View in EDS HTML Full Text PDF Full Text

Weighing down Thought: The Effect of Motor Disruption in Executive Function Tasks among Three-Year-Old Children

Saved in:
Bibliographic Details
Title: Weighing down Thought: The Effect of Motor Disruption in Executive Function Tasks among Three-Year-Old Children
Language: English
Authors: Z. Reagan Pearce, Stephanie E. Miller
Source: Journal of Cognition and Development. 2025 26(3):398-426.
Availability: Routledge. Available from: Taylor & Francis, Ltd. 530 Walnut Street Suite 850, Philadelphia, PA 19106. Tel: 800-354-1420; Tel: 215-625-8900; Fax: 215-207-0050; Web site: http://www.tandf.co.uk/journals
Peer Reviewed: Y
Page Count: 29
Publication Date: 2025
Document Type: Journal Articles
Reports - Research
Descriptors: Executive Function, Preschool Children, Motor Reactions, Attention, Habit Formation, Accuracy
DOI: 10.1080/15248372.2024.2444970
ISSN: 1524-8372
1532-7647
Abstract: This registered report examines the effect of motor disruption in the conscious control of behavior (i.e. executive function -- EF) of three-year-old children in the Southern United States (N = 114). Specifically, we investigated how disrupting a relevant motor response influenced performance on two EF tasks. Each EF task had an initial phase intended to form a habit, and a switch phase that was the critical measure of EF. Children must inhibit the habit developed at the beginning of the task and reflect consciously on new demands. The multistep multilocation task examined the inhibition of a motor-based habit (response shift), while the Dimensional Change Card Sort examined the inhibition of a cognitive-based habit (attention shift). The motor disruption consisted of administering weighted armbands at two time points during the tasks (i.e. the beginning of the task or during switch trials only) to disrupt reaching behavior. Motor manipulation was found to have a differential effect on EF tasks among three-year-old children. While the armbands improved performance on the response shifting task relative to a no armband control condition, they did not significantly affect performance on the attention shifting task. Contrary to expectations, the armbands did not lead to more errors (habit disruption) during pre-switch trials in either task. These findings highlight the significance of considering both motor and cognitive factors in early childhood EF development. Further research is needed to explore the underlying mechanisms and implications of incorporating movement in EF development.
Abstractor: As Provided
Notes: https://doi.org/10.17605/OSF.IO/2QDSG
Entry Date: 2026
Accession Number: EJ1495028
Database: ERIC
Full text is not displayed to guests.
FullText Links:
  – Type: pdflink
    Url: https://content.ebscohost.com/cds/retrieve?content=AQICAHj0k_4E0hTGH8RJwT4gCJyBsGNe_WN95AvKlDbXJGqwxwF42xZAa4H9W3rI0dgFzsJpAAAA4jCB3wYJKoZIhvcNAQcGoIHRMIHOAgEAMIHIBgkqhkiG9w0BBwEwHgYJYIZIAWUDBAEuMBEEDDxQbGOrh1TZIemIKwIBEICBmrujqj-zYnG_NUhtDV9QTzEUefT8TjaEdSt32jlZbXNj1vI3vxq34UBYtHzuDYKmBzTHloysEZDxgwmCGLHYbE-B5eO5HbpNw6wKpXIVAAwJnKmAlEcgYY3s4_i3-UCWDJo_fUq6Zrrcp4r7HWRtyac7F-jYmdmTWxbbpQj5Y_CmFLsYTEqgOWzi36N-LaONCpMU38T5ff_ZVYc=
Text:
  Availability: 1
  Value: <anid>AN0186129866;7m701may.25;2025Jun26.02:29;v2.2.500</anid> <title id="AN0186129866-1">Weighing Down Thought: The Effect of Motor Disruption in Executive Function Tasks among Three-Year-Old Children </title> <p>This registered report examines the effect of motor disruption in the conscious control of behavior (i.e. executive function – EF) of three-year-old children in the Southern United States (N = 114). Specifically, we investigated how disrupting a relevant motor response influenced performance on two EF tasks. Each EF task had an initial phase intended to form a habit, and a switch phase that was the critical measure of EF. Children must inhibit the habit developed at the beginning of the task and reflect consciously on new demands. The multistep multilocation task examined the inhibition of a motor-based habit (response shift), while the Dimensional Change Card Sort examined the inhibition of a cognitive-based habit (attention shift). The motor disruption consisted of administering weighted armbands at two time points during the tasks (i.e. the beginning of the task or during switch trials only) to disrupt reaching behavior. Motor manipulation was found to have a differential effect on EF tasks among three-year-old children. While the armbands improved performance on the response shifting task relative to a no armband control condition, they did not significantly affect performance on the attention shifting task. Contrary to expectations, the armbands did not lead to more errors (habit disruption) during pre-switch trials in either task. These findings highlight the significance of considering both motor and cognitive factors in early childhood EF development. Further research is needed to explore the underlying mechanisms and implications of incorporating movement in EF development.</p> <p>The term executive function (EF) refers to higher-order cognitive processes essential to conscious control and goal-directed problem solving (P. Zelazo et al., [<reflink idref="bib43" id="ref1">43</reflink>]). EF is a critical piece to cognitive, social, and physical development and has been shown to relate to school readiness (e.g., Blair & Razza, [<reflink idref="bib6" id="ref2">6</reflink>]; Distefano et al., [<reflink idref="bib13" id="ref3">13</reflink>]; Willoughby et al., [<reflink idref="bib42" id="ref4">42</reflink>]), social competence (e.g., Caporaso et al., [<reflink idref="bib8" id="ref5">8</reflink>]), and developmental psychopathologies (e.g., Diamond, [<reflink idref="bib12" id="ref6">12</reflink>]; White et al., [<reflink idref="bib38" id="ref7">38</reflink>]). As such, understanding the early development of EF is crucial and preschool is a period of substantial growth with significant improvements in the control of motor and cognitive responses from 3 to 5 years of age (Wiebe et al., [<reflink idref="bib39" id="ref8">39</reflink>]; P. D. Zelazo & Carlson, [<reflink idref="bib46" id="ref9">46</reflink>]). Although movement is an integral component of assessing this goal-directed behavior (Koziol et al., [<reflink idref="bib18" id="ref10">18</reflink>]), more attention to the interplay between locomotion and cognition in EF literature is needed. Thestudy reported in this registered report observed how motor manipulation impacts EF performance of three-year-olds, as this age is commonly considered a critical time period for major cognitive development (P. D. Zelazo et al., [<reflink idref="bib45" id="ref11">45</reflink>]).</p> <hd id="AN0186129866-2">Representational models of EF perseveration</hd> <p>EF is commonly examined by presenting a problem that is most likely solved through the execution of conscious control. This is usually studied by examining whether behavior is guided by an incorrect but prepotent response or by consciously controlling behavior with a novel response that fits the situation. Repeating habitual responses in situations when it is no longer appropriate is a common EF error, known as perseveration. For example, in the A-not-B task (Piaget, [<reflink idref="bib29" id="ref12">29</reflink>]) infants are required to actively search for a toy in one of two locations (A or B). In the preswitch phase, infants watch as the experimenter hides the toy in location A. They are then prompted to search for the toy. They complete this hide and search event multiple times before switching to the postswitch phase, in which the experimenter begins hiding the toy at location B. At this point, 9-month-olds typically show perseverative behavior and return to location A, despite watching the toy being hidden at location B. Researchers have studied perseverative behavior in A-not-B type tasks as an early measure of EF (e.g., Devine et al., [<reflink idref="bib10" id="ref13">10</reflink>]; Marcovitch & Zelazo, [<reflink idref="bib20" id="ref14">20</reflink>]; Miller & Marcovitch, [<reflink idref="bib25" id="ref15">25</reflink>]), suggesting that the A-not-B tasks requires all components of EF – updating representation of the hidden object to a new location, shifting a response to a new location, and inhibiting the dominant response of search at A.</p> <p>Although there are several models that attempt to explain EF behavior related to perseveration early in development (see Diamond, [<reflink idref="bib12" id="ref16">12</reflink>]; Garon et al., [<reflink idref="bib16" id="ref17">16</reflink>]), the Hierarchical Competing Systems Model (HCSM, Marcovitch & Zelazo, [<reflink idref="bib20" id="ref18">20</reflink>]) is a developmental framework based on early EF and perseverative A-not-B errors that provides a single computational model with utility for understanding perseverative EF behavior across the lifespan. In the HCSM, two systems are computationally specified. The <emph>habit system</emph> [Prob(searchA) =</p> <p>Graph</p> <p> <ephtml> <math xmlns="http://www.w3.org/1998/Math/MathML"><mi mathvariant="italic">α</mi><mo>+</mo><mi mathvariant="italic">lo</mi><mrow><msub><mi>g</mi><mi>β</mi></msub></mrow></math> </ephtml> (numA)] models the probability of search at location A, with</p> <p>Graph</p> <p> <ephtml> <math xmlns="http://www.w3.org/1998/Math/MathML"><mi mathvariant="italic">α</mi></math> </ephtml> as the baseline probability of searching at location A after one trial,</p> <p>Graph</p> <p> <ephtml> <math xmlns="http://www.w3.org/1998/Math/MathML"><mi mathvariant="italic">β</mi></math> </ephtml> as the rate of increase in habit strength, and numA as the number of A trials. This model suggests that habit strength (i.e., probability of searching at the initially correct location A) is positively related to task experience to an extent, as the HCSM proposes a unique, asymptotic function that accelerates negatively until a maximum habit strength is obtained. Thus, children in the A-not-B task strengthen their habit to reach toward hiding location A through repeated successful searches in the preswitch trials until they reach a point of maximum habit strength (e.g., searching for an object at location A 110 times does not make it any more likely that they will execute a habit and search at A as compared to 100 times). The <emph>representational system</emph> [prob (CR) = 1-</p> <p>Graph</p> <p> <ephtml> <math xmlns="http://www.w3.org/1998/Math/MathML"><mi mathvariant="italic">t</mi><mo stretchy="false">(</mo><mrow><mn>1</mn><mo>−</mo><mi mathvariant="italic">γ</mi><mrow><mrow><msup><mo stretchy="false">)</mo><mrow><mi mathvariant="italic">numA</mi></mrow></msup></mrow></mrow></mrow><mo stretchy="false">)</mo><mo stretchy="false">]</mo></math> </ephtml> models the cumulative probability of reflection (e.g., the likelihood that a participant will appropriately reflect on a mental representation cumulates across trials and increases with task experience). At the point at which EF is required (e.g., once the object is moved to location B in the A-not-B task), the process of reflection within the representational system is beneficial because it refers to an individual's ability to actively reprocess relevant representations or mental coding of information linked to memory (i.e., in this case their representation of the object hidden at location B). Specifically, once the object is moved to location B, children can either rely on the developed habit and reach back to location A or rely on conscious representation of the hidden object at the new location and reach toward the correct hiding location (Marcovitch & Zelazo, [<reflink idref="bib20" id="ref19">20</reflink>]). The parameter γ represents the baseline probability of reflection at any given trial, which remains constant within a task. However, as task experience increases, the cumulative probability of reflection over trials increases. Importantly, these baseline probabilities in habit and reflection on mental representations depend on many factors (e.g., the probability of reflection will be higher for children at higher developmental levels), and thus will vary based on the population and changes in the task environment. The habit and representational system combined CombinedProb (searchA_ = [1-Prob(CR)] X [ProbSearchA] can be used to model predicted perseveration on the A-not-B task, suggesting that incorrect search would result from a lack in reflecting on the representation of the stimulus and from the influence of the habit system to search at location A.</p> <p>Importantly, although the HCSM models perseveration based on the A-not-B task, its intent is to provide a foundation for a lifespan developmental model that suggests at the heart of the ability to control behavior (and overcome perseveration) is representational ability (i.e., the ability to mentally describe the environment linked to semantic memory) and reflection (i.e., the ability to reprocess representations to be utilized in working memory). For example, in the A-not-B task, choosing the correct hiding location involves forming and holding a mental representation of the hiding event at the new location to guide search behavior (Marcovitch & Zelazo, [<reflink idref="bib20" id="ref20">20</reflink>]).</p> <hd id="AN0186129866-3">Embodiment models of EF perseveration</hd> <p>Although the HCSM proposes that movement is an important part of perseveration in the A-not-B task (e.g., habit is based on repeated successful searches at location A), embodied approaches to perseveration and EF may place a higher emphasis on movement as a key element of behavior that is not only based in the mind but is based on the interrelation between mind, body, and environment. For example, systems perspectives suggest that motor experience based on repeated and consistent actions is important to goal-directed behavior (e.g., Bertenthal & Barton, [<reflink idref="bib3" id="ref21">3</reflink>], [<reflink idref="bib4" id="ref22">4</reflink>]). Developmentally, actions will strengthen as children repeatedly execute them, which impacts children's ability to coordinate their actions within the physical and social environment and execute goal-directed behavior.</p> <p>For example, Berger ([<reflink idref="bib2" id="ref23">2</reflink>]) demonstrated that when infants were presented with high motor demands in the A-not-B task, they were more likely to perseverate, and the extent of the perseverative errors depended on locomotor experience. The author suggested that, because the goal-directed behavior required to search at location B includes both motor and cognitive demands, increased demands and less experience in motor skills led to a cognition-action trade-off. Specifically, when motor demands were greater children were more likely to make cognitive errors. Clearfield et al. ([<reflink idref="bib9" id="ref24">9</reflink>]) have also demonstrated the importance of considering motor experience in cognitive tasks, with their work showing that younger infants with less motor control are less likely to perseverate in the A-not-B task. Consistent with a systems perspective, they again referred to the importance of considering the role of motor experience within these cognitive tasks, and suggested that those with less motor experience showed less perseveration because these motor deficits made it less likely that infants built stability in their reach toward location A. Thus, without a strong pull toward search at the previously correct location A (based on repeated motor responses), strong inhibition toward A in response to a new hiding event at location B was not needed.