habitually use a default thumb-up grasp. Phase 2 represents the behavior of children roughly between 5-10 years, when the 140

General Discussion majority starts to show the ESC effect for the first time. Phase 3 resembles the manifestation of the ESC effect in children(typically) older than 10 years, and in adults. In the example of the child going into the kitchen to have a cup of milk, the cup standing upside down in the cabinet serves as the situational condition. The grasping action to be performed, namely to take the cup out of the cabinet in order to have a cup of milk, is the voluntary action. The(body-internal) effect anticipation is the intended end-posture and the real effect is the reafferent perception of the actual endposture, which is attained when the manual rotation action is completed. This defines the major components of the three-phase model. In Phase 1 of the model, younger children will always(automatically) select a default grasp to reach for the cup, as has been suggested by Weigelt and Schack (2010). Thus, they may not be able to anticipate an action outcome other than ending the movement in a thumb-up posture(as a default posture). This may be due to the "freezing" of additional DOF for bette r task control, as was already explained in Chapter 4.3.(Bernstein, 1967). As a consequence, they will finish the manipulation in an uncomfortable thumb-down posture. Because task experience does not seem to influence grasp selection in children of this age(as was observed in the experimental study, Chapter 3), the comparison of the real effects(i.e. uncomfortable thumb-down posture) with the effect anticipation(default thumb-up posture) in order to form action effect associations may be incomplete, delayed, or may not take place at all. Therefore, neither the primary formation of action-effect associations nor the secondary contextualization of action-effect associations is realized. 141

General Discussion Figure 30. Three-phase adaptation of the ABC theory by Hoffmann(2009), adjusted to account for children's performance in ESC tasks. In Phase 1, children automatically select a default grasp(thumb-up grasp), as they are not able to anticipate other effects. As the real effect does not match the effect anticipation, no action-effect associations are formed and no contextualization to the situational condition takes place. In Phase 2, children are able to anticipate different action outcomes. Now, the real effect matches the effect anticipation and action-effect anticipations are formed(i.e. initial thumb-down grasp results in final thumb-up posture after rotation), but these are not yet contextualized to the situational condition. In Phase 3, children(and adults) are able to precisely anticipate desired action effects, based on strong action-effect associations. As the real effect reliably matches the effect anticipation, the action-effect association is now contextualized to the situational condition(i.e. inverted cup). 142

General Discussion In Phase 2, children begin to "free" additional DOF as their motor actions become more variable, resulting in new task solutions. This is accompanied by eminent processes of motor reorganization(e.g. Bard et al., 1990; Thibaut& Toussaint, 2010; see Chapter 4.1.1.). Thus, from various experiences with different task solution, they are now able to anticipate different action outcomes, such as a thumb-up posture(as the optimal posture to finish the action). Now, the real effect matches the effect anticipation and action-effect associations are formed(i.e. initial thumb-down grasp results in final thumb-up posture after rotation). The situational condition(i.e. inverted cup), however, is not yet contextualized to this new action-effect association. In Phase 3, an optimal space of DOF is exploited(Bernstein, 1967) and actions are flexibly selected to achieve intended goal-states. Older children and adults are able to precisely anticipate desired(body-internal) action effects(i.e. to end comfortably with a thumb-up grasp) based on strong action-effect associations. Again, the real effect matches the effect anticipation. As the flexible selection of grasping actions demonstrates, these action-effect associations are contextualized to the situational condition, enabling the actors to choose optimal body postures(Rosenbaum et al., 1996). 4.4. Criticism on the experimental study First, there is one point to criticize about the review study provided in Chapter 2. In this study, a search procedure was applied, based on previous systematic reviews like the one of Blauw-Hospers& Hadders-Algra(2005) or Greaves et al. (2010), containing eight distinct steps for literature search. Quite recently, another, standardized system for systematic literature search was discovered by the author of this thesis. This schema is named"PRISMA ­ Transparent reporting of systematic reviews and metaanalyses"(Moher, Liberat i, Tetzlaff,& Altman, 2009). The PRISMA Statement consists of a 27 item checklist. If this schema would have been applied to 143

General Discussion the systematic review in Chapter 2, this might have made the review more comparable to other reviews and improved in its structure. Second, the tasks to measure ESC planning abilities might have been too different to exhibit relationships between their developmental patterns. They differed regarding many details, like the degree of object rotation vs. object translation, their motivational character, or the precision requirements(see Chapter 2.4.2.). Maybe, these are too many factors that prevented the finding of any correlations between them. Moreover, all tasks were external-directed, what has been shown elsewhere to gather rather small effects than when tasks are self-directed(like the spoon-handlingtask, for example, which showed even toddlers to plan their actions in advance; Claxton et al., 2009; McCarthy et al., 1999). Therefore, the use of more self-directed tasks, and of tasks, that were more similar regarding their task constraints, may reveal more similar developmental trajectories than the tasks used in the experimental study(Chapter 3). The third concern reading the EF tasks is, that task selection may have been appropriate, but not for the big range of considered ages. As different test protocols were needed for different age groups, results of the tests were difficult to compare. The developmental trajectories in the different tasks must be seen critical. For example: A 5-year old child solved 40% of the TOH task, a 6-year-old solved 60%. One would expect a 7-year-old to solve more than 60%, maybe 70% of the items, but as the test protocol changed and the items became more difficult, a 7-year old child might perform even worse. This may have sophisticated the results on EF in children. One solution would be to examine children of a smaller age-range, while using tests with the same protocol within the age range(i.e. 3- to 5-year-olds, as the test protocols stay the same in all three EF tasks for that age range). Another possibility is to find tasks, that are able to measure EF over a larger age range, but with the same protocol. However, the author is not aware of any test to fulfil this criterion. Either, the 144

General Discussion cognitive development requires an increase in difficulty, or there are ceiling effects at a certain age(e.g. in the famous Stanford marshmallow task to measure inhibitory control, where the test procedure stays the same, but children as old as 7-years do show ceiling effects; Mischel, Ebbesen,& Raskoff Zeiss, 1972). Fourth, it might be that the selection of EF tasks may not have been appropriate to examine a relationship to ESC planning. Maybe, the EF tasks chosen in the experimental study(i.e. problem solving, visual perception, and attention) are no underlying mechanisms of ESC planning abilities. Therefore, other tests to assess EF may be used. However, it needs to be assumed, that a study on correlations between motor and intellectual functions in children between 7 and 18 years, which uses different measures of intelligence as well as several motor tasks, is not able to find a relationship between motor and cognitive functions. Such cognitive measures should be applied in combination with ESC tasks to examine a possible relationship of more superior cognitive skills and the more subordinate motor ability of ESC planning. One possible measure could be inhibitory control. Results of a study conducted by Eigsti et al.(2006) revealed, that the proportion of time preschoolers directed their attention away from a rewarding stimulus was positively associated with efficiency(here: greater speed without reduced accuracy at responding targets in a go-/no-go task). As the ESC effect can also be seen as an efficiency constraint of motor planning, the influence of inhibitory control on the ESC effect may be another possible mediating factor on ESC planning. Another possible underlying cognitive function could be memory performance. In a study conducted by Weigelt et al.(2009), a close relationship was revealed for perceptual-motor control(ESC) and cognitive processing(i.e. to memorize letters). Logan and Fischman(2011) conducted an experiment, in which participants had to view a series of letters and then perform an ESC task. Results revealed that independent of task complexity, the commonly found recency effect in memory recall was absent. This suggests a reciprocal influence of physical action and cognitive processes, which were interpreted as a basic concurrence cost. In an145

General Discussion other recent study, Spiegel and colleagues(2014) investigated the influence of available attentional resources on task-costs. Results revealed complex interactions, in that the processes involved in movement planning, spatial attention, and visuospatial working memory are functionally associated, but not linked in a mandatory fashion. Therefore, these cognitive skills could provide more insight into the functional linkage of cognition and action planning. It may be, that this and other measures of cognitive functions are more related to the development of ESC than the EF tasks chosen in our study. 4.5. Directions of future studies Even if a number of important questions regarding the development of cognitive and motor planning skills could have been answered here, some open questions remain, which should be the focus of future research. A selection of open topics based on what was discussed above will be listed in the following. In the review study, several task constraints were identified, which may have an influence on the presence of ESC in the different tasks. However, still not much is known about the role of these constraints in children, including the influence of the number of action steps, the required degree of object rotation, precision requirements, and so on(see Chapter 2.4.2). In future studies, the influence of each of these constraints should be further examined in children(and adults). Moreover, it is possible that children's perception of comfort is different from the perception of comfort in adults. There have been several studies investigating perceived comfort ratings for ESC tasks(e.g., Johnson, 2000; Rosenbaum et al., 1993; Seegelke, Hughes,& Schack, 2011). In these studies, participants perceived grasps at both ends of their supination or pronation range as considerably more awkward than those in the middle of their comfortable ranges of motion. By contrast, only one study has investigated perceived comfort levels in children and compared 146

General Discussion these to adults(Wilmut& Byrne, 2014b). Here, only the ratings of 10- to 12-year old children were similar to those of adults, suggesting that younger children perceive postures at both ends of their supination or pronation range as less awkward. It must be noted, however, that children may not conceptualize a given rating scale(and thus may not provide their comfort ratings) in the same way as adults do. In any case, the findings from these studies, as well as those reported here, suggest that future research should give greater consideration to the comfort ratings provided by children. As mentioned above, the influence of internally- versus externally-oriented tasks in children should be examined. In toddlers, it has been demonstrated that selfdirected motor planning tasks may elicit more effective motor planning than externally-oriented tasks, as the consequences of the actions are more obvious to the participants(Claxton et al., 2009). As nothing is known about the developmental trajectories of internally- versus externally-oriented tasks in children, it should be the aim of future research to examine this relationship. Furthermore, nothing is known about the influence of habitual vs. goal-directed stimuli(Herbort& Butz, 2011). A recent study by Herbort and Butz(2011) demonstrated that the ESC effect may be compromised for such habitual grasping behavior, when participants were presented with objects of everyday live. It may be that such habitual grasping effects also influenced the planning strategies in some of the children, especially, when the objects used were everyday life objects or objects, children are familiar with. This assumption should be examined in future studies with children and adults. Another cognitive function possibly linked to ESC is mental rotation. To anticipate the future state of an object, and the grasp posture which is attained with a dedicated starting position, the to-be-manipulated object needs to be mentally rotated into the desired end position in order to anticipate grasp outcomes. Marmor(1975) 147

General Discussion was the first to study mental rotation in children and found out, that 5-year old children were able to solve mental rotation tasks. Comparing this outcome with the results of the bar-transport-task and the sword-rotation-task of the experimental study, it is obvious, that this is exactly the age where the majority of children is sensitive for ESC. A study conducted by Frick, Daum, Walser, and Mast(2005) investigated the relationship of mental and manual rotation. They found, that mainly in children younger than 11 years mental and motor processes are coupled more closely than in older children and adults. Therefore, the ability to mentally rotate objects may be a premise for the anticipation of end postures in object manipulation tasks. Despite the fact, that no relationship between ESC and EF was found in the experimental study in Chapter 3, it might be that other EF tasks could be more appropriate for future investigations, such as the Wisconsin Card Sort task(Grant& Berg, 1948) or a Go-/ no Go task(Nosek& Banaji, 2001). Alternative to EF tasks, more general tests of cognitive functions could be applied, especially those, that test for intelligence in children(e.g. the Wechsler Intelligence Scale for Children, Wechsler, 2003). Also, the influence of memory performance on the ESC effect should be within the focus of future research, as can be derived from other studies (Weigelt et al., 2009; Logan& Fischman, 2011). The main restriction will be to find appropriate measures for the majority of age groups tested in the experimental study. The search for new cognitive tasks, however, may not be easy, as a most recent study on correlations between motor and intellectual functions in normally developing children between 7 and 18 years showed that motor and intellectual domains in normally developing children are largely independent and should be, for the most part, considered separately(Jenni et al., 2013). Hence, the question remains, whether cognitive and motor development is in other ways related, and if so, which measures(of cognitive and executive functions) could be used to examine this relationship appropriately. Therefore, future studies should take other cognitive functions 148

