Therapists’ Use of Instructions and Feedback in Motor Learning Interventions in Children with Developmental Coordination Disorder: A Video Observation Study

Abstract Aim This qualitative study explored therapists’ use of instructions and feedback when teaching motor tasks to children with developmental coordination disorder (DCD) as a first step in developing practical recommendations. Methods A conventional content analysis approach was used to analyze videotaped treatment sessions of physical therapists using a newly developed analysis plan. Inductive coding was used to code purposively selected video segments. The codes were sorted into categories to identify key themes. Analyses were performed independently by two researchers until data saturation was reached. Results Ten video-taped sessions were analyzed and 61 segments were coded. Three key themes were identified: (1) therapists’ intention with the instructions and feedback was to motivate or to provide information; (2) the preferred therapists’ teaching style was either direct or indirect; and (3) parameters to shape specific instructions and feedback were the focus of attention, modality, information content, timing and frequency. Conclusion Therapists used numerous instructions and feedback with different information content, often shaped by multiple focuses and/or modalities to motivate children or to provide specific information about task performance. Although therapists adapted instructions and feedback to child and task, future research should explore how characteristics of child and task can guide therapists’ clinical decision-making.


Introduction
Teaching motor skills is a fundamental part of the professions of occupational and physical therapists (Atun-Einy & Kafri, 2019;Kleynen et al., 2016;MacWilliam et al., 2021).A think-aloud study in which physical therapists (PTs) reflected while watching videos of their own treatment sessions with patients with acquired brain injury (ABI) showed that they used a great variety of motor learning strategies to teach motor skills, which they constantly adapted to the individual situation (Kleynen et al., 2016).In an interview study, PTs acknowledged the importance of using motor learning strategies, but they experienced teaching motor skills as a complex construct which they largely addressed intuitively (Atun-Einy & Kafri, 2019).Moreover, they emphasized that more insight was needed in translating theory into clinical practice to improve implementation of motor learning strategies (Atun-Einy & Kafri, 2019).Interview studies exploring how therapists perceive and experience using motor learning strategies to teach children motor skills are currently lacking.However, one study explored occupational therapists' (OTs) use of motor learning strategies in video-taped treatment sessions of children with ABI (MacWilliam et al., 2021).Results showed that therapists used various motor learning strategies which they adapted to child, task and environmental characteristics (MacWilliam et al., 2021).Published research shows that theories of motor learning lack the clarity and simplicity needed for application in pediatric practice (Zwicker & Harris, 2009).To support implementation of motor learning theories into clinical settings, several theoretical frameworks with diverse approaches, addressing different motor learning strategies, have been described (Guadagnoli & Lee, 2004;Kafri & Atun-Einy, 2019;Kleynen et al., 2015;Wulf & Lewthwaite, 2016).All consider instructions and feedback to be important (Guadagnoli & Lee, 2004;Kafri & Atun-Einy, 2019;Kleynen et al., 2015;Wulf & Lewthwaite, 2016).As such, practical recommendations on how to use instructions and feedback may enhance teaching motor skills.
Children with developmental coordination disorder (DCD) experience severe problems with motor coordination and learning, impairing their participation in daily life (O'Dea et al., 2021;Subara-Zukic et al., 2022).In order to improve daily functioning, OTs and/or (pediatric) physical therapists (PPTs) use activity-or participation-oriented interventions to teach motor skill to children with DCD implicitly and/or explicitly (Blank et al., 2019).With implicit motor learning processes, children learn motor skills without awareness and without no or minimal increase in verbal knowledge about the movement performance, whereas with explicit motor learning processes the learning process involves cognitive processes and verbal knowledge is generated about the movement performance (Kleynen et al., 2014).Evidence-based activity-and participation oriented interventions (Smits-Engelsman et al., 2018), such as Neuromotor Task Training (NTT) or Cognitive Orientation to daily Occupational Performance (CO-OP), are underpinned by theories of motor learning and motor control (Blank et al., 2019;Missiuna et al., 2001;Schoemaker et al., 2003).Both approaches are child-centered, and based on the theoretical foundation that motor performance is the result of interactions between characteristics of child, task, and environment as described by the Dynamic Systems Theory (Kamm et al., 1990;Missiuna et al., 2001;Schoemaker et al., 2003;Thelen, 2005;Wilson et al., 2017).Furthermore, motor learning strategies derived from motor learning research are addressed to enhance motor performance by manipulating the interaction of child, task and environment (Levac et al., 2009;Missiuna et al., 2001;Schoemaker et al., 2003).A pilot video-observation study in children with DCD showed that PPTs used a variety of instructions and feedback to improve motor performance (Niemeijer et al., 2006).However, little is known about the effectiveness of instructions and feedback in children with DCD.Only the effectiveness of the focus of attention has been investigated, and results from this were inconclusive (Jarus et al., 2015;Khatab et al., 2018;Li et al., 2019;Psotta et al., 2020;van Cappellen-van Maldegem et al., 2018).In these studies, two types of focuses were distinguished: (1) an external focus (EF), directing the attention of the child to the impact of the movement on the environment; and (2) an internal focus (IF), directing the attention of the child to its body movements (Wulf et al., 1998).Two studies found beneficial effects favoring EF (Li et al., 2019;Psotta et al., 2020), while three studies found no significant differences between EF and IF (Jarus et al., 2015;Khatab et al., 2018;van Cappellen-van Maldegem et al., 2018).As instructions and feedback are shaped by their content (e.g., a specific focus of attention) and modality (e.g., visual, tactile, auditory or verbal), and applied with a chosen frequency and timing (e.g., determined by therapist or child) (Levac et al., 2011), more insight is needed into how instructions and feedback can and should be shaped for children with DCD.
In summary, therapists experience teaching motor skills as challenging, and some practical recommendations on the use of instructions and feedback may enhance their teaching skills (Atun-Einy & Kafri, 2019;Kafri & Atun-Einy, 2019;Zwicker & Harris, 2009).Although therapists constantly adapt instructions and feedback to the child, task, and environment (Kleynen et al., 2016;MacWilliam et al., 2021), this seems to be based largely on intuition (Atun-Einy & Kafri, 2019).In interventions with children with DCD, teaching motor skills is additionally hindered because of limited knowledge of the effectiveness of instructions and feedback in such children.As a first step in developing practical recommendations on the use of instructions and feedback with these children, more insight is needed into therapists' current use of their instructions and feedback.The aim of this study was therefore to explore how therapists use instructions and feedback during video-taped treatment sessions with children with DCD aged 5 to 12 years.Existing tools to analyze video-taped sessions score the frequencies of a set of predefined items (Levac et al., 2011;MacWilliam et al., 2021;Niemeijer et al., 2003), providing insufficient insight into the full spectrum of therapists' use of instructions and feedback.Furthermore, these existing tools do not provide insight into how therapists act if their instructions and feedback do not seem to meet their expectations of immediate improvement in the motor task being practiced.Accordingly, the researchers developed a new and comprehensive analysis plan to explore and analyze these sessions in four consecutive steps: (1) splitting the video into smaller segments; (2) writing comments to each segment; (3) selecting segments for in-depth analysis; and (4) coding the selected segments.