</p> <p>Taken together, systems perspectives suggest that consideration of motor experience is critical to understanding goal-directed behavior on EF tasks. Perhaps, the most widely applied systems theory to early childhood EF development is the Dynamic Systems Theory (DST, J. P. Spencer et al., [<reflink idref="bib33" id="ref25">33</reflink>]; Thelen & Smith, [<reflink idref="bib32" id="ref26">32</reflink>]; see also Perone et al., [<reflink idref="bib28" id="ref27">28</reflink>]), which again stresses that movement within the environment is critical for cognition. Thelen and Smith ([<reflink idref="bib36" id="ref28">36</reflink>]) developed this approach suggesting that behavior is guided by multiple interacting systems (e.g., genetic, neural, social, motor) that are inherently flexible and self-organize into habit-forming states that guide behavior. The DST has been extensively applied to the A-not-B task and predicts that children receive multiple sources of input to gradually develop a habit-forming state of searching at location A (Thelen et al., [<reflink idref="bib35" id="ref29">35</reflink>]): <emph>task input</emph>, or environmental changes in the task parameters (i.e., number of hiding locations in the A-not-B task), <emph>specific input</emph>, or the specific cues provided by the experimenter uses to specify the hiding location (i.e., waving the object before hiding it), and <emph>memory input</emph> or the individual's memory of past search actions (i.e., the history of previous responses). Specific to embodiment, the DST suggests that the emergence of perseveration in the A-not-B task is a result of a motor habit to reach toward location A based on repeating the same action multiple times in succession – although this habit may be modified based on task and specific input.</p> <p>Furthermore, the DST (like the HCSM) suggests that perseveration is not limited to infant behavior in the A-not-B task, advocating that perseveration occurs across the lifespan. The DST not only accurately predicts behavior across a multitude of variations of the A-not-B task, including different ages, delays, and reaching experience (see J. R. Spencer et al., [<reflink idref="bib34" id="ref30">34</reflink>]), it has also been applied to more cognitive tasks like the DCCS. For instance, Buss and Spencer ([<reflink idref="bib7" id="ref31">7</reflink>]) extended the basic concepts of the DST with the addition of an autonomous dimensional attention system focused on the attentional nodes (e.g., nodes within the frontal cortex related to the current contents of attention – such as shape or color in the DCCS) that is coupled with cortical fields (e.g., ventral and dorsolateral pathways related to color and shape perception of a target card). The authors suggest a model in which attentional nodes can be differentially linked to other cortical fields depending on experience. For example, the neural interaction between these nodes for younger children may be poorly organized, thus activation of the color node may boost baseline activation in the color field and weaken activation in the shape field. For older children with more organized connectivity, the activation of the color node should have little impact on the shape field. Empirical work has shown promising results testing the predictions of a DST model to a more attentionally-based shifting task in the DCCS (Buss & Spencer, [<reflink idref="bib7" id="ref32">7</reflink>]), although the role of motor experience and disruption within a task like the DCCS has not yet been extensively examined from this perspective.</p> <hd id="AN0186129866-4">Motor influences on perseveration</hd> <p>Interestingly, both the theoretical viewpoints of representation and embodied systems would likely hypothesize that modifications to motor behavior during the formation of the motor habit have the potential to alter responses on an EF task. For example, from an HCSM perspective, Marcovitch et al. ([<reflink idref="bib21" id="ref33">21</reflink>]) hypothesized that changing the number of trials in the preswitch phase would affect the formation of the habit (i.e., more A trials would strengthen the habit to an asymptotic point) and likelihood of reflection during the postswitch phase (i.e., more A trials or experience during the task would increase the likelihood of reflecting on the representation of the hiding object). In support of this prediction, the authors found an inverted U-shaped pattern of perseveration where 9-month-old infants were more likely to show perseveration after six preswitch trials as opposed to one or eleven preswitch trials, likely because 6 A trials were the point where the A trials reached maximum habit formation but the further experience gained with 11 A trials made it more likely that infants would reflect on their mental representation of the object's location as they had extensive experience with the task (i.e, related to the cumulative probability of reflection, Marcovitch et al., [<reflink idref="bib21" id="ref34">21</reflink>]).</p> <p>Despite the growing support for movement having a function in EF, few studies have extended the work by inhibiting or modifying movement in EF tasks. A study conducted by Rivière and Lécuyer ([<reflink idref="bib31" id="ref35">31</reflink>]) manipulated movement in a C-not-B task measuring EF in 2-year-old toddlers. This task was like the A-not-B task except that it was a three-location search task and children did not encounter a habit-forming preswitch phase. In this task, children watched as the experimenter's closed fist moved underneath 3 cloth locations from location A to location B, and finally to location C. The experimenter discreetly placed the toy at location B and left a clue (i.e., lump underneath the hiding location), suggesting that the object was hidden at location B. Children displayed a perseverative error by searching for the object in the last place the experimenter's hand was seen (i.e., location C) despite the visual clue representing the object's true hiding place. Rivière and Lécuyer hypothesized that failure in the C-not-B task resulted from children's inability to inhibit the motor response (i.e., searching at the last place they saw the experimenter's hand).</p> <p>To investigate the role of motor skills in EF, Rivière and Lécuyer tested children on the C-not-B task under three experimental conditions: (<reflink idref="bib1" id="ref36">1</reflink>) a <emph>weighted armband</emph> condition, in which children wore a 200 gram weight on each arm when performing the task; (<reflink idref="bib2" id="ref37">2</reflink>) a visual (<emph>unweighted</emph>) condition, where children wore weightless colored bracelets on each arm when performing the task; and (<reflink idref="bib3" id="ref38">3</reflink>) a free (<emph>no armband</emph>) condition, in which children performed the C-not-B task without manipulation. In this study, Rivière and Lécuyer observed an increase in success rates in the children with arm weights. Conversely, search performance was not influenced on the free and visual conditions, suggesting that improvement on the C-not-B task did not occur because of the increased attention to arm movements (Rivière & Lécuyer, [<reflink idref="bib31" id="ref39">31</reflink>]). Arterberry et al. ([<reflink idref="bib1" id="ref40">1</reflink>]) further supported this finding by extending the weighted armband manipulation to a younger sample and different search task. The authors found that 24-month-olds performed significantly better on the C-not-B task and the Door task, in which children had to predict the final location of a ball rolling down a ramp (Berthier et al., [<reflink idref="bib5" id="ref41">5</reflink>]). The results of these experiments encourage further investigation of the role of movement in perseverative errors in EF tasks.</p> <p>Theoretically, it is important to note that both the DST and HCSM would likely predict that a motor disruption during the habit-forming phase of a motor-based EF task should lead to a weakened habit (i.e., reaching toward the A location), thus increasing the likelihood of a non-perseverative response to location B. However, the HCSM links motor behavior not only to a weakened habit but also to the opportunity to reflect (e.g., more task experience increases the likelihood of reflection, Marcovitch & Zelazo, [<reflink idref="bib20" id="ref42">20</reflink>]). The addition of this reflective system suggests that motor behavior likely influences behavior in multiple ways. For example, a motor disruption during the postswitch phase of the A-not-B task may also influence performance by changing the motor response and drawing additional attentional resources – and possibly reflection – to the goal-directed movement.</p> <p>Further, it is likely that motor manipulations may encourage reflection in cognitive based tasks as well. For example, although the HCSM was based on the response switching A-not-B task, its basis in reflection on mental representations is such that it may easily be applied to an attention shifting task (i.e., tasks that require not a switch in a motor-based response – from location A to location B – but a switch in attention – like switching a sorting rule from sorting by shape to sorting by color). For example, the Dimensional Change Card Sort task requires children to sort a picture with two different dimensions (e.g., yellow flower) to two potential target pictures (e.g., green flower or yellow car). Children sort by one dimension first (e.g., in the color game the yellow flower would go with the yellow car). After multiple trials, the sorting rule will switch to the new dimension (e.g., in the shape game the yellow flower would go with the green flower). From a representational model of perseveration, introducing a motor disruption in post switch trials of an attention shifting task may encourage conscious thought (specifically reflection on task relevant information) because it may change the latency and trajectory of the reach. This change has the potential to draw attention to the goal-directed motor behavior executed in their response and allow time for reflection on appropriate mental representations. The integration of movement in both habit and reflection could lead to new ideas on how movement impacts perseveration and EF more broadly.</p> <hd id="AN0186129866-5">Present study</hd> <p>The purpose of the present study was to examine how a motor manipulation consisting of weighted armbands on children's wrists affects performance on a response shifting and attention shifting task. To answer this question, 3-year-old children performed two EF tasks, an age-appropriate variant of the A-not-B task – the multistep multilocation search task (response shift, MSML, Zelazo et al., [<reflink idref="bib47" id="ref43">47</reflink>]) and the Dimensional Change Card Sort (attention shift, DCCS, P. D. Zelazo, [<reflink idref="bib44" id="ref44">44</reflink>]). This age group was chosen because of the transition from using movement to language as a communication tool and the substantial development of EF at this age. Each task is comprised of a preswitch phase and a postswitch phase. The purpose of the preswitch phase is to form a habit. To pass each task, participants must perform correctly in the postswitch phase, by opposing the habit that was formed during the preswitch phase.</p> <p>The study employed a 2 × 3 factorial design manipulating the level of motor manipulation and the time point of motor manipulation. The motor manipulation variable consisted of three levels (<emph>weighted armband, unweighted armband, no armband</emph>), while the point of administration variable consisted of two levels (armband administered during <emph>full task</emph>, armband administered during <emph>postswitch trials only</emph>), resulting in six conditions (see Tables 1 and 2).</p> <p>Table 1. Summary of conditions.</p> <p> <ephtml> <table><thead><tr><td>Motor Manipulation</td><td /><td /></tr></thead><tbody><tr><td><italic>weighted armband</italic></td><td>determines whether disrupting motor movement impacts task performance</td></tr><tr><td><italic>unweighted armband</italic></td><td>determines whether weighted armband effects could be due to drawing attention to the arm by placing a band on it</td></tr><tr><td><italic>no armband</italic></td><td>baseline control comparison (i.e., typical task administration)</td></tr><tr><td>Point of Administration</td><td /><td /></tr><tr><td><italic>Full Task</italic></td><td>determines whether disrupting motor movement during preswitch and postswitch impacts performance differently across tasks</td></tr><tr><td><italic>Postswitch Only</italic></td><td>determines whether disrupting motor movement during postswitch only impacts performance differently across tasks</td></tr></tbody></table> </ephtml> </p> <p>1 This table provides a description of the purpose of each of the six experimental conditions.</p> <p>Table 2. Overview of 3 (motor manipulation) x 2 (point of administration) design and conditions.</p> <p> <ephtml> <table><thead><tr><td /><td>Full Task</td><td>Postswitch Only</td></tr></thead><tbody><tr><td><italic>weighted armband</italic></td><td>weighted armbands worn on pre and postswitch trials</td><td>weighted armbands worn on postswitch trials only</td></tr><tr><td><italic>unweighted armband</italic></td><td>unweighted armbands worn on pre and postswitch trials</td><td>unweighted armbands worn on postswitch trials only</td></tr><tr><td><italic>no armband</italic></td><td>no armbands worn on pre and postswitch trials*</td><td>no armbands worn on postswitch trials only</td></tr></tbody></table> </ephtml> </p> <ulist> <item>2 This table provides a description of each of the six experimental conditions.</item> <item>3 *For all conditions, children hold out their arm for approximately 10 seconds between the preswitch and postswitch trials to equate the delay and movement between pre and postswitch. Although we do not believe that this very brief delay would be the reason for any performance differences, the no armband full task control is run without any additional delay between pre and postswitch to examine this assumption.