General Discussion into account, which might be more correlated with the performance in ESC tasks than the tasks in the experimental study used here. Given the positive developmental trajectory for ESC planning in children, it still remains an open question how the ability for anticipatory planning develops at old ages. If cognitive(or executive) factors are really an determining mechanism for the development of ESC, does the effect decline in higher age, as cognitive functions do? Many studies have demonstrated that cognitive decline takes place with increasing age, even with different intraindividual extends(Salthouse, 1991). The course of the developmental curve for ESC planning over the lifespan still needs to be clarified. A first step into this direction was already taken by the author(Wunsch, Weigelt,& Stöckel, in prep): Older adults between 60 and 80 years were tested in the bartransport-task. Results showed that there is a decline in ESC sensitivity in higher ages, which is similar in unimanual and bimanual grasping tasks. Older participants tended not to select ESC as consistent as their younger counterparts. Based on these results, it was suggested, that the ability for motor planning may rely on cognitive planning skills, which are known to decline in higher ages. Hence, the ability for anticipatory planning at the end of the lifespan should be examined in future studies, using different tasks and cognitive measures. 4.6. Conclusion The review study(Chapter 2) provided insights into the status quo of research on the developmental pattern of the ESC effect as a signification of anticipatory planning abilities in children. As the results showed, the ESC effect develops from early childhood until adolescence, with different trajectories for different ESC tasks. The experimental study(Chapter 3) supported the pattern of results from previous studies across different ESC tasks, as has been revealed in the review study(Chapter 2). In addition, no relationship between performance in the EF tasks and the developmental trajectory of the ESC effect was observed. Thus, contrary to what one would as149

General Discussion sume from the most prominent cognitive developmental theories(e.g. Piaget, 1936; Vygostki, 1977), there may not be such a tight link between cognitive and motor development in children. This lack of evidence, however, should be addressed again in future studies, as well as the influence of other factors on the development of the anticipatory planning abilities and motor skills during childhood. 150

References and Supplements References and Supplements 151

References References Adalbjornsson, C. F., Fischman, M. G.,& Rudisill, M. E.(2008). The end-state comfort effect in young children. Research Quarterly for Exercise and Sports, 79(1), 36-41. Adolph, K. E.,& Berger, S. A.(2006). Motor development. In W. Damon, R. Lerner, D. Kuhn, & R.S. Siegler(Eds.), Handbook of child psychology: Cognition, perception, and language(pp. 161-213). New York: Wiley. Adolph, K. E., Eppler, M. A.,& Gibson, E. J.(1993). Development of perception of affordances. In C. Rovee-Collier& L. P. Lipsitt(Eds.), Advances in infancy research(pp. 5198). Norwood: Ablex Publishing Co. Anderson, P.J.(2002). Assessment and development of executive function(EF) during childhood. Child Neuropsychology, 8(2), 71-82. Anderson, P. J.(2008). Towards a developmental model of executive function. In V. Anderson, R. Jacobs,& P.J. Anderson(Eds.), Executive functions and the frontal lobes: a lifespan perspective(pp. 3-21). New York: Taylor& Francis. Anderson, P. J., Anderson, V.,& Garth, J.(2001). Assessment and development of organizational ability: The Rey Complex Figure Organizational Strategy Score(RCF-OSS). The Clinical Neuropsychologist, 15, 81-94. Anderson, P. J., Anderson, V.,& Lajoie, G.(1996). The Tower of London Test: Validation and standardization for pediatric populations. The Clinical Neuropsychologist, 10, 5465. Anderson, P. J., Anderson, V., Northam, E., Jacobs, R.,& Catroppa, C.(2001). Development of executive functions through late childhood and adolescence in an Australian sample. Developmental Neuropsychology, 20(1), 385-406. Anderson, S., Bechara, A., Damasio, H., Tranel, D.,& Damasio, A.(1999). Impairment of social and moral behavior related to early damage in human prefrontal cortex. Nature Neuroscience, 2, 1032-1037. Andreasen, N.(2002). Brave New Brain. Geist, Gehirn, Genom. Berlin: Springer. Baddeley, A.,& Della Salla, S.(2003). Working memory and executive control. In A.C. Roberts, T.W. Robbins,& L. Weiskrantz(Eds.), The prefrontal cortex. Executive and cognitive functions(pp. 9-21). Oxford: Oxford University Press. 152

References Bard, C., Hay, L.,& Fleury, M.(1990). Timing and accuracy of visually directed movements in Children: Control of direction and amplitude components. Journal of Experimental Child Psychology, 50, 102-118. Barrash, J., Tranel, D.,& Anderson, S.(2000). Acquired personality disturbances associated with bilateral damage to the ventromedial prefrontal region. Developmental Neuropsychology, 18, 355-381. Barrett, T. M., Traupman, E.,& Needham, A.(2008). Infants' visual anticipation of object structure in grasp planning. Infant Behavior and Development, 31(1), 1-9. Becker, M., Isaac, W.,& Hynd, G.(1987). Neuropsychological development of nonverbal behaviors attributed to``Frontal Lobe'' functioning. Developmental Neuropsychology, 3, 275-298. Bell, M.,& Fox, N.(1992). The relations between frontal brain electrical activity and cognitive development during infancy. Child Development, 63, 1142-1163. Berger, S. E.,& Adolph, K. E.(2007). Learning and development in infant locomotion. Progress in Brain Research, 164, 237-255. Bernstein, N.(1967). The coordination and regulation of movements. Oxford: Pergamon Press. Blakemore, S.-J.,& Choudhury, S.(2006). Development of the adolescent brain: implications for executive function and social cognition. Journal of Child Psychology and Psychiatry, 47(4), 296-312. Blauw-Hospers, C. H.,& Hadders-Algra, M.(2005). A systematic review of the effects of early intervention on motor development. Developmental Medicine and Child Neurology, 47(6), 421-432. Boncoddo, R., Dixon, J. A.,& Kelley, E.(2010). The emergence of a novel representation from action: Evidence from preschoolers. Developmental Science, 13, 370-377. Borys, S. V., Spitz, H. H.,& Dorans, B. A.(1982). Tower of Hanoi performance of retarded young adults and nonretarded children as a function of solution length and goal state. Journal of Experimental Child Psychology, 33, 87-110. Bower, T. G. R.(1977). A primer of infant development. San Francisco: Freeman. Brickenkamp, R.(1962). Test d2 Aufmerksamkeits-Belastungs-Test. Göttingen: Hogrefe. Brickenkamp, R., Schmidt-Atzert, L.,& Liepmann, D.(2010). Test d2 ­ Revision. Aufmerksamkeits- und Konzentrationstest. Göttingen: Hogrefe. 153

References Butz, M. V., Sigaud, O.,& Gérard, P.(2003). Anticipatory behavior: Exploiting knowledge about the future to improve current behavior. In M. V. Butz, O. Sigaud,& P. Gérard (Eds.), Anticipatory behavior in adaptive learning systems( pp. 1-10). Heidelberg: Springer. Carlson, S. M.(2003). Executive function in context: Development, measurement, theory, and experience. Monographs of the Society for Research in Child Development, 68, 138-151. Casey, B. J., Jones, R. M.,& Hare, T. A.(2008). The adolescent brain. In A. Kingstone& M. B. Miller(Eds.), Annals of the New York Academy of Sciences: Vol. 1124. The Year in Cognitive Neuroscience(pp. 111-126). New York: New York Academy of Sciences. Chapman, K. M., Weiss, D. J.,& Rosenbaum, D. A.(2010). Evolutionary roots of motor planning: The end-state comfort effect in lemurs. Journal of Comparative Psychology, 124(2), 229-232. Chelune, G.,& Baer, R.(1986). Developmental norms for the Wisconsin Card Sorting Test. Journal of Clinical and Experimental Neuropsychology, 8, 219-228. Claxton, L. J., Keen, R.,& McCarty, M. E.(2003). Evidence of motor planning in infant reaching behavior. Psychological Science, 14(4), 354-356. Claxton, L.J., McCarthy, M. E.,& Keen, R.(2009). Self-directed action affects planning in tool-use tasks with toddlers. Infant Behavior and Development, 32(2), 230-233. Cohen, R. G.,& Rosenbaum, D. A.(2004). Where objects are grasped reveals how grasps are planned: Generation and recall of motor plans. Experimental Brain Research, 157, 486-495. Cole, P. M., Armstrong, L. M.,& Pemberton, C. K.(2010). The role of language in the development of emotion regulation. In S. D. Calkins& M. A. Bell(Eds.), Child development at the intersection of emotion and cognition(pp. 59-77). Washington, DC: APA. Connolly, K.,& Dalgleish, M.(1989). The emergence of a tool-using skill in infancy. Developmental Psychology, 25(6), 894-912. Cowie, D., Smith, L.,& Braddick, O.(2010). The development of locomotor planning for endstate comfort. Perception, 39(5), 661-670. Crajé, C., Aarts, P., Nijhuis-van der Sanden, M.,& Steenbergen, B.(2010). Action planning in typically and atypically developing children(unilateral cerebral palsy). Research in Developmental Disabilities, 31(5), 1039-1046. 154

References Crajé, C., van Elk, M., Beeren, M., van Schie, H. T., Bekkering, H.,& Steenbergen, B. (2010). Compromised motor planning and motor imagery in right hemiparetic cerebral palsy. Research in Developmental Disabilities, 31(6), 1313-1322. Daum, M.& Aschersleben, G.(2014). Experimentelle Handlungsforschung: Die ontogenetische Perspektive. In W. Prinz(Hrsg.), Experimentelle Handlungsforschung(pp.158216). Stuttgart: Kohlhammer. Davidson, M. S., Amso, D., Anderson, L. C.,& Diamond, A.(2006). Development of cognitive control and executive functions from 4 to 13 years: Evidence from manipulations of memory, inhibition, and task switching. Neuropsychologia, 44(11), 2037-2078. De Luca, C., Wood, S. J., Anderson, V., Buchanan, J. A., Proffitt, T., Mahony, K.,& Pantelis, C.(2003). Normative data from CABTAB: Development of executive function over the lifespan. Journal of Clinical and Experimental Neuropsychology, 25(2), 242-254. Dempster, F. N.(1992). The rise and fall of the inhibitory mechanism. Toward a unified theory of cognitive development and aging. Developmental Review, 12, 45-75. Dennis, M.(1989). Language and young damaged brain. In T. Boll& B. Bryant(Eds.), Clinical neuropsychology and brain function: Research, measurement and practice(pp. 89-123). Washington: American Psychological Association. Diamond, A.(1985). Development of the ability to use recall to guide action, as indicated by infants' performance on AB. Child Development, 56, 868-883. Diamond, A.,& Doar, B.(1989). The performance of human infants on a measure of frontal cortex function, the Delayed Response task. Developmental Psychobiology, 22, 271294. doi: Diamond, A.,& Taylor, C.(1996). Development of an aspect of executive control: Development of the abilities to remember what I said and to``do as I say, not as I do''. Developmental Psychobiology, 29, 315-334. Edelman, G. M.(1987). Neural Darwinismus ­ The Theory of Neural Group Selection. New York: Basic Books. Elsner, B.(2007). Infant's imitation of goal -directed actions: The role of movements and action effects. Acta Psychologica, 124, 44-59. Elsner, B.,& Aschersleben, G.(2003). Do I get what you get? Learning about effects of selfperformed and observed actions in infants. Consciousness& Cognition, 12(4), 732751. 155