Design
This qualitative study used a conventional content analysis approach to analyze videotaped treatment sessions and subsequent interviews with PPTs to explore therapists' use of motor learning strategies in the treatment of children with DCD.This article describes the results of analyses of treatment sessions, which focused on therapists' use of instructions and feedback.In order to prevent the observations being unduly influenced by the interview, sessions were recorded at least one week before the interview.The study was approved by the Medical Ethical Review Board of Maastricht University (2019-1338) for Dutch participants, and Hasselt University (CME2019/060) for Flemish participants.

Procedure
A convenience sampling strategy was used to recruit therapists (Moser & Korstjens, 2018).A flyer was distributed within two regional networks of PPTs in the southern Netherlands (RVFK and network Den Bosch), within the University of Hasselt's network for PT clinical internships, and at educational activities for therapists (e.g., symposia).A heterogeneous sample was required to obtain rich data (Moser & Korstjens, 2018).Therefore, therapists with different backgrounds in terms of experience in treating children with DCD and work settings (e.g., primary and secondary care) were selected.Therapists completed a short questionnaire to supply their demographic characteristics (age, work setting, graduation year, and years of experience in treating children with DCD).The Flemish data and copies of the Dutch data were stored, coded and secured, on the server of the University of Hasselt.Only one researcher (IvdV) had access to the code-key, the others having access to the coded videos and documents.Before the recording of the treatment sessions, therapists contacted children receiving their care (and their parents) to inform them about the study.After receiving written and oral information, therapists and parents gave written consent for participation, with children assenting orally.
The therapists were asked to videotape themselves during a regular treatment session in which they taught the child motor skills, with both child and therapist visible and audible.Each video was recorded with the therapist's own video camera or telephone, tripod-mounted or held by the parent.Each therapist provided one video, accompanied with information about the treatment goal(s) of that session.
Data were collected until no new meaningful information was to be gained (Boeije & Bleijenbergh, 2019;Moser & Korstjens, 2018).An additional two more sessions were analyzed to confirm that data saturation had been reached (Moser & Korstjens, 2018).Based on a previous video-observation study exploring OTs' use of motor learning variables in children with ABI, it was expected that 8-10 videos would be sufficient (MacWilliam et al., 2021).