</item> </ulist> <p>Utilizing the HCSM framework, we hypothesized that the weighted armbands would differentially affect performance in the motor-based response shifting task (MSML) and cognitively based attention shifting task (DCCS). In the MSML, weighted armbands are expected to improve performance regardless of the time point at which the motor manipulation is administered (either <emph>full task</emph> or <emph>postswitch only</emph>). Weighted armbands administered for the full duration of the task (i.e., applied before preswitch trials and worn throughout the whole task) will likely weaken the formation of the habit formed, making it easier to use representation and reflection to guide behavior once the hiding location is switched to location B. Administering weighted armbands during the postswitch phase of the task should alter the reaching action, which may heighten awareness and slow down responses on postswitch trials, encouraging reflection on the representational system. In the cognitively based attention shifting task, the DCCS, weighted armbands are expected to improve performance when the motor manipulation is administered at <emph>postswitch only</emph>. When weights are applied throughout the whole task, the cognitive habit will be developed regardless of weight condition (e.g., a habit of sorting by shape will be formed similarly regardless of the reach trajectory), resulting in a similar habit formed across all 3 weight conditions. However, similar to the MSML, a noticeable change in action will occur when weights are applied during the postswitch phase, which should encourage cognitive reflection when weights are applied in the postswitch trials only.</p> <p>It is important to acknowledge that we did not include a <emph>preswitch only</emph> level within our point of administration manipulation. We chose to examine the effect of preswitch motor disruption by having children wear armbands during the full task so that the change in the task requirements is not emphasized (e.g., a preswitch only condition would highlight the change in hiding location, see Miller & Marcovitch, [<reflink idref="bib23" id="ref45">23</reflink>]; Rivière & Lécuyer, [<reflink idref="bib31" id="ref46">31</reflink>]). Although it is possible that improvements in performance could be due to wearing weights during the postswitch trials regardless of the preswitch trial experience, we did not include a <emph>preswitch only</emph> level because we are able to draw on task comparisons to examine this possibility (i.e., if effects were solely due to wearing weights during the postswitch we would not see different effects for the MSML and the DCCS). Further, we also aimed to control the delay between preswitch and postswitch, as previous literature has noted that the introduction of a delay between the two phases of a perseverative task can alter task performance (Diamond, [<reflink idref="bib11" id="ref47">11</reflink>]). Although we did not expect a brief delay to impact performance, to equate the delay across conditions, we included an equal delay at the midpoint (pre- to postswitch) for both tasks and for all conditions with the exception of our full task no armband control condition. The <emph>full task, no armband</emph> control condition was included without a delay to examine its impact, see Table 2.</p> <hd id="AN0186129866-6">Method</hd> <p>The methodology and planned analyses for the present study were preregistered on the Open Science Framework (OSF) and can be accessed, along with the raw data and Stage 1 manuscript, at https://osf.io/enbt9/?view_only=6a6f402f7ee34062a7a03c488446920a.</p> <hd id="AN0186129866-7">Participants</hd> <p>An <emph>a priori</emph> power analysis using Fixed Effects ANOVA with main effects and interactions was conducted through G*Power software to determine that 84 participants were required to provide sufficient power for this study. An estimated medium to large effect size (η<sups>2</sups> = 0.40) was used in the analysis. The estimation was based on the large effect of motor inhibition (η<sups>2</sups> = 0.5) found by Rivière and Lécuyer ([<reflink idref="bib31" id="ref48">31</reflink>]). Additionally, a p-value of 0.05 and power of 0.90 were used. The final sample included data from 114 children between 36 and 47 months (<emph>M</emph><subs>ag<emph>e</emph></subs> = 43.38, <emph>SD =</emph> 3.44, 46 females) to meet our pre-registered power analysis of 84 children for each task (see the section titled "missing data based on preregistered exclusion criteria" of the results for details). Recruitment occurred through a preexisting database of parents interested in having their child participate in research on cognitive development, as well as through child-care facilities and preschools located in a southern city of the United States. The race of the participants' parents who did report (96.5%) was approximately 87% White, 10% Asian, and 3% African American for mothers; and 86% White, 8% Asian, 4% African American, 2% Hispanic, and 1% more than one race for fathers.</p> <hd id="AN0186129866-8">Exclusion criteria</hd> <p>Children were replaced in the dataset if they were unable to pass either the training or preswitch phase of either task (i.e., the MSML or DCCS). If children only passed the preswitch of one task, their postswitch data was included for the task that they did pass, and they had missing data for the postswitch trials of the task that they did not pass. Based on previous work, we anticipated that the proportion of children unable to pass the preswitch to be low, with a maximum percentage of approximately 10% (9 children that need to be replaced, Pearce et al., [<reflink idref="bib27" id="ref49">27</reflink>]; P. D. Zelazo, [<reflink idref="bib44" id="ref50">44</reflink>]). If there were technical issues with the equipment (e.g., a computer failure at any point in the task), children were also replaced in the data set. The section titled "missing data based on preregistered exclusion criteria" in the results fully describes all missing and replaced data.</p> <hd id="AN0186129866-9">Materials and procedure</hd> <p>Upon receiving parental permission, participants completed two computer-based EF tasks (counterbalanced) administered individually by a graduate student or trained undergraduate research assistant in an on-campus laboratory or quiet area of a preschool or childcare center. Two computerized EF tasks were utilized for this study. The MSML (response shift) is an age-appropriate version of the A-not-B task. Previous studies have suggested that a 20 second delay and a five-part multistep sequence are necessary to yield sufficient variability in responses (Marcovitch & Zelazo, [<reflink idref="bib19" id="ref51">19</reflink>]; Miller & Marcovitch, [<reflink idref="bib23" id="ref52">23</reflink>]). The standard DCCS (P. D. Zelazo, [<reflink idref="bib44" id="ref53">44</reflink>]) was computerized and used to measure attentional shift. These tasks were programmed using SuperLab Pro software. Tasks were presented on a Microsoft Surface Pro tablet (Versions 3 and 4)[<reflink idref="bib1" id="ref54">1</reflink>] with a 12-inch touch screen display. Children were seated directly in front of the computer and saw the stimuli presented in full screen view on a white background. Children were given a small toy for participating in the study.</p> <p>This 2 × 3 factorial design based on the level of motor manipulation (<emph>no armband, unweighted armband, weighted armband</emph>) and time point of administration (<emph>full task, postswitch only</emph>) resulted in six conditions.[<reflink idref="bib2" id="ref55">2</reflink>] Importantly, children completed both EF tasks in their assigned condition (i.e., both the MSML and DCCS in counterbalanced order). There were 3 levels to motor manipulation. In the <emph>no armband</emph> conditions, children performed the two tasks without wristbands. In the <emph>unweighted armband</emph> conditions, children performed the tasks with unweighted black cotton wristbands on each wrist (weight 0.00044 lbs.) In the <emph>weighted armband</emph> conditions, children performed the tasks with black weighted wristbands (weight 0.5 lbs) on each wrist. Wristband weight was determined based on previous work conducted by Rivière and Lécuyer ([<reflink idref="bib31" id="ref56">31</reflink>]) with 2-year-olds and adjusted using the US Center for Disease Control's growth charts to better match 3-year-olds. The <emph>unweighted armband</emph> and <emph>no armband</emph> conditions served as control conditions. The <emph>unweighted armband</emph> conditions were necessary to eliminate the possibility of change in response due to heightened awareness toward arm movements (Rivière & Lécuyer, [<reflink idref="bib31" id="ref57">31</reflink>]). The <emph>no armband</emph> conditions were utilized as a comparison for the <emph>unweighted armband</emph> conditions.</p> <p>There were 2 levels to the time point of administration. The administration refers to the time point during each task in which the motor manipulation is administered. The <emph>full task</emph> conditions were intended to determine if motor manipulation changed the development of habit formation during the preswitch phase of each task. The <emph>postswitch only</emph> conditions were intended to determine if motor manipulation affects the cognitive reflection process. In this condition, children were able to form a strong habit during the preswitch phase without any motor disruption, eliminating the possibility that the change in response was due to a change in habit formation. The point of administration did not apply to the <emph>no armband</emph> conditions; however, the two conditions differ by the administration of a delay between preswitch and postswitch trials of each task, see Table 2.</p> <hd id="AN0186129866-10">Multistep multilocation task</hd> <p></p> <hd id="AN0186129866-11">Training</hd> <p>During training in the MSML, children were introduced to the purpose and procedure of the task – specifically, how to search for a star. The procedure was introduced in a backward step manner so that children begin with the simplest response first (searching for the star) and build up to the full procedure (see Figure 1). This phase trained children to search for the star with one single gray box centrally located on the screen and children were trained on each element of the search procedure. In the final phase of training, children independently searched for the star in a trial that consisted of the complete multistep sequence where children watched as the lid of the gray box opened, a yellow star entered the box, and the lid closed. After the initial hiding event, a gray wall appeared on the screen for a 20 second delay before turning red, prompting the participant to lower the two walls by pressing the green buttons and perform a multistep sequence of removing blocks in order (red, yellow, green) before the box was revealed and children searched for the star by pressing the box on the screen.</p> <p>Graph: Figure 1. The multistep procedure (Miller & Marcovitch, [<reflink idref="bib23" id="ref58">23</reflink>]; Miller et al., [<reflink idref="bib26" id="ref59">26</reflink>]).</p> <hd id="AN0186129866-12">Testing-Preswitch trials</hd> <p>Once children passed training, they moved to the preswitch trials. In the full task, no armband condition, children immediately moved onto the preswitch trials without a pause. In the full task, weighted condition children were all asked to hold their hands out in front of them and put the weighted armbands on both wrists. In the full task, unweighted condition, children were asked to hold their hands out in front of them and put the unweighted armbands on both wrists. In all three postswitch only conditions, children were asked to extend their arms out in front of them, either to put on armbands or to wait for approximately 10 seconds, before performing the post-switch phase.[<reflink idref="bib3" id="ref60">3</reflink>] After the experimental manipulation, preswitch trials were like the training phase except that five gray boxes were presented on the screen during the hiding event so that the star hid in one of multiple hiding locations, see Figure 1(a). The correct hiding locations during preswitch and postswitch were counterbalanced. The middle box was not used as a hiding location to minimize interference from training trials.</p> <p>Children watched the star enter the box at the preswitch location. The experimenter pointed where the star was located and said, "The star is hiding in this box. You will find the star right here." The participant then experienced the 20 second delay, during which children were encouraged to count out loud with the experimenter. Once the delay was complete, the wall changed color and the participant performed the multistep sequence to find the star. The final block of the multistep sequence covered the hiding locations, so once the green block was pressed, the hiding locations (5 gray boxes) were revealed to the participant. For correct searches, an appetitive sound chimed, and children saw the star come out of the box. If children were incorrect, an aversive sound played, and they saw a red X come out of the incorrect boxes. Participants completed six preswitch trials. The number of errors made during the preswitch phase was measured. Children were required to perform four out of the six preswitch trials correctly to be considered to have formed a habit. This passing criterion was based on previous work showing that most 3-year-olds (approximately 90%) are able to complete an identical task with 2 or fewer errors (Pearce et al., [<reflink idref="bib27" id="ref61">27</reflink>]). Children who did not pass the preswitch phase of the task did not move onto the postswitch phase (this is like the preswitch criteria in the DCCS, P. D. Zelazo, [<reflink idref="bib44" id="ref62">44</reflink>]).[<reflink idref="bib4" id="ref63">4</reflink>]</p> <hd id="AN0186129866-13">Testing-Postswitch trials</hd> <p>After six searches at the initial location A, children moved onto the postswitch trials (i.e., see the star enters a new hiding location). Children in the full task administration conditions remained in their assigned conditions (i.e., <emph>no armband, unweighted armband, weighted armband</emph>). To establish a similar procedure and delay for all conditions, children in the full task, armband conditions extended both arms for approximately 10 seconds before transitioning onto postswitch. The <emph>full task, no armband</emph> condition served as a control and had no delay between preswitch and postswitch conditions. For the <emph>postswitch only, no armband</emph> condition, children were asked to extend both arms directly after the completion of the preswitch phase for approximately 10 seconds before moving onto postswitch. For the <emph>postswitch only, weighted</emph> condition, children were asked to put the weighted armbands on both wrists and wear the armbands for the duration of the postswitch phase. For the <emph>postswitch only, unweighted</emph> condition children were asked to put the unweighted armbands on both wrists and wear the armbands for the duration of the postswitch phase. The purpose of the arm extension was to produce a similar pause that was elicited in the <emph>unweighted</emph> and <emph>weighted armband</emph> conditions. The <emph>no armband</emph> conditions ensured that the changes in task performance are due to the change in movement rather than due to having a delay before the postswitch phase. Children performed six postswitch trials and accuracy on the first postswitch trial and performance across all 6 postswitch trials was measured. The experimenter helped the children in armband conditions remove armbands before moving to the next task. Armbands were not worn during the transition between the MSML or DCCS tasks.