References Elsner, B.,& Hommel, B.(2001). Effect anticipation and action control. Journal of Experimental Psychology: Human Perception and Performance, 27, 229-240. Eslinger, P.,& Damasio, A.(1985). Severe disturbance of higher cognition after bilateral frontal lobe ablation. Neurology, 35, 1731-1741. Fischer, K. W.(1980). A theory of cognitive development: The control and construction of hierarchies of skills. Psychological Review, 87(6), 477-531. Fischman, M. G.(1997). End-state Comfort in object manipulation. Research Quarterly for Exercise and Sports, Suppl., A-60. Fischman, M. G., Stodden, D. F.,& Lehman, D. M.(2003). The end-state comfort effect in bimanual grip selection. Research Quarterly for Exercise and Sports, 74(1), 17-24. Fleishman, E. A., Quaintance, M. K.,& Broedling, L. A.(1984). Taxonomies of human performance: The description of human tasks. Orlando: Academic Press. Frey, S. H.,& Povinelli, D. J.(2012). Comparative investigations of manual action representations: evidence that chimpanzees represent the cost of potential future actions involving tools. Philosophical Transactions of the Royal Society of Biological Sciences, 367(1585), 48-58. Frick, A., Daum, M. M., Walser, S.,& Mast, F. W.(2005). Developmental changes in the interference of motor processes with mental rotation. In B. G. Bara, L. Barsalou,& M. Bucciarelli(Eds.), Proceedings of the XXVII annual conference of the cognitive science society(CogSci2005)(pp. 720 ­ 725). Stresa Italy: University of Turin. Gesell, A.(1933). Maturation and the patterning of behavior. In C. Murchison(Ed.), A handbook of child psychology(pp. 209-235). New York: Russell& Russell Publishers. Gesell, A.,& Thompson, H.(1934). Infant behavior: Its genesis and growth. New York: McGraw-Hill Book Company. Geurts, H. M., Verté, S., Oosterlan, J., Roeyers, H.,& Sergeant, J. A.(2004). How specific are executive functioning deficits in attention deficit hyperactivity disorder and autism? Journal of Child Psychology and Psychiatry, 45(4), 836-854. Gibson, J. J.(1966). The senses considered as perceptual systems. Boston: Houghton Mifflin. Gioia, G. A., Isquith, P. K., Guy, S. C.(2001). Assessment of executive function in children with neurological impairments. In: R. J. Simeonsson& S. L. Rosenthal(Eds.), Psy156

References chological and developmental assessment: Children with disabilities and chronic conditions(pp. 317-256). New York: The Guilford Press. Gioia, G., Isquith, P., Guy, S.,& Kenworthy, L.(2000). BRIEF ­ Behavior Rating Inventory of Executive Function. Professional manua l. Odessa, FL: Psychological Assessment Resources. Golden, C.(1981). The LuriaNebraska Children's Battery: Theory and formulation. In G. Hynd& J. Obrzut(Eds.), Neuropsychological assessment of the school-aged child (pp. 277-302). New York: Grune& Stratton. Goldenberg, G.,& Hagmann, S.(1998). Tool use and mechanical problem solving in apraxia. Neuropsychologia, 36(7), 581-589. Gollwitzer, P. M.,& Brandstätter, V.(1997). Implementation intentions and effective goal pursuit. Journal of Personality and Social Psychology, 73(1), 186-199. Gollwitzer, P. M., Fujita, K.,& Oettingen, G.(2004). Planning and the implementation of goals. In R. Baumeister& K. Vohs(Eds.), Handbook of self-regulation: Research, theory and applications(pp. 211-228). New York: Guilford Press. Gonzalez, C. L. R., Mills, K. J., Genee, I., Li, F., Piquette, N., Rosen, N.,& Gibb, R.(2014). Getting the right grasp on executive function. Frontiers in Psychology, 5, 1-11 Gordon, A. M., Charles, J.,& Steenbergen, B.(2006). Fingertip force planning during grasp is disrupted by impaired sensorimotor integration ion children with hemiplegic cerebral palsy. Pediatric Research, 60, 587-591. Grant, D.A.,& Berg, E.A.(1948). A behavioral analysis of reinforcement and ease of shifting to new responses in a Weigl-type card sorting problem. Journal of Experimental Psychology, 38, 404-411. Grattan, L.,& Eslinger, P.(1991). Frontal lobe damage in children and adults: A comparative review. Developmental Neuropsychology, 7, 283-326. Greaves, S., Imms, C., Dodd, K.,& Krumlinde-Sundholm, L.(2010). Assessing bimanual performance in young children with hemiplegic cerebral palsy: A systematic review. Developmental Medicine and Child Neurology, 52(5), 413-421. Grob, A., Meyer, C. S.,& Hagmann-von Arx, P.(2009). IDS: Intelligence and Development Scales. Intelligenz- und Entwicklungsskalen für Kinder von 5-10 Jahren. Bern: Verlag Hans Huber. 157

References Grob, A., Reimann, G., Gut, J.,& Frischknecht, M.-C.(2013). Intelligence and Development Scales ­ Preschool(IDS-P). Intelligenz- und Entwicklungsskalen für das Vorschulalter. Bern: Verlag Hans Huber. Grusec, J. E.(2011). Socialization processes in the family: Social and emotional development. Annual Review of Psychology, 62, 243-269. Haehl, V., Vardaxis, V.,& Ulrich, B. D.(2000). Learning to cruise: Bernstein's theory applied to skill acquisition during infancy. Human Movement Science, 19, 685-715. Henry, F. M.,& Rogers, D. E.(1960). Increased response latency for complicated movements and a`memory drum' theory of neuromotor reaction. Research Quarterly, 31, 448-458. Herbort, O.(2012). Where to grasp a tool? Task-dependent adjustments of tool transformations by tool users. Zeitschrift für Psychologie, 220(1), 37-43. Herbort, O.,& Butz, M. V.(2011). Habitual and goal-directed factors in(everyday) object handling. Journal of Experimental Brain Research, 213, 371-382. Heriza, C.(1991). Motor development: Traditional and contemporary theories. In M. J. Lister (Ed.), Contemporary management of motor control problems(pp. 99-126). Alexandria: Foundation for Physical Therapy. Hermens, F., Kral, D.,& Rosenbaum, D. A.(2014). Limits of end-state planning. Acta Psychologica, 148, 148-162. Hill, E. L.(2004). Executive dysfunction in autism. Trends in Cognitive Psychology, 8(1), 2623. Hoffmann, J.(1993). Vorhersage und Erkenntnis: Die Funktion von Antizipationen in der menschlichen Verhaltenssteuerung und Wahrnehmung(Anticipation and Cognition: The function of anticipations in human behavioral control and perception). Oxford: Hogrefe. Hoffmann, J.(2003). Anticipatory Behavioral Control. In M. V. Butz, O. Sigaud,& P. Gérard (Eds). Anticipatory behavior in adaptive learning systems(pp 44-65). Heidelberg: Springer. Hoffmann, J., Stöcker, C.,& Kunde, W.(2004). Anticipatory control of actions. International Journal of Sport and Exercise Psychology, 2, 346-361. 158

References Hoffmann, J., Berner, M., Butz, M. V., Herbort, O., Kiesel, A., Kunde, W.,& Lenhard, A. (2007). Explorations of anticipatory behavioral control(ABC): A report from the cognitive psychology unit of the University of Würzburg. Cognitive Process, 8, 133-142. Hommel, B., Muesseler, J., Aschersleben, G.,& Prinz, W.(2001). The theory of event coding (TEC): A framework for perception and action planning. Behavioral and Brain Sciences, 24, 849-878. Houk, J. C.,& Rymer, J. C.(1981). Neural control of muscle length and tension. In V.B. Brooks(Ed.), Handbook of Physiology: Motor Control(pp. 257-323). Bethesda.: American Physiological Society. Hughes, C.(1996). Brief report: Planning problems in autism at the level of motor control. Journal of Autism and Developmental Disorders, 26(1), 99-107 Hughes, C. M.,& Franz, E. A.(2008). Goal-related planning constraints in bimanual grasping and placing of objects. Experimental Brain Research, 188, 541-550. Hughes, C. M. L., Reißig, P.,& Seegelke, C.(2011). Motor planning and execution in leftand right-handed individuals during a bimanual grasping and placing task. Acta Psychologica, 138, 111-118. Hughes, C. M. L., Seegelke, C.,& Schack, T.(2012).The influence of initial and final precision on motor planning: Individual differences in end-state comfort during unimanual grasping and placing. Journal of Motor Behavior, 44(3), 195-201. Jacques, S.,& Zelazo, P.(2001). The Flexible Item Selection Task(FIST): A measure of executive function in preschoolers. Developmental Neuropsychology. 20(3), 573-591. James, W.(1890). The principles of psychology. New York: Henry Holt and Company. Jansen, P.(2014). Macht Bewegung unsere Kinder wirklich schlauer? Neue Erkenntnisse zum Zusammenhang von Bewegung und kognitiven Fähigkeiten bei Kindern und Jugendlichen. Sports, Orthopaedics and Traumatology, 30, 267-273. Janssen, L., Crajé, C., Weigelt, M.,& Steenbergen, B.(2010). Motor planning in bimanual object manipulation: Two plans for two hands? Motor Control, 14, 240-254. doi: Jeannerod, M.(1981). Intersegmental coordination during reaching at natural objects. In J. Long& A. Baddeley(Eds.), Attention and Performance IX(pp. 153-169). Hillsdale: Erlbaum. James, W.(1890). The principles of psychology. New York: Holt. 159

References Jenni, O. G., Chaouch, A., Calfish, J.,& Rousson, V.(2013). Correlations between motor and intellectual functions in normally developing children between 7 and 18 years. Developmental Neuropsychology, 38(2), 98-113. Johns Hopkins Medical Institutions(2008, February 27). Autism's origins: Mother's antibody production may affect fetal brain. ScienceDaily, Retrieved August 6, 2014 from www.sciencedaily.com/releases/2008/02/080225213444.htm Johnson, S. H.(2000). Thinking ahead: The case for motor imagery in prospective judgements of prehension. Cognition, 74, 33-70. Jongbloed-Pereboom, M., Nijhuis-van der Sanden, M. W. G., Saraber-Schiphorst, N., Crajé, C.,& Steenbergen, B.(2013). Anticipatory action planning increases from 3 to 10 years of age in typically developing children. Journal of Experimental Child Psychology, 114, 295-305. Jovanovic, B.,& Schwarzer, G.(2011). Learning to grasp efficiently: The development of motor planning and the role of observational learning. Vision Research, 51(8), 945954. Kail, R.(1986). Sources of age differences in speed of processing. Child Development, 57, 969-987. Keele, S. W.(1968). Movement control in skilled motor performance. Psychological Bulletin, 70, 387-409. Keller, P. E., Wascher, E., Prinz, W., Waszak, F., Koch, I.,& Rosenbaum, D. A.(2006). Differences between intention-based and stimulus-based actions. Journal of Psychophysiology, 20(1), 9-20. Kelly, T.(2000). The development of executive function in school-aged children. Clinical Neuropsychological Assessment, 1, 38-55. Kirpatrick, E. V., Pearse, J. E., Eyre, J. A.,& Basu, A. P.(2013). Motor planning ability is not related to lesion side or functional manual ability in children with hemiplegic cerebral palsy. Experimental Brain Research, 231, 239-247. Knudsen, B., Henning, A., Wunsch, K., Weigelt, M.,& Aschersleben, G.(2012). The endstate comfort effect in 3- to 8-year-old children in two object manipulation tasks. Frontiers in Psychology, 3, 1-10. Kuhtz-Buschbeck, J. P., Stolze, H., Boczek-Funcke, A., Joehnk, K., Heinrichs, H.,& Illert, M. (1998). Kinematic analysis of prehension movements in children. Behavioral Brain Research, 93, 131-141. 160