</p> <hd id="AN0186129866-14">Dimensional change card sort</hd> <p></p> <hd id="AN0186129866-15">Training</hd> <p>In the training phase, participants were shown a screen with a green flower and yellow car on the bottom of the screen (left and right location counterbalanced). The experimenter labeled each target picture and presented children with a rule for sorting pictures (i.e., yellow flowers and green cars) to the target picture (i.e., sort by color or shape, counterbalanced). The experimenter demonstrated the task by sorting one picture by the rule (e.g., saying, "We are playing the color game, if it's green, it goes here; but if it's yellow, it goes there. Here's a green one. Where does it go?"). The correct response was to touch the correct target on the screen. The experimenter repeated the rules and then asked children to do the next practice card. Children were praised if correct and corrected and asked to sort the card again if incorrect (P. D. Zelazo, [<reflink idref="bib44" id="ref64">44</reflink>]). All participants performed the training phase without the motor manipulation condition.</p> <hd id="AN0186129866-16">Testing-Preswitch trials</hd> <p>The motor and point of administration manipulations for the DCCS were administered in the same manner as they were in the MSML. Following the administration of the armbands, the experimenter prompted the participant by reminding them of the rule (e.g., "Now it is your turn. Remember, if it is green, it goes here, but if it is yellow it goes here"). Children were then presented with 6 trials for sorting (e.g., "Here is a green one, where does it go?") and no feedback was given during preswitch trials. Children were required to be correct on four of the six preswitch trials to move onto the postswitch phase.[<reflink idref="bib5" id="ref65">5</reflink>] Children who did not pass preswitch did not move onto postswitch trials.[<reflink idref="bib6" id="ref66">6</reflink>] The number of errors made during the preswitch phase was measured (P. D. Zelazo, [<reflink idref="bib44" id="ref67">44</reflink>]).</p> <hd id="AN0186129866-17">Testing-Postswitch trials</hd> <p>Children in the full task administration conditions remained in their assigned conditions (i.e., no armband, unweighted armband, weighted armband) and the delay between pre and postswitch trials and administration of the armbands were identical to the MSML. After the administration of the armbands, the experimenter introduced the new sorting rule (e.g., say, "Now we are going to play a new game, we are going to play the shape game. In the shape game, all the flowers go here, and all of the cars go there"). The correct response was to touch the correct target on the screen. The target pictures remained in the same location on the screen. Once the test card appeared, the experimenter labeled it by the dimension specified by the game (e.g., shape game) and asked the participant where it went. Children still performed six postswitch trials and accuracy on the first postswitch trial and performance across all 6 postswitch trials were measured.</p> <hd id="AN0186129866-18">Post DCCS matching task</hd> <p>Finally, to examine potential mechanisms for how motor manipulation affected performance (e.g., through shifting attention to task relevant representations or assisting in the formation of higher-order rule structures), children completed a matching task after the DCCS was complete. In this task, children were presented with the hierarchical rule structure of the task (see Figure 2a) and asked to match the sorting card (Figure 2b) to the target card according to the hierarchical rules. For example, for the leftmost yellow flower, the experimenter would say "OK, if we are playing the color game, the yellow flower (pointing to the yellow flower) goes with which card?" Children were asked to place their selection for the sorting cards in the slot underneath the yellow flower. Total correct out of 4 was measured.</p> <p>Graph: Figure 2a. Sorting board for DCCS post-task.</p> <p>Graph: Figure 2b. Sorting cards for DCCS post-task.</p> <hd id="AN0186129866-19">Methodological note</hd> <p>The above tasks were similar in that they both consisted of preswitch trials that involved forming a habit (i.e., searching or sorting behavior) that later conflicted with new information presented in the postswitch trials. However, it is also important to note the differences among the MSML and DCCS tasks. First and foremost, performance on the MSML depends on the formation of a motor habit (i.e., response shift), while performance on the DCCS depends more heavily on a cognitive habit (i.e., attention shift). Moreover, the MSML includes feedback for correct searching. The DCCS, on the other hand, only provides feedback during the training phase of the task. In an effort to follow the original protocol of the DCCS, feedback was not included in the pre or postswitch phases of the task. With the MSML task being a search task, feedback on the correct hiding location was essential to maintaining search behavior (i.e., children must see if the object is there when they search), therefore feedback remained part of the task.</p> <p>Despite these differences, there is preliminary evidence that these tasks are age appropriate and can elicit perseveration among three-year-olds. Miller and Marcovitch ([<reflink idref="bib23" id="ref68">23</reflink>]) utilized a similar computerized version of the MSML (i.e., 10 second delay, removal of three blocks) with older 2-year-olds and found an approximate error rate of 45% on the first postswitch trial. With older children (i.e., 2.5 to 4-years-old) a delay and multistep procedure identical to the one proposed in the present study revealed an approximate error rate of 48% on the first postswitch trial after children were required to complete 6 correct preswitch trials (Pearce et al., [<reflink idref="bib27" id="ref69">27</reflink>]). The DCCS (P. D. Zelazo, [<reflink idref="bib44" id="ref70">44</reflink>]) has been extensively administered to 3-year-olds and there is usually a substantial number of errors on the postswitch trials at this age.</p> <hd id="AN0186129866-20">Data analysis plan</hd> <p>Analyses were conducted separately for the MSML and DCCS, as the impact of motor manipulation is hypothesized to operate differently for a response-shifting task (i.e., MSML) compared to an attention-shifting task (i.e., DCCS). For each dependent measure of interest, a 2 (point of administration) x 3 (motor manipulation) factorial omnibus test (i.e., ANOVA for continuous data and logistic regression for dichotomous data) was conducted. Bonferroni and Tukey's LSD post-hoc comparisons were examined to further probe any statistically significant main effects and interactions. Further, a priori contrasts and Bayesian t-tests were performed for any hypothesized differences and Bayesian t-tests were performed for hypothesized null effects registered before data collection to supplement traditional frequentist approaches and examine evidence for the null and alternative hypotheses.</p> <hd id="AN0186129866-21">Results</hd> <p></p> <hd id="AN0186129866-22">Missing data based on preregistered exclusion criteria</hd> <p>In accordance with the preregistered methodology, data collection continued until we reached the number of participants estimated in our power analysis (i.e., 84 participants or roughly 14 participants in each experimental condition for each task). To meet this criteria we worked with a total of 119 children. Four children were excluded due to being outside of the specified age range at the time of testing and 1 child refused to participate in the study, resulting in a final sample of 114 children (demographics reported in participants section).</p> <p>Of the 114 children, 3 children refused to wear weighted armbands on both tasks, 4 children took the weighted armbands off after the first task and refused to put them back on, and 3 children refused to wear the unweighted armbands on both tasks. We did not specify in the preregistered methodology a procedure in cases of child refusal. During testing, we chose to have children who refused the armbands to complete the task(s) without the armband (e.g., full no armband condition). However, we removed these cases in analyses given that potential systematic correlates of refusal (e.g., being more impulsive) could nullify random assignment to conditions. This resulted in a final sample of 104. It is important to note that children performed slightly worse on preswitch trials than expected in the preregistered methodology as 17% failed training on the DCCS and 21% failed training on the MSML, which did not differ by task, <emph>McNemar χ</emph><sups><emph>2</emph></sups> =.37, <emph>p = 0.54.</emph></p> <p>The preregistered methodology specified that children would be replaced in the dataset if they were unable to pass the training or preswitch phase of either task (i.e., the MSML or DCCS) and that we would consider partial data in the analyses (i.e., complete data from one task if they failed the other). Therefore, 27 children were excluded due to failure to complete the MSML task (e.g., <emph>n</emph> = 5 failed to complete preswitch trials, <emph>n</emph> = 21 incorrectly searched 3 or more times during preswitch trials, and <emph>n</emph> = 1 failed to complete postswitch trials), resulting in 77 children who passed MSML preswitch trials with complete data on the pre and postswitch trials of the MSML. For analyses examining the effect of weighted armbands and point of administration on the DCCS task, 18 children were excluded due to failure to complete the task (e.g., <emph>n</emph> = 3 failed training, <emph>n</emph> = 2 failed to complete preswitch trials, <emph>n</emph> = 13 incorrectly searched 3 or more times during preswitch trials). The remaining 86 children had complete data for preswitch and postswitch trials of the DCCS. Of the 86 children who had complete data for the DCCS, 84 of them completed the post DCCS questions, as 2 children did not complete the post DCCS (<emph>n</emph> = 1 failure to present the task to the child or <emph>n</emph> = 1 child refusal). Results suggested better preswitch accuracy in the DCCS (<emph>M</emph> = 5.21, <emph>SD</emph> = 1.47) compared to the MSML (<emph>M</emph> = 4.55, <emph>SD</emph> = 1.44), <emph>t</emph>(<reflink idref="bib93" id="ref71">93</reflink>) <emph>=</emph> 3.88, <emph>p</emph> < 0.001 and a reverse pattern for postswitch accuracy across 6 trials with better performance in the MSML (<emph>M</emph> = 4.70, <emph>SD</emph> = 1.59) compared to the DCCS (<emph>M</emph> = 1.56, <emph>SD</emph> = 2.23), <emph>t</emph>(<reflink idref="bib82" id="ref72">82</reflink>) <emph>=</emph> 10.37, <emph>p</emph> < 0.001.</p> <p>The number of children in each condition varied (see Tables 3 and 4 for descriptive statistics and a breakdown by condition for children who passed preswitch trials). Although we did not specify an examination of response time latency in stage one preregistered methodology, we conducted exploratory analyses to examine whether weighted armbands slowed down responding – which may be interpreted as a type of manipulation check for the weighted armbands. For both tasks, of the children that had latency data after outliers (more than 3 SD from the mean) were excluded, we found no evidence that weighted or unweighted armbands slowed response times compared to no armbands on preswitch trials, see Supplementary Material.</p> <p>Table 3. Descriptive statistics for preswitch errors by task and condition for children who passed preswitch.</p> <p> <ephtml> <table><thead><tr><td>Effects of Motor Manipulation on Habit Formation</td></tr><tr><td>Point of Administration</td><td>Motor Manipulation</td><td>M <italic>(SE)</italic></td><td>95% CI</td><td>Range</td><td>n</td></tr></thead><tbody><tr><td>MSML Response Shift</td></tr><tr><td><italic>Full Task</italic></td><td><italic>No Armband</italic></td><td>.82 <italic>(.18)</italic></td><td>[.41, 1.22]</td><td>0–2</td><td>11</td></tr><tr><td /><td><italic>Unweighted</italic></td><td>.86 <italic>(.23)</italic></td><td>[.36, 1.36]</td><td>0–2</td><td>14</td></tr><tr><td /><td><italic>Weighted</italic></td><td>.62 <italic>(.21)</italic></td><td>[.15, 1.08]</td><td>0–2</td><td>13</td></tr><tr><td><italic>Postswitch Only</italic></td><td><italic>No Armband</italic></td><td>1.00 <italic>(.18)</italic></td><td>[.61, 1.39]</td><td>0–2</td><td>14</td></tr><tr><td /><td><italic>Unweighted</italic></td><td>.69 <italic>(.21)</italic></td><td>[.24,1.15]</td><td>0–2</td><td>13</td></tr><tr><td /><td><italic>Weighted</italic></td><td>1.00 <italic>(.21)</italic></td><td>[.53,1.47]</td><td>0–2</td><td>12</td></tr><tr><td>DCCS Attention Shift</td></tr><tr><td><italic>Full Task</italic></td><td><italic>No Armband</italic></td><td>.21 <italic>(.11)</italic></td><td>[−.0346]</td><td>0–1</td><td>14</td></tr><tr><td /><td><italic>Unweighted</italic></td><td>.12 <italic>(.08)</italic></td><td>[−.0529]</td><td>0–1</td><td>17</td></tr><tr><td /><td><italic>Weighted</italic></td><td>.31 <italic>(.18)</italic></td><td>[−.0769]</td><td>0–2</td><td>13</td></tr><tr><td><italic>Postswitch Only</italic></td><td><italic>No Armband</italic></td><td>.21 <italic>(.16)</italic></td><td>[−.1255]</td><td>0–2</td><td>14</td></tr><tr><td /><td><italic>Unweighted</italic></td><td>.29 <italic>(.13)</italic></td><td>[.02,.56]</td><td>0–1</td><td>14</td></tr><tr><td /><td><italic>Weighted</italic></td><td>.57 <italic>(.23)</italic></td><td>[.08,-1.06]</td><td>0–2</td><td>14</td></tr></tbody></table> </ephtml> </p> <p>Table 4. Descriptive statistics for postswitch accuracy by task and condition.</p> <p> <ephtml> <table><thead><tr><td>Effects of Motor Manipulation on Accuracy</td></tr><tr><td>Point of Administration</td><td>Motor Manipulation</td><td>M <italic>(SE)</italic></td><td>95% CI</td><td>Range</td><td>n</td></tr></thead><tbody><tr><td>MSML Response Shift (passing first postswitch trial)</td></tr><tr><td><italic>Full Task</italic></td><td><italic>No Armband</italic></td><td>.64 <italic>(.15)</italic></td><td>[.30,.98]</td><td>0–1</td><td>11</td></tr><tr><td /><td><italic>Unweighted</italic></td><td>.93 <italic>(.07)</italic></td><td>[.77, 1.08]</td><td>0–1</td><td>14</td></tr><tr><td /><td><italic>Weighted</italic></td><td>.77 <italic>(.12)</italic></td><td>[.50, 1.03]</td><td>0–1</td><td>13</td></tr><tr><td><italic>Postswitch Only</italic></td><td><italic>No Armband</italic></td><td>.57 <italic>(.14)</italic></td><td>[.27, 87]</td><td>0–1</td><td>14</td></tr><tr><td /><td><italic>Unweighted</italic></td><td>.85 <italic>(.10)</italic></td><td>[.62, 1.07]</td><td>0–1</td><td>13</td></tr><tr><td /><td><italic>Weighted</italic></td><td>.92 <italic>(.08)</italic></td><td>[.73, 1.10]</td><td>0–1</td><td>12</td></tr><tr><td>DCCS Attention Shift (passing 5 of 6 postswitch trials)</td></tr><tr><td><italic>Full Task</italic></td><td><italic>No Armband</italic></td><td>.43 <italic>(.14)</italic></td><td>[.13,.73]</td><td>0–1</td><td>14</td></tr><tr><td /><td><italic>Unweighted</italic></td><td>.00 <italic>(.00)</italic></td><td>[−.00,.00]</td><td>0–1</td><td>17</td></tr><tr><td /><td><italic>Weighted</italic></td><td>.15 <italic>(.10)</italic></td><td>[−.07–.38]</td><td>0–1</td><td>13</td></tr><tr><td><italic>Postswitch Only</italic></td><td><italic>No Armband</italic></td><td>.14 <italic>(.10)</italic></td><td>[−.07,.35]</td><td>0–1</td><td>14</td></tr><tr><td /><td><italic>Unweighted</italic></td><td>.08 <italic>(.08)</italic></td><td>[−.09,.24]</td><td>0–1</td><td>13</td></tr><tr><td /><td><italic>Weighted</italic></td><td>.20 <italic>(.11)</italic></td><td>[−.03,.43]</td><td>0–1</td><td>15</td></tr></tbody></table> </ephtml> </p> <p>Our main analyses only examined children who passed the preswitch trials despite the limited range of errors (i.e., children could only make 2 errors in preswitch to pass) for several reasons. First, this allowed us to examine the effect of motor disruption on habit formation and accuracy in the same sample of children (i.e., those who pass the preswitch trials). Second, as previously noted, our replacement rule led to unequal sample sizes, leading some conditions to have more children who failed training (and thus higher preswitch errors). Table 5 depicts the preswitch errors made for all children (regardless of whether they passed training) and we provide footnotes detailing how analyses of preswitch errors compared between the two samples.</p> <p>Table 5. Descriptive statistics for preswitch errors by task and condition for all children, regardless of whether they passed preswitch trials.</p> <p> <ephtml> <table><thead><tr><td>Effects of Motor Manipulation on Habit Formation</td></tr><tr><td>Point of Administration</td><td>Motor Manipulation</td><td>M <italic>(SE)</italic></td><td>95% CI</td><td>Range</td><td>n</td></tr></thead><tbody><tr><td>MSML Response Shift</td></tr><tr><td><italic>Full Task</italic></td><td><italic>No Armband</italic></td><td>1.47 <italic>(.32)</italic></td><td>[.78, 2.16]</td><td>0–4</td><td>15</td></tr><tr><td /><td><italic>Unweighted</italic></td><td>2.00 <italic>(.49)</italic></td><td>[.98, 3.02]</td><td>0–6</td><td>20</td></tr><tr><td /><td><italic>Weighted</italic></td><td>1.07 <italic>(.37)</italic></td><td>[.27, 1.86]</td><td>0–5</td><td>15</td></tr><tr><td><italic>Postswitch Only</italic></td><td><italic>No Armband</italic></td><td>1.56 <italic>(.32)</italic></td><td>[.89, −2.22]</td><td>0–5</td><td>18</td></tr><tr><td /><td><italic>Unweighted</italic></td><td>.93 <italic>(.31)</italic></td><td>[.27, −1.59]</td><td>0–4</td><td>14</td></tr><tr><td /><td><italic>Weighted</italic></td><td>1.71 <italic>(.32)</italic></td><td>[1.03–2.38]</td><td>0–4</td><td>17</td></tr><tr><td>Children who completed preswitch trials (pass or fail) 99</td></tr><tr><td>DCCS Attention Shift</td></tr><tr><td><italic>Full Task</italic></td><td><italic>No Armband</italic></td><td>.40 <italic>(.21)</italic></td><td>[−.06–.86]</td><td>0–3</td><td>15</td></tr><tr><td /><td><italic>Unweighted</italic></td><td>.75 <italic>(.37)</italic></td><td>[−.02–1.52]</td><td>0–6</td><td>20</td></tr><tr><td /><td><italic>Weighted</italic></td><td>1.13 <italic>(.47)</italic></td><td>[.12–2.13]</td><td>0–6</td><td>16</td></tr><tr><td><italic>Postswitch Only</italic></td><td><italic>No Armband</italic></td><td>.63 <italic>(.32)</italic></td><td>[−.05–1.30]</td><td>0–4</td><td>16</td></tr><tr><td /><td><italic>Unweighted</italic></td><td>.60 <italic>(.34)</italic></td><td>[−.12–1.32]</td><td>0–5</td><td>15</td></tr><tr><td /><td><italic>Weighted</italic></td><td>1.24 <italic>(.41)</italic></td><td>[.37–2.10]</td><td>0–5</td><td>17</td></tr><tr><td>Children who completed preswitch trials (pass or fail) 103</td></tr></tbody></table> </ephtml> </p> <hd id="AN0186129866-23">Effect of motor manipulation on habit formation</hd> <p></p> <hd id="AN0186129866-24">Multistep multilocation response switch task</hd> <p>A 2 (administration) x 3 (motor manipulation) between-subject analysis of variance was conducted on the number of errors made in the preswitch phase to examine habit formation, means and standard deviations presented in Table 3. Results did not support the hypothesis that motor manipulation would influence the formation of a motor habit, hypothesized to be evidenced by an interaction between motor manipulation and administration. There was no significant main effect of motor manipulation, <emph>F</emph>(<reflink idref="bib2" id="ref73">2</reflink>, 71) =.23, <emph>p =</emph>.80, <emph>partial η</emph><sups><emph>2</emph></sups><subs><emph>p</emph></subs><emph>=</emph>.01, or point of administration, <emph>F</emph>(<reflink idref="bib1" id="ref74">1</reflink>, 71) =.62, <emph>p = </emph>.43, <emph>partial η</emph><sups><emph>2</emph></sups><subs><emph>p</emph></subs><emph>=</emph>.01. The hypothesized interaction was also not significant, <emph>F</emph>(<reflink idref="bib2" id="ref75">2</reflink>, 71) =.91, <emph>p =</emph>.41, <emph>partial η</emph><sups><emph>2</emph></sups><subs><emph>p</emph></subs><emph>=</emph>.03. Thus, we did not find the hypothesized significant interaction that could indicate the motor manipulation disrupted habit formation when administered during preswitch trials (i.e., the full duration of the task, see Figure 3).</p> <p>Graph: Figure 3. Average number of preswitch errors made on MSML by condition.</p> <p>We also conducted planned contrasts and Bayesian t-tests to examine the a priori hypothesized difference that children in the <emph>weighted armband</emph> condition would exhibit the most errors made during the preswitch phase indicating a habit disruption, but only when the armbands were administered during preswitch trials (i.e., for the <emph>full task)</emph>. First, we conducted a planned contrast to compare the conditions where children wore <emph>weighted armbands</emph> during preswitch trials (<emph>n</emph> = 13) to children who did not wear any armbands during the preswitch trials (<emph>n</emph> = 50, collapsed across the <emph>full no armband, post no armband, post unweighted armband, post weighted armband conditions)</emph>. This contrast was not significant, <emph>F</emph>(<reflink idref="bib1" id="ref76">1</reflink>, 71) = 1.28, <emph>p =</emph>.26, <emph>partial η</emph><sups><emph>2</emph></sups><subs><emph>p</emph></subs><emph>=</emph>.02. A Bayesian independent samples t-test using unequal variance of the same comparison revealed a Bayes Factor<subs>01</subs> =.47, suggesting that the data provided anecdotal evidence for the null hypothesis (i.e., no difference in preswitch errors between those who wore weighted armbands and those who did not).</p> <p>Next, we conducted a planned comparison of the number of errors made by children who wore <emph>weighted armbands</emph> during preswitch trials (<emph>n</emph> = 13) to the children who wore <emph>unweighted armbands</emph> during the preswitch trials (<emph>n</emph> = 14). This contrast was also not significant, <emph>F</emph>(<reflink idref="bib1" id="ref77">1</reflink>, 71) =.71, <emph>p =</emph>.40, <emph>partial η</emph><sups><emph>2</emph></sups><subs><emph>p</emph></subs><emph>=</emph>.01. A Bayesian independent samples t-test using unequal variance of the same comparison revealed a Bayes Factor<subs>01</subs> =.36, suggesting that the data provided anecdotal evidence for the null hypothesis (i.e., no difference in preswitch errors between those who wore weighted armbands and those who wore unweighted armbands).[<reflink idref="bib7" id="ref78">7</reflink>]</p> <p>A final exploratory contrast compared the number of errors made by children who wore <emph>unweighted armbands</emph> during preswitch trials (<emph>n</emph> = 14) to the children who did not wear armbands during the preswitch trials (<emph>n</emph> = 50). This contrast was also not significant, <emph>F</emph>(<reflink idref="bib1" id="ref79">1</reflink>, 71) =.01, <emph>p =</emph>.93 <emph>partial η</emph><sups><emph>2</emph></sups><subs><emph>p</emph></subs><emph><</emph>.001. A Bayesian independent samples t-test using unequal variance of the same comparison revealed a Bayes Factor<subs>01</subs> =.49, suggesting that the data provided anecdotal evidence for the null hypothesis (i.e., no difference in preswitch errors between those who wore unweighted armbands and those who did not wear any armbands during preswitch trials).</p> <hd id="AN0186129866-25">Dimensional change card sort attention shifting task</hd> <p>A 2 (administration) x 3 (motor manipulation) between-subjects analysis of variance was conducted on the number of errors made in the preswitch phase of the DCCS. To be consistent with the notion that motor disruption will not affect the formation of a cognitive habit, we hypothesized that children would make few errors with no statistically significant difference on the DCCS, regardless of condition. Consistent with our hypothesis, this analysis[<reflink idref="bib8" id="ref80">8</reflink>] revealed no significant main effect of motor manipulation, <emph>F</emph>(<reflink idref="bib2" id="ref81">2</reflink>, 80) = 1.56, <emph>p =</emph>.22, <emph>partial η</emph><sups><emph>2</emph></sups><subs><emph>p</emph></subs><emph><</emph>.04, or point of administration, <emph>F</emph>(<reflink idref="bib1" id="ref82">1</reflink>, 80) = 1.39, <emph>p =</emph>.24, <emph>partial η</emph><sups><emph>2</emph></sups><subs><emph>p</emph></subs><emph>=</emph>.02. The interaction was also not significant, <emph>F</emph>(<reflink idref="bib2" id="ref83">2</reflink>, 80) =.39, <emph>p =</emph>.68, <emph>partial η</emph><sups><emph>2</emph></sups><subs><emph>p</emph></subs><emph>=</emph>.01, (see Figure 4).</p> <p>Graph: Figure 4. Average number of preswitch errors made on DCCS by condition.</p> <p>We also conducted two Bayesian independent t-tests using unequal variance to further examine the hypothesized null effect of motor manipulation on habit formation in the DCCS. First, we compared the conditions where children wore weighted armbands during preswitch trials (<emph>n</emph> = 15) to children who did not wear any armbands during the preswitch trials (<emph>n</emph> = 56, collapsed across the <emph>full no armband, post no armband, post unweighted armband, post weighted armband conditions</emph>). The Bayesian independent samples t-test revealed a Bayes Factor<subs>01</subs> =.23, suggesting that the data provided moderate evidence for the null hypothesis (i.e., no difference in preswitch errors between those who wore weighted armbands and those who did not).</p> <p>The second Bayesian independent samples t-tests using unequal variance compared the number of errors made by children who wore <emph>weighted armbands</emph> during preswitch trials (<emph>n</emph> = 13) to the children who wore <emph>unweighted armbands</emph> during the preswitch trials (<emph>n</emph> = 17). This t-test revealed a Bayes Factor<subs>01</subs> =.43, further suggesting that the data provided anecdotal evidence for the null hypothesis that there was no difference between children who wore <emph>weighted</emph> and <emph>unweighted armbands</emph> on preswitch trial errors.[<reflink idref="bib9" id="ref84">9</reflink>]</p> <hd id="AN0186129866-26">Effect of motor manipulation on task accuracy</hd> <p></p> <hd id="AN0186129866-27">Multistep multilocation response shifting task</hd> <p>Accuracy on the postswitch trials was also measured. A 2 (administration) x 3 (motor manipulation) between-subjects logistic regression was conducted on the accuracy for the first postswitch trial. To support the hypothesis that weighted armbands would improve task performance regardless of administration in the MSML, a major effect of motor disruption was expected. However, there no significant main effect of motor manipulation, <emph>Wald</emph> χ<sups>2</sups> (<reflink idref="bib2" id="ref85">2</reflink>) = 2.74, <emph>p</emph> = 0.10, point of administration, <emph>Wald</emph> χ<sups>2</sups> (<reflink idref="bib1" id="ref86">1</reflink>) =.11, <emph>p</emph> = 0.74 or interaction between motor manipulation and point of manipulation, <emph>Wald</emph> χ<sups>2</sups> (<reflink idref="bib2" id="ref87">2</reflink>) = 1.48, <emph>p</emph> = 0.48.</p> <p>We did, however, find some support for predicted differences in task accuracy in our a priori planned contrasts, which indicated a marginally significant effect demonstrating that children in the <emph>weighted</emph> conditions (<emph>n</emph> = 25) outperformed children in the <emph>no armband</emph> conditions (<emph>n</emph> = 25), χ<sups>2</sups>(<reflink idref="bib1" id="ref88">1</reflink>) = 3.57, <emph>p</emph> = 0.06, see Figure 5. We hypothesized this would suggest that the motor manipulation could improve performance on a response shifting task by weakening the automatic response to reach toward the previous location (i.e., when administered during preswitch and postswitch trials) and encouraging cognitive reflection (i.e., when administered during postswitch trials only). A Bayesian estimation using chi-square tests in SPSS was also conducted with the same comparison and revealed a Bayes Factor<subs>01</subs> of 2.52 indicating anecdotal evidence for the alternative hypothesis (i.e., that children who wore <emph>weighted armbands</emph> had higher accuracy on the first postswitch trial of the DCCS compared to children who did not wear armbands).</p> <p>Graph: Figure 5. Accuracy on DCCS postswitch by condition.</p> <p>A priori planned contrasts using chi-square tests were also conducted to examine whether those in the <emph>weighted armband</emph> conditions would outperform the <emph>unweighted armband</emph> conditions on the first B postswitch trial accuracy. Surprisingly, children in the <emph>weighted armband</emph> conditions (<emph>n</emph> = 25) did not outperform children in the <emph>unweighted armband</emph> conditions (<emph>n</emph> = 27), χ<sups>2</sups>(<reflink idref="bib1" id="ref89">1</reflink>) =.27, <emph>p</emph> = 0.67 (Figure 5). This result indicated that children with unweighted armbands did not perform differently from those with weighted armbands on the first B postswith trial, a finding that was also evident by the fact that children in the <emph>unweighted armband</emph> conditions outperformed children in the <emph>no armband</emph> conditions (<emph>n</emph> = 25), χ<sups>2</sups>(<reflink idref="bib1" id="ref90">1</reflink>) = 5.78, <emph>p</emph> = 0.02. Bayesian estimations using chi-square tests also aligned with these findings and suggest anecdotal evidence for the null hypothesis in contrast 1 (i.e, no difference between children who wore weighted armbands and unweighted armbands for accuracy on the first postswitch trial of the MSML, Bayes Factor<subs>01</subs> =.39) and moderate evidence for the alternative hypothesis (i.e, that children who wore <emph>unweighted armbands</emph> outperformed those who wore <emph>no armbands</emph> on the first postswitch trial of the MSML, Bayes Factor<subs>01</subs> = 7.09).