References Kunde, W., Elsner, K.,& Kiesel, A.(2007). No anticipation- no action. The role of anticipation in action and perception. Cognitive Processing, 8, 71-78. Lange-Kuettner, C.(2009). Habitual size and projective size: The logic of spatial systems in children's drawings. Developmental Psychology, 45(4), 913-927. Lashley, K.(1951). The problem of serial order in behavior. In L. Jeffress(Ed.), Cerebral Mechanisms in Behavior( pp. 112-131). New York: Wiley. Levin, H., Culhane, K., Hartmann, J., Evankovich, K., Mattson, A., Harward, H., Ringholz, G., Ewing-Cobbs, L.,& Fletcher, J.(1991). Developmental changes in performance on tests of purported frontal lobe functioning. Developmental Neuropsychology, 7, 377395. Lockman, J. J.(2000). A perception-action perspective on tool use development. Child Development, 71(1), 137-144. Logan, S. W.,& Fischman, M.G.(2011). The relationship between end-state comfort effects and memory performance in serial and free recall. Acta Psychologica, 137, 292-299. Lotze, R. H.(1852). Medicinische Psychologie oder die Physiologie der Seele. Leipzig: Weidmann'sche Buchhandlung. Luria, A.(1973). The working brain: An introduction to neuropsychology. New York: Basic Books. Manoel, E.,& Conolly, K.(1997). Variability and stability in the development of skilled actions. In K. Conolly& H. Forssberg(Eds.), Neurophysiology of motor development (pp. 286-318). London: Mac Keith Press; Cambridge University Press. Manoel, E.,& Moreira, C.(2005). Planning manipulative hand movements: Do young children show the End-State Comfort Effect? Journal of Human Movement Studies, 49, 93-114. doi: Marmor, G. S.(1975). Development of kinetic images: when does the child first represent movement in mental images? Cognitive Psychology, 7, 548-559. McCarthy, M. E., Clifton, R. K.,& Collard, R. R.(1999). Problem solving in infancy: The emergence of an action plan. Developmental Psychology, 35(4), 1091-1101. McCarthy, M. E.,& Keen, R.(2005). Facilitating problem-solving performance among 9- and 12-month-old infants. Journal of Cognition and Development, 6(2), 209-228. 161

References Marteniuk, R. G., MacKenzie, C. L., Jeannerod, M., Athenes, S.,& Dugas, C.(1987). Constraints on human arm movement trajectories. Canadian Journal of Psychology, 41, 365-378. Masson, M. E. J., Bub, D. N.,& Breuer, A. T.(2011). Priming of reach and grasp actions by handled objects. Journal of Experimental Psychology: Human Perception and Performance, 37(5), 1470-1484. McGraw, M.(1953). Growth: A study of Johnny and Jimmy. New York: International Universities Press. Mechling, H.,& Munzert, J.(2003). Handbuch Bewegungswissenschaft ­ Bewegungslehre. Stuttgart: Hofmann. Meulenbroek, R. G.,& van Galen, G. P.(1988). The acquisition of skilled handwriting: Discontinuous trends in kinematic variables. Advances in Psychology, 55, 273-281. Mischel, W., Ebbesen, E.B.,& Raskoff Zeiss, A.(1972). Cognitive and attentional mechanism in delay of gratification. Journal of Personality and Social Psychology, 21(2), 204-218. Moher, D., Liberati, A., Tetzlaff, J., Altman, D.G.,& The PRISMA Group(2009). Preferred Reporting Items for Systematic Reviews and Meta-Analyses: The PRISMA Statement. PLoS Med, 6(7), 1 ­ 6. Münsterberg, H.(1888). Die Willenshandlung. Ein Beitrag zur physiologischen Psychologie. Freiburg: Mohr. Mutsaarts, M., Steenbergen, B.,& Bekkering, H.(2006). Anticipatory planning deficits and task context effects in hemiparetic cerebral palsy. Journal of Experimental Brain Research, 172, 151-162. Monsell, S.,& Driver, J.(2000). Banishing the control homunculus. In S. Monsell,& J. Driver (Eds.), Control of cognitive processes( pp. 3-32). Cambridge: MIT Press. National Institute of Mental Health(2012). Attention deficit hyperactivity disorder. Retrieved August 6, 2014 from http://www.nimh.nih.gov/health/publications/attention-deficithyperactivity-disorder/index.shtml Nosek, B.A.,& Banaji, M.R.(2001). The Go/No-Go Association Task. Social Cognition, 19(6), 625-666. 162

References Nowrangi, M. A., Lyetsos, C.,& Munro, C. A.(2014). Systematic review of neuroimaging correlates of executive functioning: Converging evidence from different clinical populations. Journal of Neuropsychiatry and Clinical Neuroscience, 26(2), 114-125. Oztop, E., Arbib, M.,& Bradley, N.(2006). The development of grasping and the mirror system. In M. Arbib(Ed.), Action to Language via the Mirror Neuron System(pp. 397423). Cambridge University Press. Oztop, E., Bradley, N.,& Arbib, M.(2004). Infant grasp learning: a computational model. Experimental Brain Research, 158(4), 480-503. Passler, M., Isaac, W.,& Hynd, G.(1985). Neuropsychological development of behavior attributed to frontal lobe functioning in children. Developmental Neuropsychology, 1, 349-370. Paulus, M., Hunnius, S., Vissers, M.,&Bekkering, H.(2011). Bridging the gap between the other and me: The functional role of motor resonance and action effects in infants' imitation. Journal of Developmental Science, 14, 901-910. Payne, V. G.,& Isaacs, L. D.(1999). Human motor development: A lifespan approach. Mountain View: Mayfield. Petermann, F.,& Petermann, U.(Hrsg.)(2008). Hamburg-Wechsler-Intelligenztest für Kinder IV(HAWIK-IV). Bern: Verlag Hans Huber. Piaget, J.(1936). Das Erwachen der Intelligenz beim Kinde. Stuttgart: Klett. Piaget, J.,& Cook, M.(1952). The origins of intelligence in children. New York: W.W. Norton & Co. Prinz, W.(1990). A common coding approach to perception and action. In O. Neumann& W. Prinz(Eds.), Relationships between perception and action(pp. 167-201). Berlin: Springer. Prinz, W.(1997). Perception and action planning. European Journal of Cognitive Psychology, 9(2), 129-154. Prior, M.,& Hoffmann, W.(1990). Brief report: Neuropsychological testing of autistic children through an exploration with frontal lobe tests. Journal of Autism and Developmental Disorders, 20(4), 581-590. Randerath, J., Li, Y., Goldenberg, G.,& Hermsdörfer, J.(2009). Grasping tools: Effects of task and apraxia. Neuropsychologia, 47(2), 497-505. 163

References Ricken, G., Fritz, A., Schuck, K. D.,& Preuß, U.(Hrsg.)(2007). Hannover-WechslerIntelligenztest für das Vorschulalter ­ III(HAWIVA-III). Bern: Verlag Hans Huber. Roberts, A. C.(2003). Introduction. In A.C. Roberts, T.W. Robbins,& L. Weiskrantz(Eds.), The prefrontal cortex. Executive and cognitive functions(pp. 1-5). Oxford: Oxford University Press. Robinson, L. E.,& Fischman, M.G.(2012). Motor planning in preschool children ­ The use of a visual cue to test the end-state comfort effect: A preliminary study. Early Child Development and Care, 183(5), 605-612. Rosenbaum, D. A.(2010). Human Motor Control. San Diego: Academic Press. Rosenbaum, D. A., Carlson, R. A.,& Gilmore, R. O.(2001). Acquisition of intellectual and perceptual-motor skills. Annual Review of Psychology, 52, 453-470. Rosenbaum, D. A., Chapman, K. M., Weigelt, M., Weiss, D. J.,& van der Wel, R.(2012). Cognition, action, and object manipulation. Psychological Bulletin, 138(5), 924-946. Rosenbaum, D. A., Cohen, R. G., Dawson, A. M., Jax, S. A., Meulenbroek, R. G., van der Wel, R.,& Vaughan, J.(2007). The posture-based motion planning framework: New findings related to object manipulation, moving around obstacles, moving in three spatial dimensions, and haptic tracking. In D. Sternad(Ed.), Progress in motor control (pp. 485-497). Heidelberg: Springer. Rosenbaum, D. A., Cohen, R. G., Jax, S. A., Weiss, D. J.,& van der Wel, R.(2007). The problem of serial order in behavior: Lashley's legacy. Human Movement Science, 26(4), 525-554. Rosenbaum, D. A., Halloran, E. S.,& Cohen, R. G.(2006). Grasping movement plans. Psychonomic Bulletin& Review, 13(5), 918-922. Rosenbaum, D. A.,& Jorgensen, M.J.(1992). Planning macroscopic aspects of manual control. Human Movement Science, 11, 61-69. Rosenbaum, D. A., Marchak, F., Barnes, H. J., Vaughan, J., Slotta, J. D.& Jorgensen, M. J. (1990). Constraints for action selection: overhand versus underhand grips. In M. Jeannerod(Ed.). Attention and Performance XIII(pp. 321-342). Hillsdale: Erlbaum. Rosenbaum, D. A., Meulenbroek, R. G. J.,& Vaughan, J.(1996). Three approaches to the degrees of freedom problem in reaching. In A. M. Wing, P. Haggard,& R. Flanagan (Eds.), Hand and Brain: Neurophysiology and psychology of hand movement(pp. 169-185). San Diego: Academic Press. 164

References Rosenbaum, D. A., van Heugten, C.,& Caldwell, G. C.(1996). From cognition to biomechanics and back: The end-state comfort effect and the middle-is-faster effect. Acta Psychologica, 94, 59-85. Rosenbaum, D. A., Vaughan, J., Jorgensen, M. J., Barnes, H. J.,& Stewart, E.(1993). Plans for object manipulation. In D. E. Meyer& S. Kornblum(Eds.), Attention and Performance IV: Silver jubilee: Synergies in experimental psychology, artificial intelligence and cognitive neuroscience(pp. 803-820). Cambridge: MIT Press. Rosenbaum, P., Paneth, N., Leviton, A., Goldstein, M.,& Bax, M.(2007). A report: The definition and classification of cerebral palsy April 2006. Developmental Medicine& Child Neurology, 49, 8-14. Rovee, C.K.,& Rovee, D.T.(1986). Conjugate reinforcement of infants' exploratory behavior. Journal of Experimental Child Psychology, 8, 33-39. Ryan, R.,& Deci, E. L.(2000). Self-determination theory and the facilitation of intrinsic motivation, social development, and well-being. American Psychologist, 55(1), 68-78. Salthouse, T. A.(1991). Theoretical perspectives on cognitive aging. Hillsdale: Erlbaum. Salthouse, T. A., Atkinson, T. M.,& Berish, D. E.(2003). Executive functioning as a potential mediator of age-related cognitive decline in normal adults. Journal of Experimental Psychology, 132, 566-594. Scharoun, S. M.,& Bryden, P. J.(2013). The development of end- and beginning-state comfort in a cup manipulation task. Journal of Developmental Psychobiology, 56(3), 407420. Scheres, A., Oosterlaan, J., Geurts, H., Morein-Zamir, S., Meiran, N., Schut, H., Vlasveld, L., & Sergeant, J. A.(2004). Executive functioning in boys with ADHD: Primarily an inhibition deficit? Archives od Clinical Neuropsychology, 19(4), 569-594. Seegelke, C., Hughes, C. M. L.,& Schack, T.(2011). An investigation into manual asymmetries in grasp behavior and kinematics during an object manipulation task. Experimental Brain Research, 215, 65-75. Shallice, T.(1990). From neuropsychology to mental structure. New York: Oxford University Press. Short, M. W.,& Cauraugh, J. H.(1999). Precision hypothesis and the end-state comfort effect. Acta Psychologica, 100, 243-252. 165

References Smyth, M. M.,& Mason, U. C.(1997). Planning and execution of action in children with and without developmental coordination disorder. Journal of Child Psychology and Psychiatry, 38(8), 1023-1037. Spiegel. M. A., Koester, D., Weigelt, M.,& Schack, T.(2012). The cost of changing and intended action: Movement planning, but not execution, interfers with verbal working memory. Neuroscience Letters, 509, 82-86. Steenbergen, B.,& Gordon, A.(2006). Activity limitation in hemiplegic cerebral palsy: Evidence for disorders in motor planning. Developmental Medicine and Child Neurology, 48, 780-783. Steenbergen, B.,& van der Kamp, J.(2004). Control of prehension in hemiparetic cerebral palsy: Similiarities and differences between the ipsi- and contra-lesional sides of the body. Developmental Medicine and Child Neurology, 46(5), 325-332. Steenbergen, B., van Nimwegen, M.,& Crajé, C.(2007). Solving a mental rotation task in congenital hemiparesis: Motor imagery versus visual imagery. Neuropsychologia, 45(14), 3324-3328. Steenbergen, B., Verrel, J.,& Gordon, A. M.(2007). Motor planning in congenital hemiplegia. Disability and Rehabilitation, 29, 13-23. Stock, A.,& Stock, C.(2004). A short history of ideo-motor action. Psychological Research, 68, 176-188. Stöckel, T., Hughes, C. M. L.,& Schack, T.(2012). Representation of grasp postures and anticipatory motor planning in children. Psychological Research, 76(6), 768-776. Stuss, D. T.(1992). Biological and psychological development of executive functions. Brain and Cognition, 20(1), 8-23. Stuss, D. T.(2011). Functions of the frontal lobes: Relation to executive functions. Journal of the International Neuropsychological Society, 17, 759-765. Stuss, D.,& Benson, D.(1984). Neuropsychological studies of the frontal lobes. Psychological Bulletin, 95, 3-28. Stuss, D. T.,& Alexander, M. P.(2000). Executive functions and the frontal lobes: a conceptual view. Psychological Research, 63, 289-298. Thelen, E.& Smith, L. B.(1994). A dynamic systems approach to the development of cognition and action. Cambridge: MIT Press. 166