</p> <hd id="AN0186129866-28">Dimensional change card sorting attention shifting task</hd> <p>In the DCCS accuracy of sorting pictures during the postswitch phase is typically measured as a dichotomous outcome (i.e., passing if participants are correct on five of the six postswitch trials), because of the bimodal distribution of responses. Therefore, a 2 (administration) x 3 (motor manipulation) between-subjects logistic regression was conducted on DCCS sorting behavior (i.e., pass or fail). There was no support for the hypothesis that an interaction would be found, such that the <emph>postswitch weighted armband</emph> condition would be significantly different from the other conditions. The main effect of motor manipulation was not significant, <emph>Wald</emph> χ<sups>2</sups>(<reflink idref="bib2" id="ref91">2</reflink>) = 2.28, <emph>p</emph> = 0.32. There were no significant effects related to point of administration, <emph>Wald</emph> χ<sups><emph>2</emph></sups>(<reflink idref="bib1" id="ref92">1</reflink>) = 2.59, <emph>p</emph> = 0.11, or a significant interaction between motor manipulation and point of administration, <emph>Wald</emph> χ<sups>2</sups>(<reflink idref="bib2" id="ref93">2</reflink>) = 1.93, <emph>p</emph> = 0.38, failing to support the hypothesis that <emph>weighted armbands</emph> would improve DCCS performance when administered for postswitch trials only, see Figure 5.</p> <p>An a priori planned contrast was performed using chi-square tests to examine our original hypothesis that the postswitch weighted armband condition (<emph>n</emph> = 15) would outperform all other conditions (<emph>n</emph> = 71) on the DCCS. This comparison was not significant, χ<sups>2</sups> (<reflink idref="bib1" id="ref94">1</reflink>) =.19, <emph>p</emph> = 0.67, suggesting that children in the <emph>postswitch weighted condition</emph> did not perform differently than in the other conditions. A Bayesian estimation using chi-square tests with the same comparison revealed a Bayes Factor<subs>01</subs> of.24 indicating anecdotal support for the null hypothesis that there was no difference on the passing behavior on postswitch trial accuracy when comparing the <emph>postswitch weighted condition</emph> to all other conditions.</p> <hd id="AN0186129866-29">Effect of motor manipulation on higher order rule structure</hd> <p>Finally, to better understand whether the motor manipulation impacts performance by aiding in the creation of higher-order rule structures, we conducted a 2(administration) x 3(motor manipulation) ANOVA on the number correct on the post-DCCS sorting task. Although exploratory, it was hypothesized that an interaction would be found, such that the <emph>postswitch weighted armband</emph> condition would be significantly different from the other conditions. Our hypotheses were not supported. The results of this analysis indicated no significant main effect of motor manipulation, <emph>F</emph>(<reflink idref="bib2" id="ref95">2</reflink>, 78) =.41, <emph>p =</emph>.29, <emph>partial η</emph><sups><emph>2</emph></sups><subs><emph>p</emph></subs> =.01, or point of administration, <emph>F</emph>(<reflink idref="bib1" id="ref96">1</reflink>, 78) = 1.35, <emph>p =</emph>.25. <emph>partial η</emph><sups><emph>2</emph></sups><subs><emph>p</emph></subs><emph>=</emph>.02. The interaction was also not significant, <emph>F</emph>(<reflink idref="bib2" id="ref97">2</reflink>, 78) = 1.04, <emph>p =</emph>.36, <emph>partial η</emph><sups><emph>2</emph></sups><subs><emph>p</emph></subs><emph>=</emph>.03. Thus, we were unable to support the exploratory hypothesis that time provided for reflection and a shift in attention due to motor disruption on postswitch trials may relate to the formation of a higher order rule structure in the DCCS, see Table 6 and Figure 6).</p> <p>Graph: Figure 6. Accuracy on post DCCS by condition.</p> <p>Table 6. Descriptive statistics for post DCCS accuracy by task and condition.</p> <p> <ephtml> <table><thead><tr><td>Effect of Motor Manipulation on Higher Order Rule Structure</td></tr><tr><td>Point of Administration</td><td>Motor Manipulation</td><td>M <italic>(SE)</italic></td><td>95% CI</td><td>Range</td><td>n</td></tr></thead><tbody><tr><td><italic>Full Task</italic></td><td><italic>No Armband</italic></td><td>1.43 <italic>(.36)</italic></td><td>[.65–2.20]</td><td>0–4</td><td>14</td></tr><tr><td /><td><italic>Unweighted</italic></td><td>1.19 <italic>(.23)</italic></td><td>[.70–1.67]</td><td>0–2</td><td>16</td></tr><tr><td /><td><italic>Weighted</italic></td><td>1.77 <italic>(.28)</italic></td><td>[1.16–2.38]</td><td>0–4</td><td>13</td></tr><tr><td><italic>Postswitch Only</italic></td><td><italic>No Armband</italic></td><td>1.57 <italic>(.31)</italic></td><td>[.90–2.24]</td><td>0–4</td><td>14</td></tr><tr><td /><td><italic>Unweighted</italic></td><td>1.83 <italic>(.37)</italic></td><td>[1.03–2.64]</td><td>0–4</td><td>12</td></tr><tr><td /><td><italic>Weighted</italic></td><td>1.87 <italic>(.35)</italic></td><td>[1.12–2.62]</td><td>0–4</td><td>15</td></tr></tbody></table> </ephtml> </p> <p>An a priori planned contrast was also performed to examine our original hypothesis that the <emph>postswitch weighted armband</emph> condition (<emph>n</emph> = 14) would outperform all other conditions (<emph>n</emph> = 69) on the DCCS. This comparison was not significant, <emph>F</emph>(<reflink idref="bib1" id="ref98">1</reflink>, 78) =.89, <emph>p =</emph>.49, <emph>partial η</emph><sups><emph>2</emph></sups><subs><emph>p</emph></subs><emph>=</emph>.06, suggesting that children in the <emph>postswitch weighted</emph> condition did not perform differently than the other conditions. A Bayesian estimation using chi-square tests with the same comparison revealed a Bayes Factor<subs>01</subs> of.33 indicating anecdotal support for the null hypothesis that there was no difference on the first postswitch trial accuracy when comparing the <emph>postswitch weighted</emph> condition to all other conditions.</p> <hd id="AN0186129866-30">Discussion</hd> <p>The present study examined the impact of motor disruption on EF in three-year-old children, specifically focusing on how disrupting a relevant motor response affects performance on two EF tasks. We hypothesized that even though both tasks measured EF, we would find a differential effect of motor disruption (related to both <emph>if</emph> and <emph>when</emph> a motor manipulation was administered) because of the role that movement plays in each task. Specifically, the MSML response shift task requires children to resist physically searching at a previously correct location (or shift motor response to a new location). The DCCS attention shifting task requires children to mentally resist sorting cards by a previously correct rule (or shift attention to a new rule or sorting dimension). We found no substantial evidence indicating that motor disruption led to more errors (i.e., habit disruption) during pre-switch trials in either the response (MSML) or attention shifting (DCCS) tasks. For performance on the critical EF trials, we did find a differential effect of motor manipulation by task. In the MSML response shifting task, <emph>weighted armbands</emph> improved performance relative to the <emph>no armband control condition</emph> as hypothesized, but not relative to the <emph>unweighted armband control</emph> condition. For the DCCS attention shifting task, we did not find any differences in postswitch performance, or the higher-order rule task related to if and when weighted armbands were administered on postswitch performance or on the higher-order rule structure task, which we hypothesized as a potential mechanism for improved EF. Taken together, results suggest that movement may play a nuanced role in how young children perform on different types of EF tasks.</p> <hd id="AN0186129866-31">The effect of motor disruption on response shifting EF task</hd> <p>Our examination of motor disruption on the response shifting MSML task is both a replication and extension of past work. Others have looked at how changing motor trajectories through weighted armbands impact perseverative performance (e.g., Arterberry et al., [<reflink idref="bib1" id="ref99">1</reflink>], 2020; Rivière & Lécuyer, [<reflink idref="bib31" id="ref100">31</reflink>]), however, this is the first study to the authors' knowledge that examines both the disruption to habit (i.e., preswitch errors) and impact on EF perseveration (i.e., accuracy on the first postswitch trial). We hypothesized that weighted armbands would influence both habit formation (preswitch errors; e.g., as likely predicted in the HCSM, Marcovitch & Zelazo, [<reflink idref="bib20" id="ref101">20</reflink>], and DST; Thelen & Smith,) and EF performance (postswitch accuracy; e.g., as predicted by the HCSM possibly drawing more attentional resources to the goal directed movement, Marcovitch & Zelazo, [<reflink idref="bib20" id="ref102">20</reflink>]). Our results partially supported these hypotheses. We found that <emph>weighted armbands</emph> improved performance on postswitch trials compared to <emph>no armband</emph> conditions but did not outperform the <emph>unweighted armband</emph> condition. There are several possibilities for these results.</p> <p>Similar to Rivière and Lécuyer ([<reflink idref="bib31" id="ref103">31</reflink>]) we found that wearing a weighted armband resulted in better accuracy in a response shifting EF task. Based on the HCSM, we originally suggested this finding would be due to 2 factors. First, children in the <emph>weighted armband, full task</emph> condition likely performed better on the postswitch EF trials because they formed a weaker habit to reach toward location A due to motor disruption on these trials. This is in line with other studies guided by both the HCSM and DST suggesting motor experience influences the strength of the habit that children must resist on postswitch trials (Berger, [<reflink idref="bib2" id="ref104">2</reflink>]; Clearfield et al., [<reflink idref="bib9" id="ref105">9</reflink>]; Marcovitch & Zelazo, [<reflink idref="bib19" id="ref106">19</reflink>]; Marcovitch et al., [<reflink idref="bib21" id="ref107">21</reflink>]). If the motor habit is weaker due to disruption of the reach toward location A, the strength of the habit in opposition to the possibility of reflection would be weaker, resulting in better postswitch performance (Marcovitch & Zelazo, [<reflink idref="bib20" id="ref108">20</reflink>]). There are two problems with this interpretation. First, we do not see an increase in errors or response time related to a motor habit disruption on preswitch trial errors. Second, children in the <emph>weighted armband</emph> condition do not perform significantly better than children in the <emph>unweighted armband</emph> condition. If improved performance was truly due to the motoric disruption leading to a weaker habit formed during preswitch trials, we would expect children in the <emph>weighted armband</emph> condition to outperform those in the <emph>unweighted armband</emph> condition, who have a wristband but no weight to disrupt movement. It is possible that the unweighted armband also disrupts habit in some way but given the fact that there is no difference in preswitch trial accuracy or response time in either condition, we propose another mechanism.</p> <p>We found that children in both <emph>weighted armband</emph> conditions (<emph>full administration</emph> and <emph>postswitch trials only</emph>) performed better in EF postswitch trials compared to those in the <emph>no armband</emph> condition. Although we initially hypothesized they would perform better for different reasons (i.e., <emph>weighted armbands</emph> during preswitch trials would disrupt habits, <emph>weighted armbands</emph> during postswitch trials would increase reflection), it is possible that both conditions had improved performance due to increased reflection on the critical postswitch trials. For example, during preswitch trials children can rely on reflection or habit to obtain the hidden object (Marcovitch & Zelazo, [<reflink idref="bib20" id="ref109">20</reflink>]). It is only on the critical postswitch trials where habit and reflection conflict and children would be better served if they relied on reflecting on their mental representation of where the object is hidden. Thus, children in both <emph>weighted armband</emph> conditions might have increased reflection during preswitch trials and postswitch trials. If this is true, this increase in reflection would only be evident in postswitch trial performance when the motor habit conflicts with the potential for reflection.</p> <p>This proposal might also help to explain why we saw a difference between the <emph>weighted armband</emph> condition when compared to the <emph>no armband</emph> condition, but not the <emph>unweighted armband</emph> control. If the increase in reflection is driven by drawing attention to goal directed movement, this may be achieved through changes in movement (as hypothesized for the <emph>weighted armband</emph> condition). But it may also be driven by changes in how children perceive and interact within the environment (e.g., Miller & Marcovitch, [<reflink idref="bib24" id="ref110">24</reflink>]). Questions related to reflection and cognizance within the context of motor actions has also been examined in classic works, in which Piaget ([<reflink idref="bib30" id="ref111">30</reflink>]/1976) demonstrated that young children (i.e., 4 years of age) can reflect on potentially more automatic sensorimotor actions when adjustments are required – though they are not always accurate and are developing in their reflections (e.g., 4-year-olds asked to on the action of crawling reflect on the fact that there is movement, but incorrectly report the order or integration of arm and leg movement in crawling).</p> <p>While we originally hypothesized that an unweighted armband would not be enough to encourage reflection because it did not change the trajectory of the movement, we did not consider that drawing any additional attention to one's body (even without changes to movement trajectory) might encourage an increase in conscious thought and reflection on the goal directed action of searching for a hidden object in the present study (e.g., Piaget, [<reflink idref="bib30" id="ref112">30</reflink>]; [<reflink idref="bib30" id="ref113">30</reflink>]). This may also help explain why we found a pattern of results different from Rivière and Lécuyer ([<reflink idref="bib31" id="ref114">31</reflink>]), who demonstrated that weighted armbands improved performance in the C-not-B compared to both no armband and unweighted armband control conditions. Their explanation of results was purely motoric and not based on increased attention – they stated that having additional weights may prevent the execution of a direct visual-motor response (i.e., searching for an object at the last place it was seen). Increased attention in their task may not have been helpful in this specific motor routine that was more established (i.e., not built within the context of the task like the MSML). More work is needed to better understand why results differed between tasks, but increased reflection within both <emph>weighted armband</emph> and <emph>unweighted armband</emph> conditions may help possibly explain current findings.