References Thibaut, J.-P.,& Toussaint, L.(2010). Developing motor planning over ages. Journal of Experimental Child Psychology, 105(1-2), 116-129. Toussaint, L., Tahej, P.-K., Thibaut, J.-P., Possamai, C.-A.,& Badets, A.(2013). On the link between action planning and motor imagery: a developmental study. Experimental Brain Research, 231, 331-339. Tucker, M.,& Ellis, R.(1998). On the relations between seen objects and components of potential actions. Journal of Experimental Psychology: Human Perception and Performance, 24(3), 830-846. Van der Wel, R.,& Rosenbaum, D.A.(2010). Bimanual grasp planning reflects changing rather than fixed constraint dominance. Journal of Experimental Brain Research, 205, 351-362. Van Swieten, L. M., van Bergen, E., Williams, J. H. G., Wilson, A. D., Plumb, M. S., Kent, S. W.,& Mon-Williams, M. A.(2010). A test of motor(not executive) planning in developmental coordination disorder and autism. Journal of Experimental Psychology: Human Perception and Performance, 36(2), 493-499. Vereijken, B., van Emmerik, R. E. A., Bongaardt, R., Beek, W. J.,& Newell, K. M.(1997). Changing coordinative structures in complex skill acquisition. Human Movement Science, 16(6), 823-844. Vereijken, B., van Emmerik, R. E. A., Whiting, H. T. A.,& Newell, K. M.(1992). Free(z)ing degrees of freedom in skill acquisition. Journal of Motor Behavior, 24, 133-142. Von Aster, M., Neubauer, A.,& Horn, R.(Hrsg.)(2006). Wechsler Intelligenztest für Erwachsene(WIE). Frankfurt am Main: Harcourt Test Services. Von Hofsten, C.(2004). An action perspective on motor development. Trends in Cognitive Science, 8(6), 266-272. Von Hofsten, C.,&Fazel-Zandy, S.(1984). Development of visually guided hand orientation in reaching. Journal of Experimental Child Psychology, 38, 208-219. Von Hofsten, C.,& Ronnqvist, L.(1988). Preparation for grasping an object: A developmental study. Journal of Experimental Psychology: Human Perception and Performance, 14, 610-621. Von Holst, E.,& Mittelstaedt, H.(1950). Das Reafferenzprinzip. Naturwissenschaften, 37, 464-476. Vygotski, L.S.(1977). Denken und Sprechen. Frankfurt: Fischer. 167

References Wechsler, D.(1997). Wechsler Adult Intelligence Scale ­ 3rd edition(WAIS®-III). San Antonio: Harcourt Assessment Services. Wechsler, D.(2002). Wechsler Primary and Preschool Scale of I ntelligenceTM ­ 3rd edition (WPPSITM -III). San Antonio: Harcourt Assessment Services. Wechsler, D.(2003). Wechsler Intelligence Scale for Children ­ 4th edition(WISC-IV®). San Antonio: Harcourt Assessment Services. Weigelt, M., Cohen, R. G.,& Rosenbaum, D. A.(2007). Returning home: Locations rather than movements are recalled in human object manipulation. Experimental Brain Research, 179, 191-198. Weigelt, M., Kunde, W.,& Prinz, W.(2006). End-state comfort in bimanual object manipulation. Experimental Psychology, 53(2), 143-148. Weigelt, M., Rosenbaum, D. A., Huelshorst, S.,& Schack, T.(2009). Moving and memorizing: Motor planning modulates the recency effect in serial and free recall. Acta Psychologica, 132, 68-79. Weigelt, M.,& Schack, T.(2010). The development of end-state comfort planning in preschool children. Experimental Psychology, 57(6), 476-782. Welsh, M. C.(1991). Rule-guided behaviour and self-monitoring on the Tower of Hanoi disktransfer task. Cognitive Development, 6(1), 59-76. Welsh, M. C., Pennington, B. F.,& Groisser, D. B.(1991). A normative-developmental study of executive function: A window on prefrontal function in children. Developmental Neuropsychology, 7, 131-149. Weiss, D. J., Wark, J. D.,& Rosenbaum, D. A.(2007). Monkey see, monkey plan, monkey do: The end-state comfort effect in cotton-top tamarins. Psychological Science, 18(12), 1063-1068. Wilmut, K.,& Byrne, M.(2014a). Grip selection for sequential movements in children and adults with and without developmental coordination disorder. Human Movement Science, 36, 272-284. Wilmut, K.,& Byrne, M.(2014b). Influences of grasp selection in typically developing children. Acta Psychologica, 148, 181-187. Wolpert, D. M., Ghahramani, Z.,& Jordan, M. I.(1995). An internal model for sensorimotor integration. Science, 269, 1880-1882. Wunsch, K., Weigelt, M.,& Stöckel, T. Motor planning in older adults. In prep. 168

References Wunsch, K., Weiss, D., Schack, T.,& Weigelt, M.(2013). Second order motor planning in children: Insights from a cup-manipulation-task. Psychological Research, 1-9. Advanced online publication. Zander, S. L., Weiss, D. J.,& Judge, P. G.(2013). The interface between morphology and action planning: a comparison of two species of New World monkeys. Animal Behaviour, 86(6), 1251-1258. Zelazo, P. D., Carter, A., Reznick, J. S.,& Frye, D.(1997). Early development of executive function: A problem-solving framework. Review of General Psychology, 1, 1-29. 169

References Supplements Supplement A: Means for the tasks of the experimental study..............................171 Supplement B: Informed consent declaration........................................................177 Supplement C: Stamp Card(randomization).........................................................179 Supplement D: Test instructions and protocols.....................................................180 Statutory Declaration............................................................................................211 Acknowledgements...............................................................................................213 170

Supplements Supplements Supplement A ­ Means for the tasks of the experimental study A1. Mean percentage of ESC grasps in the bar-transport-task divided by age groups. 3-year-olds 4-year-olds 5-year-olds 6-year-olds 7-year-olds 8-year-olds 9-year-olds 10-year-olds Adults Uncritical trials Trial 1 Trial 2 Trial 3 100 100 100 95,5 100 100 100 95,2 100 100 100 100 95,5 100 100 91,3 95,2 100 90,5 95,7 96,2 100 88,5 100 96,3 100 100 Critical trials Trial 1 Trial 2 Trial 3 14,3 47,8 42,3 54,5 57,7 59,3 69,6 85,7 96,4 28,6 47,8 61,5 63,6 61,5 74,1 82,6 95,2 96,4 33,3 47,8 61,5 59,1 65,4 63,0 82,6 100 92,9 ESC in> 50% of trials Uncritical Critical 100 23,8 100 47,8 100 61,5 100 63,6 100 61,5 100 63,0 100 78,3 100 95,2 100 96,4 Note. All values describe percentages of participants. Trials 1,2 and 3 were not assigned in the demonstrated order, but were presented randomly. The mean percentages demonstrated in the last two rows were applied to statistical analyses. 171

Supplement A ­ Means for the tasks of the experimental study A2. Mean percentage of ESC grasps in the sword-rotation-task divided by age groups. 3-year-olds 4-year-olds 5-year-olds 6-year-olds 7-year-olds 8-year-olds 9-year-olds 10-year-olds Adults Uncritical trials Critical trials Position1 Position 4 Position 5 Position 6 Position 2 Position 3 Trial 1 Trial 2 Trial 3 Trial 1 Trial 2 Trial 3 Trial 1 Trial 2 Trial 3 Trial 1 Trial 2 Trial 3 Trial 1 Trial 2 Trial 3 Trial 1 Trial 2 Trial 3 100 100 100 85,7 90,5 95,2 95,2 95,2 100 100 100 100 47,6 57,1 47,6 23,8 52,4 47,6 100 95,7 100 87,0 91,3 87,0 100 100 100 100 100 100 60,9 39,1 43,5 39,1 56,5 34,8 100 92,3 92,3 100 100 100 96,2 100 96,2 100 96,2 100 76,9 69,2 73,1 73,1 69,2 80,8 100 95,5 90,9 81,8 86,4 77,3 95,5 95,5 95,5 100 95,5 95,5 63,6 72,7 72,7 59,1 68,2 68,2 100 96,2 96,2 96,2 96,2 96,2 100 96,2 96,2 100 100 100 61,5 57,7 65,4 65,4 73,1 80,8 96,3 100 96,3 96,3 96,3 92,6 100 96,3 92,6 100 100 100 74,1 81,5 88,9 81,5 77,8 81,5 100 100 87,0 100 100 95,7 100 91,3 95,7 95,7 100 95,7 65,2 69,6 78,3 78,3 82,6 73,9 100 100 95,2 90,5 100 100 100 100 100 100 95,2 100 76,2 90,5 85,7 90,5 81,0 81,0 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 ESC in> 50% of Uncritical 100 100 100 100 100 100 100 100 100 Critical 42,9 43,5 57,7 63,6 73,1 74,1 78,3 76,2 100 Note. All values describe percentages of participants. Positions 1 to 6 and Trials 1,2 and 3 were not assigned in the demonstrated order, but were presented randomly. The mean percentages demonstrated in the last two rows were applied to statistical analyses. 172

Supplement A ­ Means for the tasks of the experimental study A3. Mean grasp heights in the grasp-height-task divided by age groups. 3-year-olds 4-year-olds 5-year-olds 6-year-olds 7-year-olds 8-year-olds 9-year-olds 10-year-olds Adults Grasp height[cm] High target(HT) Low target(LT) HtT TtH HtT TtH 48,7 37,2 53,1 70,6 48,2 40,6 54,4 76,2 54,0 47,0 56,1 74,9 53,5 49,8 52,9 70,8 51,1 50,8 55,2 72,1 59,0 59,8 64,3 75,8 58,7 59,7 60,2 70,5 58,4 60,6 60,7 69,8 41,4 50,1 56,6 56,0 ESC[% of participants] HT 19,0 34,8 19,2 31,8 38,5 33,3 39,1 45,5 92,9 LT 4,8 4,3 7,7 0,0 3,8 22,2 26,1 13,6 50,0 Mean 11,9 19,6 13,5 15,9 21,2 27,8 32,6 29,5 71,4 Legend. HtT= Home-to-Target moves; TtH= Target-to-home moves Note. Presented values for HtT and TtH moves reflect mean grasp height collapsed over all three HtT and TtH trials, respectively. The last three rows show the percentage of participants, who showed ESC. ESC was declared as being present, if grasp height was lower in TtT than in TtH moves in moves to the high target platform, and if grasp height was higher in HtT than in TtH moves in moves to the low target platform. 173

Supplement A ­ Means for the tasks of the experimental study A4. Mean performance on the Tower-of-Hanoi-task divided by age groups. 3-year-olds 4-year-olds 5-year-olds 6-year-olds 7-year-olds 8-year-olds 9-year-olds 10-year-olds Adults Accomplished items First mistake Latency Time per Superflous Rule Correctly item moves violations solved [# of items][# of item][# of moves][in s][in s][# of moves][amount] [% of accomplished items] 1,3 1,4 1,1 16,5 55,5 1,5 2,4 2,2 1,1 8,8 38,9 1,8 4,6 3,1 1,2 7,4 52,7 3,4 5,4 2,2 1,3 7,2 43,5 3,5 4,2 1,4 1,8 11,5 61,2 4,8 4,8 1,7 1,7 6,1 45,0 5,0 4,9 1,7 2,0 5,2 40,5 5,0 5,7 2,0 1,7 5,4 40,8 5,8 6,6 2,8 2,6 7,3 48,6 5,3 2,2 38,9 2,0 73,9 1,2 63,5 2,3 42,2 1,0 31,1 1,1 28,3 1,3 42,8 0,5 35,0 0,1 44,3 Solved items [% of total item s] 12,9 23,9 45,8 53,6 41,9 47,8 49,1 56,7 66,1 Note. The table shows the values for performance in the TOH-task. Every participant had the chance to complete 10 items. The task was terminated if 1) the total amount of 2xn moves was exceeded, or if 2) the participant did not make any move within 30 seconds. Correctly solved items indicate those items, which were solved in the optimal amount of moves without any superfluous moves or mistakes. Solved items, however, include all items until termination, no matter if they were completed correctly, or just within 2xn moves. For statistical analyses, the amount of solved items was used. 174

Supplement A ­ Means for the tasks of the experimental study A5. Mean performance on the Mosaic-task divided by age groups. 3-year-olds 4-year-olds 5-year-olds 6-year-olds 7-year-olds 8-year-olds 9-year-olds 10-year-olds Adults Accomplished items [# of items] 11,6 15,2 19,2 12,2 9,3 11,5 11,7 13,0 13,7 First mistake [# of item] 8,7 14,8 16,8 11,7 9,1 9,3 10,6 11,3 11,9 Time per item [in s] 12,9 23,2 25,6 30,7 26,9 30,9 33,5 35,5 35,1 Incorrect pattern [# of items] 4,6 4,3 2,7 2,4 3,1 1,9 1,4 0,7 0,3 Time exceedance [# of items] 1,1 2,1 2,8 2,5 2,8 1,9 2,0 1,1 0,8 Correctly solved [% of accomplished items] No time Time bonus Possible bonus points [# of points][# of points][# of points] 50,5 17,3 17,3 46 66,0 16,5 16,5 34 83,3 24,2 24,2 34 53,2 22,5 22,5 28 66,5 25,7 26,7 50 82,0 31,7 34,9 46 83,9 34,6 38,3 46 92,5 39,3 43,0 46 97,7 31,1 43,3 46 Points [% of total points] 37,6 48,5 71,2 55,1 51,4 69,0 75,2 85,5 67,7 Note. The table shows the values for performance in the Mosaic-task. According to the different test protocols, possibly reachable points varied(see last row). The task was terminated if participants failed to solve three consequtive patterns du to 1) time exceedance or 2) the builing of an incorrect pattern. Correctly solved items indicate those items, which were solved within the time limit, without a second trial, and without failures in the pattern. In thetest protocol of older children and adults, time boni could be reached; the faster the pattern is built, the more points participants get. For statistical anlyses, points without time bonus were used. 175

Supplement A ­ Means for the tasks of the experimental study A6. Mean performance on the D2-task divided by age groups. 3-year-olds 4-year-olds 5-year-olds 6-year-olds 7-year-olds 8-year-olds 9-year-olds 10-year-olds Adults Processed items [# of items] 26,9 34,7 -------IDS 3-5 Mistakes Points [# of items][# of points] 0,7 26,1 0,5 34,2 ---------------------Total [% of total points] 36,3 47,5 -------Processed items [# of items] --61,0 77,8 94,7 105,4 127,3 139,1 -IDS 5-10 Mistakes Points Total [# of items][# of points] [% of total points] ----------7,8 53,2 28,1 8,7 69,1 36,6 8,9 85,8 45,4 6,1 99,3 52,5 5,6 121,7 64,4 4,3 134,9 71,4 -----Processed items [# of items] --------191,5 D2-R Mistakes Points Total [# of items][KL total] [% of total points] ----------------------------------------7,2 177,5 43,1 Note. The number of processed items reflect the summarized number of items in each row up to the last item which was sketched out. Mistakes can either be missed items, or items that have falsly sorted out. Points indicate the number of processed items with the number of mistakes subtracted. The total percentage of reached points was taken for statistical analyses. 176

Supplement B ­ Informed consent declaration Supplement B ­ Informed consent declaration 177

Supplement B ­ Informed consent declaration 178

Supplement C ­ Stamp card(randomization) Supplement C ­ Stamp card(randomization) 179

Supplement D ­ Test instructions and protocols Supplement D ­ Test instructions and protocols D1. Instruction and test protocol for the bar-transport-task For task and procedure for the bar-transport-task, please see Chapter 3.2.2.1.1. D1.1. Instruction ,,Bitte stelle dich hier hinter diese Linie [ca. 10 cm vor der Tischkante]. Stehe aufrecht und lasse beide Arme seitlich am Körper locker herunterhängen, so dass die Handflächen die Oberschenkel berühren. Vor dir auf dem Tisch siehst du einen zweifarbigen Stab. Dieser ist auf einer Seite schwarz, auf der anderen Seite weiß. Kannst du mir einmal die weiße Seite zeigen? Und die schwarze? Sehr gut. Ich werde dich nun anweisen, entweder das schwarze oder das weiße Ende des Stabes in diese Öffnung zu stecken[zeigen!]. Greife daraufhin den Stab mit der Stempelhand und stecke die Seite des Stabes, die ich dir zuvor genannt habe, in die Öffnung ein. Achte darauf, dass du den Stab fest mit der ganzen Hand umschlossen hältst. Löse daraufhin deinen Griff und lasse beide Hände wieder seitlich am Körper hängen wie zuvor. Ich lege den Stab dann wieder zurück und werde dir wieder eine Farbe nennen. Hast du noch Fragen? Gut, dann starten wir jetzt ." 180

Supplement D ­ Test instructions and protocols D1.2. Test protocol Legend. Versuch= trial Orientierung= orientation Instruktion= instruction Griffauswahl= grasp choice ESC Ja/Nein= showed ESC Yes/No w= white(weiß) s= black(schwarz) Note. Here, two exemplary test protocols are depicted. Every participant had six trials(Versuche), and the start orientation(Orientierung) was counterbalanced across participants, as color-coded in the protocol. The instruction(Instruktion) was assigned randomly, with"w" meaning"white"(weiß),"s" meaning "black"(schwarz). Grasp choice was noted("OH"("Oberhand") for an overhand grip,"UH"("Unterhand") for an underhand grip), and the presence of ESC was recorded in the last row(ESC Ja/Nein). 181

Supplement D ­ Test instructions and protocols D2. Instruction and test protocol for the sword-rotation-task For task and procedure for the sword-rotation-task, please see Chapter 3.2.2.1.2. D2.1. Instruction ,,Bitte stelle dich hier hinter diese Linie [ca. 10 cm vor der Tischkante]. Stehe aufrecht und lasse beide Arme seitlich am Körper locker herunterhängen, sodass die Handflächen die Oberschenkel berühren. Vor dir auf dem Tisch siehst du ein Schwert. Du kannst es dir gerne anschauen. Aber pass auf: Die Klinge ist scharf! Du darfst das Schwert nur hier am Griff anfassen, so wie es ein Pirat auch machen würde. Jetzt kannst du uns zeigen, ob du auch schon ein richtiger Pirat bist. Wir möchten von dir lernen, wie man das Schwert richtig in die Holzbox steckt. Kannst du uns das einmal zeigen?[Versuch#1 immer in Position 1]. Bitte nimm dazu die Stempelhand. Sobald es in der Box steckt, lass es bitte los und lasse die Arme wieder seitlich neben dem Körper hängen. Gut das klappt ja schon super. Das machen wir jetzt noch einige Male, damit ich sehe, ob du das immer so gut kannst. Aber denke daran: greife das Schwert immer nur am Griff, und greife es immer fest mit der ganzen Hand. Du musst es fest halten, sodass es sich nicht in deiner Hand bewegt oder herunterfällt. Ich werde das Schwert immer wieder hier vor dich auf den Tisch legen und du versuchst dann, es mit der Klinge in die Box zu stecken. Danach lässt du es los und stellst dich wieder so hin wie jetzt gerade. Hast du noch Fragen? Gut, dann starten wir." 182

Supplement D ­ Test instructions and protocols D2.2. Test protocol Legend. Versuch= trial Orientierung= orientation; numbers 1 to 6 indicate the start orientation of the sword Griffauswahl= grasp choice ESC Ja/Nein= showed ESC Yes/No Note. Every participant had eighteen trials(Versuche), divided into three blocks, with every orientation of the sword(Orientierung) appearing once in a block in a randomized fashion. Grasp choice was noted ("hin"("towards") for a grip with the thumb pointing to the blade of the sword,"weg"("away") for a grip with the thumb pointing to the blade). Presence of ESC was recorded in the last row(ESC Ja/Nein). 183

Supplement D ­ Test instructions and protocols D3. Instruction and test protocol for the grasp-height-task For task and procedure for the grasp-height-task, please see Chapter 3.2.2.1.3. D.3.1. Instruction ,, Bitte stelle dich hier hinter diese Linie[10 cm vor den Regalböden] vor die mittlere Plattform. Stehe aufrecht und lasse beide Arme seitlich am Körper locker herunterhängen, sodass die Handflächen die Oberschenkel berühren. Ich mache jetzt ein Video von dir, wie du diese Aufgabe ausführst. Mit Hilfe dieses Videos soll später ein Roboter programmiert werden, der sich dann genau so bewegen soll wie du. Es ist daher sehr wichtig, dass du die Aufgabe genau so ausführst, wie ich sie dir gerade erkläre. Du hast gleich die Aufgabe, den Pömpel vor dir entweder nach oben, nach unten, oder in die Mitte des Regals zu stellen. Wohin genau, sage ich dir immer, sobald du wieder die Ausgangsposition eingenommen hast. Du solltest dazu mit den Füßen hinter der Linie stehen und deine Arme rechts und links am Körper hängen lassen. Ich werde dir vorher immer sagen, wohin du den Pömpel stellen sollst. Also warte bitte immer in der Ausgangshaltung, bis ich dir eine Anweisung gebe. Greife dann nach dem Pömpel und halte ihn fest in der Hand, damit er nicht verrutscht während du ihn auf die Regalhöhe stellst, die ich dir zuvor genannt habe. Benutze hierfür immer die Stempelhand. Wenn du keine Fragen mehr hast, beginnen wir. Bitte nimm die Ausgangshaltung ein." 184

Supplement D ­ Test instructions and protocols D3.2. Test protocol Vpn:_______________________ Alter des Probanden in Monaten:__________ Größe des Probanden:______________ Höhe des mittleren Regalbodens:_____________ Testreihenfolge: Oben Oben Griffhöhe:_____________ cm Griffhöhe:_____________ cm Oben Unten Griffhöhe:_____________ cm Griffhöhe:_____________ cm Unten Unten Griffhöhe:_____________ cm Griffhöhe:_____________ cm Vpn:_______________________ Alter des Probanden in Monaten:__________ Größe des Probanden:_____________ Höhe des mittleren Regalbodens:_____________ Testreihenfolge: Unten Griffhöhe:_____________ cm Unten Griffhöhe:_____________ cm Unten Griffhöhe:_____________ cm Oben Griffhöhe:_____________ cm Oben Griffhöhe:_____________ cm Oben Griffhöhe:_____________ cm Legend. Vpn= participant Alter des Probanden in Monaten= age of participant in moth Größe des Provbanden= height of participant Höhe des mittleren Regalbodens= height oft he middle shelf board Testreihenfolge= testing sequence Unten= lower shelfboard Oben= high shelf board Griffhöhe= Grasp height Note. Here, the two possible test protocols are depicted. Every participant had six trials, divided into two blocks, with shelf heights blocked. The start height was counterbalanced across participants. Grasp height was analyzed offline, using a frame-by-frame video software(Super), stopping the video at the time of the first contact with the plunger. Then, grasp height was counted relative to the plunger base. 185