</p> <hd id="AN0186129866-32">The effect of motor disruption on attention shifting EF tasks</hd> <p>To our knowledge, the examination of the effect that motor disruption has on the habit formation (i.e., preswitch errors) and EF performance (i.e., postswitch trial passing) in an attention shifting DCCS task was a novel exploration related to how motor manipulation impacts cognition. We predicted that motor manipulation would affect performance in the DCCS differently as compared to the MSML. With regard to habit formation, we hypothesized that the <emph>weighted armband</emph> condition would not result in any changes on preswitch performance relative to the <emph>no armband</emph> and <emph>unweighted armband</emph> controls because in the DCCS the habit that is formed in the preswitch trials is a cognitive habit (i.e., a habit of sorting by shape should develop regardless of reach trajectory). Although weighted armbands were not predicted to affect cognitive habit, we did hypothesize an effect of motor manipulation on postswitch performance in the DCCS. Specifically, similar to the MSML, only weighted armbands administered during postswitch trials should encourage cognitive reflection through heightened awareness and slowed responses. Our results partially supported these hypotheses. When examining habit formation on preswitch trials, we did not see evidence for differences related to when and if weighted armbands were administered, as hypothesized. However, contrary to our hypotheses, we did not find any evidence for <emph>weighted armbands</emph> administered during postswitch improving EF performance on postswitch trials.</p> <p>These results are surprising, given the significant <emph>weighted</emph> and <emph>unweighted armband</emph> effects on the MSML, which we suggested was due to increased reflection (i.e., both weighted and unweighted draw attention to goal-directed movement). It is possible that these results reflect a differential effect, where manipulations to the body (i.e., both through changing motor trajectories and increased attention to one's own body) affect performance on a motoric set-shifting EF task but not a cognitive attention-shifting EF task. If this is true, it could lead to interesting considerations for the HCSM (Marcovitch & Zelazo, [<reflink idref="bib20" id="ref115">20</reflink>]). This model suggests that the probability for reflection depends on many factors (e.g., developmental level, changes to task environment). Results from the present study emphasize that changes to the task environment are not equal across tasks and considerations of the role that the body plays in creating habits and resisting prepotent responses (J. P. Spencer et al., [<reflink idref="bib33" id="ref116">33</reflink>]; Wehrle, [<reflink idref="bib37" id="ref117">37</reflink>]) are important – especially when changes to the environment involve an individual's own body or movement. Future work integrating embodied perspectives of cognition (Gottwald et al., [<reflink idref="bib17" id="ref118">17</reflink>]; Marshall, [<reflink idref="bib22" id="ref119">22</reflink>]) may be helpful to understanding the limits of movement manipulations across EF tasks that vary in the extent that the body is involved.</p> <p>Finally, it is important to note that we may have also been influenced by a floor effect on the DCCS task. Children passed at a rate of 15%, which was below what we expected and noted within our registered report. The low performance on the DCCS task was surprising given previous research (e.g., P. D. Zelazo, [<reflink idref="bib44" id="ref120">44</reflink>]). It is currently unclear why DCCS performance was lower than anticipated, but a recent meta-analysis suggests that contextual factors, such as socioeconomic status, might influence DCCS performance and warrant further investigation (Doebel & Zelazo, [<reflink idref="bib14" id="ref121">14</reflink>]).</p> <hd id="AN0186129866-33">The lack of a motor manipulation effect on habit formation</hd> <p>Across both the MSML and DCCS, we found moderate to anecdotal evidence for the null hypothesis (i.e., that there were no differences in habit formation for those who wore weights compared to those who did not – both unweighted and no armbands). Although this result was anticipated for the DCCS, it went against hypotheses that adding weights during a motorically built habit in MSML preswitch trials would lead to habit disruption (i.e., more preswitch trial errors). These null results are important to consider, as they may reveal relevant considerations regarding the study and measurement of relatively subtle motoric manipulations on cognition in early childhood. It is possible that the weighted armbands did not significantly affect motor or cognitive habits, as suggested by both accuracy and response time data (see Supplemental Material).</p> <p>Another possibility for the lack of observed effects in our study may be due to a potentially weak manipulation. The lack of influence of weighted armbands on preswitch trial performance and response time may suggest that the weighted armbands were not strong enough to affect preswitch trial habit formation or slow down responses on postswitch trials. However, our finding that both weighted and unweighted armbands resulted in higher postswitch accuracy of the MSML may suggest that a focus on physical elements like response time might not be the best indicator of how the manipulation affected children in preswitch trials, as the presence of the armband (and not the weight which was hypothesized to lead to slower responses) seemed to influence accuracy on the first postswitch trial in the MSML. Future research may consider mixed methods such as children's qualitative responses related to their experience wearing the wristband (e.g., many of the children in our present study indicated the armbands affected them as evidenced by the unanticipated refusals to wear them during the tasks, see also Piaget, [<reflink idref="bib30" id="ref122">30</reflink>]; [<reflink idref="bib30" id="ref123">30</reflink>]), movement trajectories, or physiological responses to better understand how the mechanisms for the wristband's influence on performance.</p> <hd id="AN0186129866-34">Conclusions</hd> <p>The present work suggests that motor manipulation seems to produce differential effects on EF task performance depending on whether the task is based on a response shifting or attention shifting task. Although the proposed mechanism for improved performance within the attention shifting MSML task is reflection, our results also suggest that more work is needed to better understand the dynamic interplay between movement and cognition within EF.</p> <hd id="AN0186129866-35">Disclosure statement</hd> <p>No potential conflict of interest was reported by the author(s).</p> <hd id="AN0186129866-36">Data availability statement</hd> <p>The data that support the findings of this study are openly available on Open Science Framework upon publication under the OSF registration doi:10.17605/OSF.IO/2QDSG.</p> <hd id="AN0186129866-37">Supplementary material</hd> <p>Supplemental data for this article can be accessed online at https://doi.org/10.1080/15248372.2024.2444970</p> <ref id="AN0186129866-38"> <title> References </title> <blist> <bibl id="bib1" idref="ref36" type="bt">1</bibl> <bibtext> Arterberry, M. E., Hespos, S. J., & Herth, R. A. (2018). Integration of thought and action: Arm weights facilitate search accuracy in 24-month-old children. Infancy, 23 (2), 173 – 193. https://doi.org/10.1111/infa.12208</bibtext> </blist> <blist> <bibl id="bib2" idref="ref23" type="bt">2</bibl> <bibtext> Berger, S. E. (2010). Locomotor expertise predicts infants' perseverative errors. Developmental Psychology, 46 (2), 326. https://doi.org/10.1037/a0018285</bibtext> </blist> <blist> <bibl id="bib3" idref="ref21" type="bt">3</bibl> <bibtext> Bertenthal, B. I., & Barton, A. A. (2017). Perception and action. Encyclopedia of Infant and Early Childhood Development, 498 – 513. https://doi.org/10.1016/b978-0-12-809324-5.05862-4</bibtext> </blist> <blist> <bibl id="bib4" idref="ref22" type="bt">4</bibl> <bibtext> Bertenthal, B. I., & Barton, A. A. (2020). Perception and action. In J. B. Benson (Ed.), Encyclopedia of infant and early childhood development second edition (pp. 498 – 513). Elsevier. https://doi.org/10.1016/B978-0-12-809324-5.05862-4</bibtext> </blist> <blist> <bibl id="bib5" idref="ref41" type="bt">5</bibl> <bibtext> Berthier, N. E., DeBlois, S., Poirier, C. R., Novak, M. A., & Clifton, R. K. (2000). Where's the ball? two- and three-year-olds reason about unseen events. Developmental Psychology, 36 (3), 394. https://doi.org/10.1037/0012-1649.36.3.394</bibtext> </blist> <blist> <bibl id="bib6" idref="ref2" type="bt">6</bibl> <bibtext> Blair, C., & Razza, R. P. (2007). Relating effortful control, executive function, and false belief understanding to emerging math and literacy ability in kindergarten. Child Development, 78 (2), 647 – 663. https://doi.org/10.1111/j.1467-8624.2007.01019.x</bibtext> </blist> <blist> <bibl id="bib7" idref="ref31" type="bt">7</bibl> <bibtext> Buss, A. T., & Spencer, J. P. (2014). The emergent executive: A dynamic field theory of the development of executive function. Monographs of the Society for Research in Child Development, 79 (2), vii, 1 – 103. https://doi.org/10.1002/mono.12096</bibtext> </blist> <blist> <bibl id="bib8" idref="ref5" type="bt">8</bibl> <bibtext> Caporaso, J. S., Boseovski, J. J., & Marcovitch, S. (2019). The individual contributions of three executive function components to preschool social competence. Infant and Child Development, 28 (4), 1 – 17. https://doi.org/10.1002/icd.2132</bibtext> </blist> <blist> <bibl id="bib9" idref="ref24" type="bt">9</bibl> <bibtext> Clearfield, M. W., Diedrich, F. J., Smith, L. B., & Thelen, E. (2006). Young infants reach correctly in A-not-B tasks: On the development of stability and perseveration. Infant Behavior and Development, 29 (3), 435 – 444. https://doi.org/10.1016/j.infbeh.2006.03.001</bibtext> </blist> <blist> <bibtext> Devine, R. T., Ribner, A., & Hughes, C. (2019). Measuring and predicting individual differences in executive functions at 14 months: A longitudinal study. Child Development, 90 (5), e618 – e636. https://doi.org/10.1111/cdev.13217</bibtext> </blist> <blist> <bibtext> 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. https://doi.org/10.2307/1130099</bibtext> </blist> <blist> <bibtext> Diamond, A. (2013). Executive function. Annual Review of Psychology, 64 (1), 135 – 168. https://doi.org/10.1146/annurev-psych-113011-143750</bibtext> </blist> <blist> <bibtext> Distefano, R., Grenell, A., Palmer, A. R., Houlihan, K., Masten, A. S., & Carlson, S. M. (2021). Self-regulation as promotive for academic achievement in young children across risk contexts. Cognitive Development, 58, 101050. https://doi.org/10.1016/j.cogdev.2021.101050</bibtext> </blist> <blist> <bibtext> Doebel, S., & Zelazo, P. D. (2015). A meta-analysis of the dimensional change card sort: Implications for developmental theories and the measurement of executive function in children. Developmental Review, 38, 241 – 268. https://doi.org/10.1016/j.dr.2015.09.001</bibtext> </blist> <blist> <bibtext> Field, A. P., & Wilcox, R. R. (2017). Robust statistical methods: A primer for clinical psychology and experimental psychopathology researchers. Behaviour Research and Therapy, 98, 19 – 38. https://doi.org/10.1016/j.brat.2017.05.013</bibtext> </blist> <blist> <bibtext> Garon, N., Bryson, S. E., & Smith, I. M. (2008). Executive function in preschoolers: A review using an integrative framework. Psychological Bulletin, 134 (1), 31 – 60. https://doi.org/10.1037/0033-2909.134.1.31</bibtext> </blist> <blist> <bibtext> Gottwald, J. M., Achermann, S., Marciszko, C., Lindskog, M., & Gredebäck, G. (2016). An embodied account of early executive-function development: Prospective motor control in infancy is related to inhibition and working memory. Psychological Science, 27 (12), 1600 – 1610. https://doi.org/10.1177/0956797616667447</bibtext> </blist> <blist> <bibtext> Koziol, L. F., Budding, D. E., & Chidekel, D. (2012). From movement to thought: Executive function, embodied cognition, and the cerebellum. The Cerebellum, 11 (2), 505 – 525. https://doi.org/10.1007/s12311-011-0321-y</bibtext> </blist> <blist> <bibtext> Marcovitch, S., & Zelazo, P. (2006). Influence of number of a trials on 2-year-olds' behavior in two A-Not-B type search tasks. Journal of Cognition and Development, 7 (4), 477 – 501. https://doi.org/10.1207/s15327647jcd0704_3</bibtext> </blist> <blist> <bibtext> Marcovitch, S., & Zelazo, P. D. (2009). A hierarchical competing systems model of the emergence and early development of executive function. Developmental Science, 12 (1), 1 – 18. https://doi.org/10.1111/j.1467-7687.2008.00754.x</bibtext> </blist> <blist> <bibtext> Marcovitch, S., Zelazo, P. D., & Schmuckler, M. A. (2002). The effect of the number of a trials on performance on the A-not-B task. Infancy, 3 (4), 519 – 529. https://doi.org/10.1207/S15327078IN0304_06</bibtext> </blist> <blist> <bibtext> Marshall, P. J. (2016). Embodiment and human development. Child Development Perspectives, 10 (4), 245 – 250. https://doi.org/10.1111/cdep.12190</bibtext> </blist> <blist> <bibtext> Miller, S. E., & Marcovitch, S. (2011). Toddlers benefit from labeling on an executive function search task. Journal of Experimental Child Psychology, 108 (3), 580 – 592. https://doi.org/10.1016/j.jecp.2010.10.008</bibtext> </blist> <blist> <bibtext> Miller, S. E., & Marcovitch, S. (2012). How theory of mind and executive function co-develop. Review of Philosophy and Psychology, 3 (4), 597 – 625. https://doi.org/10.1007/s13164-012-0117-0</bibtext> </blist> <blist> <bibtext> Miller, S. E., & Marcovitch, S. (2015). Examining executive function in the second year of life: Coherence, stability, and relations to joint attention and language. Developmental Psychology, 51 (1), 101 – 114. https://doi.org/10.1037/a0038359</bibtext> </blist> <blist> <bibtext> Miller, S. E., Marcovitch, S., & Waldron, L. C. (2009, April). The influence of number of A-trials, inhibition, and language ability in a multistep multilocation search task for preschoolers. Poster presented at the Biennial Meeting of the Society for Research in Child Development, Denver, CO.</bibtext> </blist> <blist> <bibtext> Pearce, R., Marcovitch, S., & Miller, S. E. (2019, October). Testing the hierarchical competing systems Model in young preschoolers' Executive function. Poster presented at the Biennial Meeting of the Cognitive Development Society, Louisville, KY.</bibtext> </blist> <blist> <bibtext> Perone, S., Simmering, V. R., & Buss, A. T. (2021). A dynamical reconceptualization of executive-function development. Perspectives on Psychological Science, 1 – 11. https://doi.org/10.1177/1745691620966792</bibtext> </blist> <blist> <bibtext> Piaget, J. (1954). The construction of reality in the child. Basic Books.</bibtext> </blist> <blist> <bibtext> Piaget, J. (1976). The grasp of consciousness. Harvard University Press. (Original work published in 1974.</bibtext> </blist> <blist> <bibtext> Rivière, J., & Lécuyer, R. (2008). Effects of arm weight on C-not-B task performance: Implications for the motor inhibitory deficit account of search failures. Journal of Experimental Child Psychology, 100 (1), 1 – 16. https://doi.org/10.1016/j.jecp.2008.01.005</bibtext> </blist> <blist> <bibtext> Smith, L. B., & Thelen, E. (1994). A dynamic systems approach to the development of cognition and action. Bradford Books.