Supplement D ­ Test instructions and protocols D4. Instruction and test protocol for the Tower-of-Hanoi-task For task and procedure for the Tower-of-Hanoi-task, please see Chapter 3.2.2.2.1. D4.1. Instruction ,, Auf diesem Bild siehst du einen Turm. Deine Aufgabe ist es, diesen Turm stets auf dem markierten Stab(dem Gleichen, den du auf dem Bild siehst) nachzubauen. Aber die Anfangsposition ist immer eine andere. Du kannst dir das Bild jederzeit anschauen. Es gelten jedoch einige Regeln: 1. Du darfst immer nur eine Scheibe bewegen. 2. Die Scheiben sind entweder in deiner Hand oder auf einer der drei Stäbe; du darfst sie nicht auf dem Tisch oder dem Gestell ablegen. 3. Du darfst nur kleinere Scheiben auf größere Scheiben setzen, keine größere Scheibe auf eine kleinere. Bitte versuche, den ersten Turm nachzubauen. Du darfst mich gerne fragen, sollte dir etwas unklar sein. Das hat ja super geklappt. Ich werde nun einen Sichtschutz hier aufstellen, während ich deine neue Aufgabe vorbereite. Sobald ich diesen entferne, darfst du beginnen. Ich werde dann die Zeit stoppen. Aber: Es ist wichtiger, dass du die Aufgabe richtig löst, anstatt sie schnell zu lösen. Lass dir daher ruhig Zeit und denke gut nach. Hast du noch Fragen? Gut, dann beginnen wir: Fertig ­ los!" 186

Supplement D ­ Test instructions and protocols D4.2. Test protocol for 3- to 4-year old children 187

Supplement D ­ Test instructions and protocols 188

Supplement D ­ Test instructions and protocols D4.3. Test protocol for 5- to 6-year old children 189

Supplement D ­ Test instructions and protocols 190

Supplement D ­ Test instructions and protocols D4.4. Test protocol for participants aged 7 years and older 191

Supplement D ­ Test instructions and protocols 192

Supplement D ­ Test instructions and protocols Legend. VP= participant Übungsitem= familiarization trial Anzahl benötigter Züge= number of moves Anzahl Fehler= Amount of mistaken moves Zeitpunkt 1. Fehler= move withf first mistake Regelverstoß= breach of a rule Latenz= latency Abbruch ja/nein= breakup yes/no Zeit= time needed schwarz= black blau= blue grün= green rot= red gelb= yellow Note. Above, the three different test protocols were depicted: one for 3- to 4-year-olds, one for 5- to 6year-olds, and one for all participants aged 7 years and older(including adults). Each participant started with a familiarization trial("Übungsitem"). On the very left of each table, the number of moves is depicted, which are needed to solve the item correctly, without any additional moves. The next three columns indicate the start position of the discs(i.e. which disc is on which peg at the beginning). In the next column, the number of moves were checked, while participants performed their moves(" Anzahl benötigter Züge") The number of mistakes were captured in the next column("Anzahl Fehler"), and the move, where the first mistake was made, was recorded in the last row("Zeitpunkt 1. Fehler"). Below, an exa mple is shown(test protocol for 5- to 6-year-olds, item 6). Figure 31. Example for an item-protocol of the Tower-of-Hanoi-task. 193

Supplement D ­ Test instructions and protocols D5. Instruction and test protocol for the Mosaic-task For task and procedure for the Mosaic-task, please see Chapter 3.2.2.2.2. In the following, an example is given, according to the test protocol of the HAWIVA-III. The instructions were quite similar for all the tests, only the point in time, when a model or a picture was shown, if a second trials was given or not, and the points for each accomplished pattern differed regarding to the test protocol. D5.1. Instruction Wir spielen nun ein Spiel mit Würfeln. Dafür habe ich rote, weiße und rot-weiße Würfel mitgebracht(zeigen!). Nun lass uns mit den Würfeln spielen. Schau mir zu/ schau dir das Bild an[Modell bauen/ Bild mit zu bauendem Muster aufstellen! Das Kind erhält die genaue Anzahl der benötigten Würfel]. Jetzt mach du das mal. Versuche, das Muster so schnell wie möglich nachzubauen und sage mir, wenn du fertig bist. Hast du noch Fragen? Los geht's!" 194

Supplement D ­ Test instructions and protocols D5.2. Test protocol for 3- to 4-year old children(HAWIVA-III) 195

Supplement D ­ Test instructions and protocols D5.3. Test protocol for 4- to 6--year old children(HAWIVA-III) 196

Supplement D ­ Test instructions and protocols D5.4. Test protocol for 7- to 10-year old children(HAWIK-IV) 197

Supplement D ­ Test instructions and protocols D5.5. Test protocol for adults(WIE) Legend. Muster= pattern Raster= grid Benötigte Würfel= Amount of blocks needed Versuche= trials 198

Supplement D ­ Test instructions and protocols (Präsentations-)Methode(Modell vs. Bild)= method of presentation(model vs. picture) Zeitlimit/ Zeitgrenze= time limit (Lösungs-)Zeit in Sekunden= solution time in seconds Punkte/ erreichte Punktzahl= reached points Rot= red Weiß= white Rohwertsumme= total value of reached points Richtiges Muster= right pattern Nachgebautes Muster= recreated pattern Richtige/ falsche Lösung= right/ wrong solution Note. Above, the four different test protocols were depicted: one for 3- to 4-year-olds, one for 4- to 6year-olds(HAWIVA-III; differ only in the number of start item), one for 7-10-year-olds(HAWIK-IV) and one for adults(WIE). In the first row, the different goal-patterns are depicted. For younger children, the endposture was painted in the boxes beside them. Then, the number of cubes is identified("Benötigte Wü rfel"), and the number of trials, which can be possibly obtained("Versuch"). For younge children, it is declared afterwards, if a model("Modell") or a picture("Bild") serves as the target pattern. The time -limit ("Zeitlimit") indicates the given time, b efore the items is not successfully accomplished. Then, the time needed is recoreded, and points are given depending ont either the time needed, or the successful accomplishment of building the correct pattern. 199

Supplement D ­ Test instructions and protocols D6. Instruction and test protocol for the D2-task For task and procedure for the D2-task, please see Chapter 3.2.2.2.3. D6.1. Instruction and test protocol for 3- to 5-year old children(IDS 3-5) D6.1.1. Instruction " Vor dir siehst du einen Stift auf dem Tisch liegen. Ich gebe dir nun immer nacheinander einen Stapel Karten. Auf diesen Karten siehst du verschiedene Enten. Wir wollen jetzt alle Enten finden, die einen gelben Schnabel haben. Diese Enten legen wir zu diesem Stift. Die Enten mit weißem Schnabel legen wir dahin(links vom Stift). Bei einigen Enten scheint noch die Sonne, aber das ist nicht wichtig. Du musst nur auf den Schnabel schauen. Versuch du jetzt, alle Enten mit einem gelben Schnabel zu finden, und lege sie hier zum Stift.[Eingewöhnungsstapel; hier wird auch noch korrigiert]. Du sollst nun aus den Karten, die ich dir gebe, alle Enten mit einem gelben Schnabel suchen und hier zum Stift legen, und zwar sollst du das so schnell wie möglich machen! Ob die Sonne scheint oder nicht, ist nicht wichtig." Figure 32. Example for correct ducks in the IDS 3-5. All ducks with a yellow snail need to be sorted out, no matter if the sun is shining or not. 200

Supplement D ­ Test instructions and protocols D6.1.2. Test protocol Legend. Erreichte Punktzahl nach 30 Sekunden= reached points after 30 seconds GSK(Gesamtzahl sortierter Karten)= total amount of sorted cards GFA(Gesamtzahl falschlicherweise ausgelassener Karten)= total amount of mistakenly missed cards GFS(Gesamtzahl fälschlicerweise aussortierter Karten)= total amount of mistakenly cards sorted out Wert= value/ amount 201

Supplement D ­ Test instructions and protocols D6.2. Instruction and test protocol for 6- to 10-year old children(IDS 5-10) D6.2.1. Instruction ,,Hier siehst du drei Enten. Jede dieser Enten hat zwei orange Körperteile. Die Er ste Ente hat 2 orange Schnabelhälften, die zweite Ente hat zwei orange Füße und die dritte Ente hat eine orange Schnabelhälfte und einen orangen Fuß ­ zusammengezählt sind das auch zwei orangene Körperteile. Jede Ente, die zwei orange Körperteile hat, sollst du durchstreichen. Mach das jetzt bei diesen drei Enten. Auch in dieser Reihe sollst du alle Enten mit zwei orangen Körperteilen durchstreichen. Pass gut auf, denn alle anderen Enten dürfen nicht durchgestrichen werden. Es gibt auch noch eine zweite Regel! Neben den Enten die auf diese Seite(rechts) sehen, gibt es auch Enten die auf die andere Seite sehen(links). Enten die auf die andere Seite(links) sehen, dürfen nie durchgestrichen werden, egal, wie viele orange Körperteile sie hat. Streich jetzt jede Ente durch, die nach rechts schaut und zwei orange Körperteile hat. Wir schauen nun gemeinsam, ob du auch alle richtigen Enten durchgestrichen hast. Auf der Rückseite befinden sich viele Reihen mit Enten. In jeder Reihe sollst du wieder Enten mit zwei orangen Körperteilen, die nach rechts schauen durchstreichen, wie du das eben getan hast. Nach 15 Sekunden sage ich jeweils ,,Stop! Nächste Reihe!". Dann geh sofort zur nächsten Riehe und streiche dort wieder alle Enten mit 2 orangen Körperteilen durch. Versuch so schnell wie möglich vorwärtszukommen, aber auch keine Fehler zu machen. Falls du eine falsche Ente durchgestrichen hast, mach einfach ein Kreuz und fahre schnell mit der nächsten Ente fort. Hast du noch Fragen? Ich drehe das Blatt nun um. Nimm den Stift in die Hand und beginne hier oben mit der ersten Zeile wenn ich los sage. Achtung, Los!" 202

Supplement D ­ Test instructions and protocols D6.2.2. Test protocol 203

Supplement D ­ Test instructions and protocols 204

Supplement D ­ Test instructions and protocols Legend. BE(bearbeitete Enten)= amount of processed items/ ducks per line F1(Fehler 1)= total amount of mistakenly missed items/ ducks F2(Fehler 2)= total amount of mistakenly items/ ducks sketched out GBE(Gesamtzahl bearbeiteter Enten)= total amount of processed items/ ducks GF(Gesamtzahl Fehler)= total amount of F1 and F2 D6.3. Instruction and test protocol for adults(D2-R) D6.3.1. Instruction ,, Wir wollen mit dem folgenden Versuch feststellen, wie gut du dich auf eine bestimmte Aufgabe konzentrieren kannst. Fülle bitte auf dem Blatt ,,Kurzanleitung" die Fragen zu deiner Person aus. Pass jetzt bitte gut auf! Schaue in die Kurzanleitung. Unter den Fragen zur Person, die du gerade beantwortet hast, wird deine Aufgabe genannt: Du sollst gleich jedes d(wie Dora), das zwei Striche hat, durchstreichen! Im Kästchen siehst du die gesuchten Zeichen. Jeder Buchstabe d hat zwei Striche. Das erste d hat zwei Striche oben, das zweite zwei Striche unten, und das dritte d hat einen Strich oben und einen Strich unten. Rechts daneben siehst du Zeichen, die du nicht durchstreichen sollst: Das kann ein d sein, das weniger oder mehr als zwei Striche hat. Ebenfalls nicht durchstreichen sollst du den Buchstabe p(wie Paula), egal mit wie vielen Strichen es versehen ist. Führe Übung 1 durch. Streiche jedes d mit zwei Strichen durch. Die gesuchten Zeichen sind hier markiert. Führe nun Übung 2 durch. Streiche jedes d mit zwei Strichen durch ­ nun ohne Hilfe. Falls du einmal ein falsches Zeichen durchstreichst, durchkreuze einfach den Strich. 205