</bibtext> </blist> <blist> <bibtext> Spencer, J. P., Austin, A., & Schutte, A. R. (2012). Contributions of dynamic systems theory to cognitive development. Cognitive Development, 27 (4), 401 – 418. https://doi.org/10.1016/j.cogdev.2012.07.006</bibtext> </blist> <blist> <bibtext> Spencer, J. R., Smith, L. B., & Thelen, E. (2001). Tests of a dynamic systems account of the A-Not-B error: The influence of prior experience on the spatial memory abilities of two-year-olds. Child Development, 72 (5), 1327 – 1346. https://doi.org/10.1111/1467-8624.00351</bibtext> </blist> <blist> <bibtext> Thelen, E., Schöner, G., Scheier, C., & Smith, L. B. (2001). The dynamics of embodiment: A field theory of infant perseverative reaching. Behavioral and Brain Sciences, 24 (1), 1 – 34. https://doi.org/10.1017/S0140525X01003910</bibtext> </blist> <blist> <bibtext> Thelen, E., & Smith, L. B. (2006). Dynamic systems theories. In R. M. Lerner & W. Damon (Eds.), Handbook of child psychology (pp. 258 – 312). https://doi.org/10.1002/9780470147658.chpsy0106</bibtext> </blist> <blist> <bibtext> Wehrle, M. (2021). 'Bodies (that) matter': The role of habit formation for identity. Phenomenology and the Cognitive Sciences, 20 (2), 365 – 386. https://doi.org/10.1007/s11097-020-09703-0</bibtext> </blist> <blist> <bibtext> White, L. K., Moore, T. M., Calkins, M. E., Wolf, D. H., Satterthwaite, T. D., Leibenluft, E., & Gur, R. E. (2017). An evaluation of the specificity of executive function impairment in developmental psychopathology. Journal of the American Academy of Child and Adolescent Psychiatry, 56 (11), 975 – 982.e3. https://doi.org/10.1016/j.jaac.2017.08.016</bibtext> </blist> <blist> <bibtext> Wiebe, S. A., Sheffield, T., Nelson, J. M., Clark, C. A. C., Chevalier, N., & Espy, K. A. (2011). What is event history analysis? Journal of Experimental Child Psychology, 108 (3), 436 – 452. https://doi.org/10.1016/j.jecp.2010.08.008</bibtext> </blist> <blist> <bibtext> Wilcox, R. (2012). Comparing two groups. Introduction to robust estimation and hypothesis testing (pp. 137 – 213). https://doi.org/10.1016/b978-0-12-386983-8.00005-6.</bibtext> </blist> <blist> <bibtext> Wilcox, R. R., & Rousselet, G. A. (2023). Methods for dealing with bad data and inadequate models: Distributions, linear models, and beyond. In H. Cooper, M. N. Coutanche, M. M. L., T. P. A., D. Rindskopf, & K. J. Sher (Eds.), APA handbook of research methods in psychology: Data analysis and research publication (2nd ed., pp. 5 – 25). American Psychological Association. https://doi.org/10.1037/0000320-001.</bibtext> </blist> <blist> <bibtext> Willoughby, M. T., Magnus, B., Vernon-Feagans, L., & Blair, C. B. (2017). Developmental delays in executive function from 3 to 5 years of age predict kindergarten academic readiness. Journal of Learning Disabilities, 50 (4), 359 – 372. https://doi.org/10.1177/0022219415619754</bibtext> </blist> <blist> <bibtext> Zelazo, P., Carter, A. S., Reznik, J., & Frye, D. (1997). Early development of executive function: A problem-solving framework. Review of General Psychology, 1 (2), 198 – 226. https://doi.org/10.1037/1089-2680.1.2.198</bibtext> </blist> <blist> <bibtext> Zelazo, P. D. (2006). The dimensional change card sort (DCCS): A method of assessing executive function in children. Nature Protocols, 1 (1), 297 – 301. https://doi.org/10.1038/nprot.2006.46</bibtext> </blist> <blist> <bibtext> Zelazo, P. D., Anderson, J. E., Richler, J., Wallner-Allen, K., Beaumont, J. L., & Weintraub, S. (2013). II. NIH toolbox Cognition Battery (CB): Measuring executive function and attention. Monographs of the Society for Research in Child Development, 78 (4), 16 – 33. https://doi.org/10.1111/mono.12032</bibtext> </blist> <blist> <bibtext> Zelazo, P. D., & Carlson, S. M. (2012). Hot and cool executive function in childhood and adolescence: Development and plasticity. Child Development Perspectives, 6 (4), 354 – 360. https://doi.org/10.1111/j.1750-8606.2012.00246.x</bibtext> </blist> <blist> <bibtext> Zelazo, P. D., Reznick, J. S., & Spinazzola, J. (1998). Representational flexibility and response control in a multistep multilocation search task. Developmental Psychology, 34 (2), 203 – 214. https://doi.org/10.1037/0012-1649.34.2.203</bibtext> </blist> </ref> <ref id="AN0186129866-39"> <title> Footnotes </title> <blist> <bibtext> The preregistered methodology reported the use of a Surface Pro 3 tablet, however, a second Surface Pro 4 tablet with the same monitor size was also utilized to allow for data collection at multiple sites.</bibtext> </blist> <blist> <bibtext> In the preregistered report, we initially planned for 14 children per condition, but group sizes varied (approximately equal to 14) due to circumstances further discussed in the results section.</bibtext> </blist> <blist> <bibtext> We had 2 instances when the experimenter forgot to place weighted armbands on children's wrists. These children were assigned to the full no armband condition.</bibtext> </blist> <blist> <bibtext> Initially, the software was programmed to end the task session if the touchscreen recorded 3 or more errors. However, following testing with 26 children, we changed the programming so that all participants were presented with postswitch trials because of 4 instances where the touchscreen inaccurately registered a response incorrect (e.g., armband may have hit the screen as they prepared to touch the screen). Allowing children to move on to the postswitch regardless of what the touchscreen recorded permitted us to override 8 cases in which the touchscreen inaccurately scored preswitch trials as incorrect based on visual observation during testing or video review (e.g., something bumping the touchscreen) to avoid prematurely ending the testing when children should have moved forward to the postswitch. We made this change for both the MSML and the DCCS.</bibtext> </blist> <blist> <bibtext> The traditional DCCS protocol considers children to pass the preswitch phase if they complete five out of the six preswitch trials correctly. We made the passing criteria more lenient to align with the protocol and passing criteria of the multistep multilocation search task.</bibtext> </blist> <blist> <bibtext> See footnote 4.</bibtext> </blist> <blist> <bibtext> Analyses of MSML preswitch errors with children who both passed and failed preswitch trials revealed an identical pattern of null results.</bibtext> </blist> <blist> <bibtext> Significant shapiro-wilks tests for normality were evident across all six conditions of the DCCS (<emph>Ws</emph> <.53, <emph>ps</emph> <.001). Outliers were also identified in the data. Results from the ANOVA analyses we originally proposed were compared to the results of the robust 2-way ANOVA (Wilcox, [40]; Field & Wilcox, [15]) that have been suggested to address distributions that violate normality and/or outliers (Wilcox & Rousselet, [41]). The results of the robust ANOVA were similar in that no significant main effects or interactions were present, Fs < 1.97, <emph>ps ></emph>.39.</bibtext> </blist> <blist> <bibtext> Analyses of DCCS preswitch errors with children who both passed and failed preswitch trials revealed an identical pattern of null results.</bibtext> </blist> </ref> <aug> <p>By Z. Reagan Pearce and Stephanie E. Miller</p> <p>Reported by Author; Author</p> </aug> <nolink nlid="nl1" bibid="bib43" firstref="ref1"></nolink> <nolink nlid="nl2" bibid="bib13" firstref="ref3"></nolink> <nolink nlid="nl3" bibid="bib42" firstref="ref4"></nolink> <nolink nlid="nl4" bibid="bib12" firstref="ref6"></nolink> <nolink nlid="nl5" bibid="bib38" firstref="ref7"></nolink> <nolink nlid="nl6" bibid="bib39" firstref="ref8"></nolink> <nolink nlid="nl7" bibid="bib46" firstref="ref9"></nolink> <nolink nlid="nl8" bibid="bib18" firstref="ref10"></nolink> <nolink nlid="nl9" bibid="bib45" firstref="ref11"></nolink> <nolink nlid="nl10" bibid="bib29" firstref="ref12"></nolink> <nolink nlid="nl11" bibid="bib10" firstref="ref13"></nolink> <nolink nlid="nl12" bibid="bib20" firstref="ref14"></nolink> <nolink nlid="nl13" bibid="bib25" firstref="ref15"></nolink> <nolink nlid="nl14" bibid="bib16" firstref="ref17"></nolink> <nolink nlid="nl15" bibid="bib33" firstref="ref25"></nolink> <nolink nlid="nl16" bibid="bib32" firstref="ref26"></nolink> <nolink nlid="nl17" bibid="bib28" firstref="ref27"></nolink> <nolink nlid="nl18" bibid="bib36" firstref="ref28"></nolink> <nolink nlid="nl19" bibid="bib35" firstref="ref29"></nolink> <nolink nlid="nl20" bibid="bib34" firstref="ref30"></nolink> <nolink nlid="nl21" bibid="bib21" firstref="ref33"></nolink> <nolink nlid="nl22" bibid="bib31" firstref="ref35"></nolink> <nolink nlid="nl23" bibid="bib47" firstref="ref43"></nolink> <nolink nlid="nl24" bibid="bib44" firstref="ref44"></nolink> <nolink nlid="nl25" bibid="bib23" firstref="ref45"></nolink> <nolink nlid="nl26" bibid="bib11" firstref="ref47"></nolink> <nolink nlid="nl27" bibid="bib27" firstref="ref49"></nolink> <nolink nlid="nl28" bibid="bib19" firstref="ref51"></nolink> <nolink nlid="nl29" bibid="bib26" firstref="ref59"></nolink> <nolink nlid="nl30" bibid="bib93" firstref="ref71"></nolink> <nolink nlid="nl31" bibid="bib82" firstref="ref72"></nolink> <nolink nlid="nl32" bibid="bib24" firstref="ref110"></nolink> <nolink nlid="nl33" bibid="bib30" firstref="ref111"></nolink> <nolink nlid="nl34" bibid="bib37" firstref="ref117"></nolink> <nolink nlid="nl35" bibid="bib17" firstref="ref118"></nolink> <nolink nlid="nl36" bibid="bib22" firstref="ref119"></nolink> <nolink nlid="nl37" bibid="bib14" firstref="ref121"></nolink>
Header DbId: eric
DbLabel: ERIC
An: EJ1495028
AccessLevel: 3
PubType: Academic Journal
PubTypeId: academicJournal
PreciseRelevancyScore: 0
IllustrationInfo
Items – Name: Title
  Label: Title
  Group: Ti
  Data: Weighing down Thought: The Effect of Motor Disruption in Executive Function Tasks among Three-Year-Old Children
– Name: Language
  Label: Language
  Group: Lang
  Data: English
– Name: Author
  Label: Authors
  Group: Au
  Data: <searchLink fieldCode="AR" term="%22Z%2E+Reagan+Pearce%22">Z. Reagan Pearce</searchLink><br /><searchLink fieldCode="AR" term="%22Stephanie+E%2E+Miller%22">Stephanie E. Miller</searchLink>
– Name: TitleSource
  Label: Source
  Group: Src
  Data: <searchLink fieldCode="SO" term="%22Journal+of+Cognition+and+Development%22"><i>Journal of Cognition and Development</i></searchLink>. 2025 26(3):398-426.
– Name: Avail
  Label: Availability
  Group: Avail
  Data: Routledge. Available from: Taylor & Francis, Ltd. 530 Walnut Street Suite 850, Philadelphia, PA 19106. Tel: 800-354-1420; Tel: 215-625-8900; Fax: 215-207-0050; Web site: http://www.tandf.co.uk/journals
– Name: PeerReviewed
  Label: Peer Reviewed
  Group: SrcInfo
  Data: Y
– Name: Pages
  Label: Page Count
  Group: Src
  Data: 29
– Name: DatePubCY
  Label: Publication Date
  Group: Date
  Data: 2025
– Name: TypeDocument
  Label: Document Type
  Group: TypDoc
  Data: Journal Articles<br />Reports - Research
– Name: Subject
  Label: Descriptors
  Group: Su
  Data: <searchLink fieldCode="DE" term="%22Executive+Function%22">Executive Function</searchLink><br /><searchLink fieldCode="DE" term="%22Preschool+Children%22">Preschool Children</searchLink><br /><searchLink fieldCode="DE" term="%22Motor+Reactions%22">Motor Reactions</searchLink><br /><searchLink fieldCode="DE" term="%22Attention%22">Attention</searchLink><br /><searchLink fieldCode="DE" term="%22Habit+Formation%22">Habit Formation</searchLink><br /><searchLink fieldCode="DE" term="%22Accuracy%22">Accuracy</searchLink>
– Name: DOI
  Label: DOI
  Group: ID
  Data: 10.1080/15248372.2024.2444970
– Name: ISSN
  Label: ISSN
  Group: ISSN
  Data: 1524-8372<br />1532-7647
– Name: Abstract
  Label: Abstract
  Group: Ab
  Data: This registered report examines the effect of motor disruption in the conscious control of behavior (i.e. executive function -- EF) of three-year-old children in the Southern United States (N = 114). Specifically, we investigated how disrupting a relevant motor response influenced performance on two EF tasks. Each EF task had an initial phase intended to form a habit, and a switch phase that was the critical measure of EF. Children must inhibit the habit developed at the beginning of the task and reflect consciously on new demands. The multistep multilocation task examined the inhibition of a motor-based habit (response shift), while the Dimensional Change Card Sort examined the inhibition of a cognitive-based habit (attention shift). The motor disruption consisted of administering weighted armbands at two time points during the tasks (i.e. the beginning of the task or during switch trials only) to disrupt reaching behavior. Motor manipulation was found to have a differential effect on EF tasks among three-year-old children. While the armbands improved performance on the response shifting task relative to a no armband control condition, they did not significantly affect performance on the attention shifting task. Contrary to expectations, the armbands did not lead to more errors (habit disruption) during pre-switch trials in either task. These findings highlight the significance of considering both motor and cognitive factors in early childhood EF development. Further research is needed to explore the underlying mechanisms and implications of incorporating movement in EF development.
– Name: AbstractInfo
  Label: Abstractor
  Group: Ab
  Data: As Provided
– Name: Note
  Label: Notes
  Group: Note
  Data: https://doi.org/10.17605/OSF.IO/2QDSG
– Name: DateEntry
  Label: Entry Date
  Group: Date
  Data: 2026
– Name: AN
  Label: Accession Number
  Group: ID
  Data: EJ1495028
PLink https://search.ebscohost.com/login.aspx?direct=true&site=eds-live&db=eric&AN=EJ1495028
RecordInfo BibRecord:
  BibEntity:
    Identifiers:
      – Type: doi
        Value: 10.1080/15248372.2024.2444970
    Languages:
      – Text: English
    PhysicalDescription:
      Pagination:
        PageCount: 29
        StartPage: 398
    Subjects:
      – SubjectFull: Executive Function
        Type: general
      – SubjectFull: Preschool Children
        Type: general
      – SubjectFull: Motor Reactions
        Type: general
      – SubjectFull: Attention
        Type: general
      – SubjectFull: Habit Formation
        Type: general
      – SubjectFull: Accuracy
        Type: general
    Titles:
      – TitleFull: Weighing down Thought: The Effect of Motor Disruption in Executive Function Tasks among Three-Year-Old Children
        Type: main
  BibRelationships:
    HasContributorRelationships:
      – PersonEntity:
          Name:
            NameFull: Z. Reagan Pearce
      – PersonEntity:
          Name:
            NameFull: Stephanie E. Miller
    IsPartOfRelationships:
      – BibEntity:
          Dates:
            – D: 01
              M: 01
              Type: published
              Y: 2025
          Identifiers:
            – Type: issn-print
              Value: 1524-8372
            – Type: issn-electronic
              Value: 1532-7647
          Numbering:
            – Type: volume
              Value: 26
            – Type: issue
              Value: 3
          Titles:
            – TitleFull: Journal of Cognition and Development
              Type: main
ResultId 1