Supplement D ­ Test instructions and protocols Lege nun den Stift hin und höre gut zu. Drehe den Testbogen erst um, wenn ich dich dazu auffordere. Auf dem Testbogen befinden sich 14 Zeilen mit den gleichen Zeichen wie in den Übungen. Fange gleich links oben mit der ersten Zeile an. Streiche ­ wie in den Übungen- jedes d mit zwei Strichen durch. Nach 20 Sekunden sage ich ,,Halt! Nächste Zeile!". Dann hörst du sofort aus und fängst mit der nächsten Zeile an. Nach weiteren 20 Sekunden erfolgt wieder der Zuruf ,,Halt! Nächste Zeile!". Beginne dann wieder sofort mit der nächsten Zeile. Arbeite so schnell wie möglich- aber möglichst ohne Fehler. Hast du noch eine Frage? In der Kurzanleitung unten sind noch einmal die wichtigsten Punkte genannt, die du bitte im Test beachtest. Schaue in die Kurzanleitung: Von links nach rechts arbeiten" Bei ,,Halt" Nächste Zeile!" sofort mit der nächsten Zeile anfangen. Schnell und dabei möglichst fehlerfrei arbeiten. Nimm den Stift zur Hand. Drehe das Testblatt um. Fange auf mein Kommando an. Achtung ­ Los! 206

Supplement D ­ Test instructions and protocols D6.3.2. Test protocol 207

Supplement D ­ Test instructions and protocols 208

Supplement D ­ Test instructions and protocols Note. Above, the tree different test protocols for the D2-task were depicted: one for 3- to 5-year-olds (preschoolers, IDS 3-5), one for 5- to 10-year-olds(school children, IDS 5-10), and one for adults(D2R). All participants had to sort out cards, or to cross out items with a specific feature: duck with yellow snails(in the task for the youngest children), ducks with two yellow bodyparts(in the task for the school children), or"d's" with two lines. Evaluation was almost the same: The number of correctly d etected items was counted, as well as mistakes(those items, which were mistakenly not sorted out, and those items, which were sorted out, but did not coincide with the required pattern). In the end, a total score was computed. 209

Statutory Declaration Statutory Declaration I declare that I have developed and written the enclosed thesis entitled" Anticipatory planning in childhood ­ The development of anticipatory planning and ist relationship to executive functions " entirely by myself and have not used sources or means without declaration in the text. Any thoughts or quotations, which were inferred from these sources, are clearly cited as such. This thesis was not submitted in the same or in a substantially similar version, not even partially, to any other authority to achieve an academic grading and was not published elsewhere. In accordance with§9 section 4 of the examination regulations of the Faculty of Natural Science at the University of Paderborn from the 12 th of November 2012, Chapter 2 is based on a publication in the Journal of Motor Learning and Development in 2013. Other contents of the thesis have not been published to this day. Both studies(Chapter 2 and 3) originated in cooperation with other researchers. My contributions comprise the conception of the study designs, the implementation of testing, preparation and evaluation of the data, the documentation and the discussion(and in Chapter 2: the publication) of results. Paderborn, 05.11.2014 Kathrin Wunsch 210

Eidesstattliche Versicherung Eidesstattliche Versicherung Hiermit versichere ich an Eides statt, dass die vorliegende Arbeit mit dem Titel ,, Anticipatory planning in childhood ­ The development of anticipatory planning and ist relationship to executive functions " von mir selbstständig und ausschließlich unter Verwendung der angegebenen Quellen angefertigt wurde. Alle Stellen, die wörtlich oder sinngemäß aus Veröffentlichungen entnommen sind, habe ich als solche kenntlich gemacht. Die Arbeit wurde bisher in gleicher oder ähnlicher Form, auch nicht in Teilen, keiner anderen Prüfungsbehörde vorgelegt. Im Einklang mit§9 Absatz 4 der Promotionsordnung der Fakultät für Naturwissenschaften der Universität Paderborn vom 12. November 2012 basiert Kapitel 2 auf einer Veröffentlichung im Journal of Motor Learning and Development von 2013. Weitere Inhalte der Dissertation wurden bislang nicht veröffentlicht. Beide Teilstudien(Kapitel 2 und 3) entstanden in Kooperation mit weiteren WissenschaftlerInnen. Mein eigener Beitrag in beiden Teilen umfasst sowohl die Konzeption der Studiendesigns, die Durchführung der Studie, die Aufbereitung sowie die Auswertung der Daten sowie die Dokumentation(und in Kapitel 2 die Veröffentlichung) der Ergebnisse. Paderborn, 05.11.2014 Kathrin Wunsch 211

Acknowledgments Acknowledgments A scientific paper is never the contribution of only one person. Therefore, I would like to thank all those people, who supported me on my way to this dissertation. I would like to express my special appreciations and thanks to my advisor Professor Dr. Matthias Weigelt, who always supported and encouraged me and has been a tremendous mentor for me. He always had a friendly ear and gave me new though-provoking impulses. It is due to him, that I became that enthusiastic about scientific research, and that I am going to continue my way into a scientific career. His advice on both, my research and my career, have been priceless. I would also like to thank Professor Dr. Gisa Aschersleben, who kindly agreed to be my second surveyor. She and her expertise on developmental psychology had a great influence on my personal advancement. She and the members of her workgroup, Professor Dr. Anne Henning and Dr. Birgit Knudsen, helped me a lot with developmental issues regarding the studies provided in this thesis. I thank them all very much for the integration in the research group and for the broad support I received. Another special thanks goes to my colleagues at the University of Paderborn, especially my research group, including Dr. Timo Klein-Soetebier, Mustafa Alaboud, Andrè Berwinkel, and Yvonne Steggemann-Weinrich. In numerous conversations they drew my attention to interesting approaches on my research. I am especially grateful to Timo, as he always was there for me, being it for statistical questions or for other topics. I received great countenance of my research students, who were always eager and have been loyal and faithful. Especially I would like to thank Vanessa Heit212

Acknowledgments breder and Jacqueline Sievers, who supported me in data collection, as well as Imke Gerling. Moreover, I would like to thank all of my cooperation partners, who definitely promoted my development in scientific writing. More precisely, I thank Associate Professor Dr. Daniel Weiss for his time and effort he investigated in me and our research paper, and certainly, Gisa Aschersleben, Anne Henning, and Birgit Knudsen, for the great support with our three co-authored publications. Dr. Roland Pfister taught me a lot about statistics, what I greatly appreciate. I thank him for his patience and endurance in analyzing the large data set. Moreover, I thank Dr. Tino Stöckel and Dr. Charmayne Hughes for their professional collaboration on studies and publications to be submitted. They offered me great chances to extend my view on research. Last but not least, a special thanks goes to my family and friends. I can only imagine how proud my father would be in these days. He was a special contributor to my personal development. He supported and encouraged me all the time, even in difficult times. Unfortunately, he encumbered the most difficult time in my life with debts. At this point, I like to thank my mother Martha. We always supported each other, and without her help I would certainly be far away from my thesis at this moment. Moreover I give a great thanks to my boyfriend, Philipp Knapp, who built me up, encouraged me, and sometimes helped me to get my mind off things. He always supported me in every decision and held my arbitrations up . Another big"thank you" goes to his parents, Anton und Luzia, who always made me feel home and gave me (and my mother) the impression of being a part of the family. Moreover, I thank all my friends, who endorsed me to blow the cobwebs away. THANK YOU! 213

Danksagung Danksagung Eine wissenschaftliche Arbeit ist nie das Werk einer einzelnen Person, deshalb möchte ich mich im Folgenden bei allen Menschen bedanken, die mich auf dem Weg zu meiner Dissertation unterstützt haben. Zuerst möchte ich meinem Betreuer und Mentor, Professor Dr. Matthias Weigelt, einen besonderen Dank aussprechen. Er hat mich über die gesamte Dauer meiner Dissertation immer unterstützt und ermutigt. Immer hatte er ein offenes Ohr für mich und gab mir zahlreiche Denkanstöße im Bezug zu meiner Arbeit. Meine Begeisterung für das wissenschaftliche Arbeiten, und die Tatsache, dass ich diesen Weg auch weiterhin verfolgen möchte, habe ich ihm zu verdanken. Seine zahlreichen Ratschläge, seien es bezüglich meiner Arbeit oder meines weiteren Werdeganges, waren unbezahlbar und überaus hilfreich. Weiterhin möchte ich Frau Professor Dr. Gisa Aschersleben danken, die freundlicherweise zugestimmt hat, das Zweitgutachten für diese Arbeit zu übernehmen. Sie hatte mit ihrer Expertise im Bereich der Entwicklungspsychologie, großen Einfluss auf meine persönliche Entwicklung. Sowohl sie, als auch die Mitarbeiterinnen ihres Arbeitsbereiches, Professor Dr. Anne Henning und Dr. Birgit Knudsen, haben mir bezüglich entwicklungspsychologischer Fragestellungen in Bezug auf die in dieser Thesis dargestellten Studien sehr geholfen. Ich danke ihnen allen, dass ich mich immer als Teil der Arbeitsgruppe fühlen durfte und für die große Unterstützung, die ich von ihnen erhielt. Ein weiteres großes Dankeschön möchte ich an meine Kollegen an der Universität Paderborn, besonders die Mitglieder meiner Arbeitsgruppe richten. Hier sind vor allem Timo Klein-Soetebier, Mustafa Alaboud, André Berwinkel, und Yvonne Steggemann-Weinrich zu nennen. In zahlreichen Gesprächen brachten sie mich immer wieder auf neue Ideen und neue Ansätze hinsichtlich meiner Forschung. Ein 214

Danksagung besonderer Dank gilt an dieser Stelle Timo, der mir immer mit Rat und Tat zur Seite stand, egal ob es um statistische Fragen oder andere Dinge ging. Große Unterstützung habe ich von meinen Hilfskräften erfahren, welche mich immer eifrig und zuverlässig unterstützt haben. Mein Dank gilt hier vor Allem Vanessa Heitbreder und Jacqueline Sievers, die mich in der Datenaufnahme unterstützt haben, sowie Imke Gerling. Des Weiteren danke ich all meinen Kooperationspartnern, die mich sicherlich in meiner Fähigkeit zum wissenschaftlichen Schreiben unterstützt und verbessert haben. Genauer gesagt danke ich Associate Professor Dr. Daniel Weiss für die Zeit und die Mühe, die er in mich und unser gemeinsames Projekt investiert hat. Natürlich gilt mein Dank an dieser Stelle auch Gisa Aschersleben, Anne Henning, and Birgit Knudsen für die große Unterstützung und Hilfe bei unseren drei gemeinsamen Veröffentlichungen. Dr. Roland Pfister lehrte mich viele statistische Dinge, wofür ich ihm sehr dankbar bin. Ich danke ihm außerdem für seine Geduld und Ausdauer beim Auswerten der großen Datenmenge. Weiterhin möchte ich Professor Dr. Tino Stöckel und Dr. Charmayne Hughes für ihre professionelle Zusammenarbeit in unterschiedlichen gemeinsamen Studien danken, welche zeitnah publiziert werden sollen. Sie boten mir große Chancen, meine Sicht auf wissenschaftliches Arbeiten zu erweitern. Mein letzter großer Dank geht an meine Familie und Freunde. Leider kann ich nur erahnen, wie stolz mein Vater jetzt gerade auf mich wäre. Er hat immer ein große Rolle in meinem Leben gespielt und hat viel Positives zu meiner persönlichen Entwicklung beigetragen. Er hat mich immer unterstütz und ermutigt, selbst in schwierigen Zeiten. Leider hat nun er selbst die schwierigste Zeit meines Lebens verursacht. An diesem Punkt möchte ich meiner Mutter Martha danken. Wir haben uns immer gegenseitig gestützt und wieder aufgebaut, und ohne ihre Hilfe wäre ich heute sicherlich noch weit von der Abgabe meine Dissertation entfernt. Außerdem möchte ich meinem Freund Philipp Knapp danken, welcher mich in schlechten Zeiten auf215

Danksagung baute, mich ermutigte, und mir manchmal auch half, den Kopf wieder frei zu bekommen. Er stand immer hinter mit und hat jede meiner Entscheidungen mit mir getragen. Ein weiteres großes Dankeschön richte ich an seine Eltern, Anton und Luzia, bei denen ich mich immer wie ,,zu Hause" gefühlt habe, und die mir(und meiner Mutter) immer das Gefühl gaben, ein Teil der Familie zu sein. Zuletzt danke ich noch meinen Freunden, welche mir halfen, meine Gedanken zu zerstreuen. VIELEN DANK! 216