Participants
Dutch and Flemish PTs, with at least one year of experience in treating children with DCD, could participate if they were able to video-tape their treatment sessions with a child (aged 5-12 years) with (probable) DCD.The child should preferably have been diagnosed with DCD: however, in the Netherlands, the mean age of receiving a diagnosis of DCD is 7.02 years (SD 1.79) and the process of diagnosis takes an average of 2.79 years (SD 2.13) (Lust et al., 2022).Therefore, children with probable DCD were included if the criteria for DCD according to the international recommendations were met: (I) scoring 16 th percentile of the Movement Assessment Battery for Children 2nd edition (Movement-ABC-2NL); (II) suspected of having DCD according to the Developmental Coordination Disorder Questionnaire (DCDQ); (III) no other condition that could account for the motor skills deficits was reported by the therapist; (IV) there had been an early onset of symptoms, as reported by the therapist (Blank et al., 2019).

A video-based analysis plan
Three members of the research group (IvdV, NvdW, ER) developed a plan to analyze video-taped treatment sessions, which allowed for reviewing the sessions at two levels (Pearce et al., 2014): (1) the treatment session as whole; and (2) in-depth analysis of specific segments using an inductive coding strategy (Boeije & Bleijenbergh, 2019;Hsieh & Shannon, 2005).See Appendix 1 for more information about the development of the plan.
Analyses were conducted in two phases: Phase 1 comprised of three steps and Phase 2 of one step.To promote internal validity of the investigation, all steps within the analyses were performed independently by two researchers.Phase 1 was performed by Researchers 1 (IvdV or NvdW) and 2 (MG): afterwards, differences were discussed to advance the understanding of the data.In Phase 2, segments were coded by Researcher 3, who had no prior knowledge of the video (IvdV or NvdW).The open coding was checked and complemented by Researcher 1, and subsequently discussed by both researchers.If these could not reach a consensus, an independent researcher (ER) was consulted.Throughout the analysis process, notes were made of first impressions and thoughts.Frequent meetings were organized to continuously reflect on the process and results of the analyses with the whole research group, comprising researchers with methodological and/or clinical expertise.
Within the analyses of Phase 1, three consecutive steps were conducted.In Step 1, videos were split into relevant smaller segments.We assumed that the instructions and feedback provided by therapists would be related to each other.Thus, in order to better understand therapists' use of instructions and feedback, each segment had to contain: the instructions; the task performance; and the therapists' reactions to that performance (e.g., feedback).In Step 2, each segment was provided with comments about its content, firstly, whether it concerned practising tasks or other activities (e.g., organizing practice situations) and, secondly, if it concerned practising tasks, (a) the task practiced, (b) the motor learning strategies applied, (c) the amount of therapists' actions to enhance motor learning, and (d) whether the segment contained unique information when compared with other segments of that video.Further, each segment received a label regarding the type(s) of motor learning used.The label was based on: (1) the used focus of attention (e.g., EF), because an EF promotes implicit motor learning, and an IF explicit motor learning (Wulf et al., 2001); and (2) the amount of information given because the amount of information that needs to be processed relates to the involvement of the working memory (Kleynen et al., 2015).The following labels were used: (1) implicit motor learning (IML); (2) more implicit than explicit motor learning (IML > EML); (3) more explicit than implicit motor learning (EML > IML); or (4) explicit motor learning (EML).If the segment showed the child practising without active guidance or showed other activities, the segment received one of the following labels: (5) no motor learning (NML); or (6) Others (Table 1 describes these labels).Finally, in Step 3, some segments were selected for further analysis.The selection process focused on segments that provided rich data or showed unique elements of motor learning.Because instructions and feedback are adapted to the child and task (Kleynen et al., 2016;MacWilliam et al., 2021;Wilson et al., 2017), and are shaped differently depending on the type of motor learning (Kleynen et al., 2015), it was important to select segments with different tasks and labels.In Phase 2, Step 4, all observable actions of the therapist that might enhance the motor learning process of the child in the selected segments were coded, using an inductive strategy.See Appendix 1 for more detailed descriptions of the steps.

Data analysis
Median age (with range), and frequencies of gender, and nationality were presented for the therapists and children separately.For therapists, a range of years of experience in treating children with DCD was presented as well.An overview of the tasks practised in the segments selected for coding was provided.The processes of recruitment and data collection were described.
A conventional content analysis approach, using ATLAS.tiversion 8, was used for qualitative analyses (Hsieh & Shannon, 2005).Videos were analyzed independently from interviews.For Phase 1, the following results were described: (1) the total length in minutes of the analyzed treatment sessions and selected segments; (2) the number of segments in total and selected for coding; (3) the distribution of the labels assigned to the segments.The coding procedure in Phase 2 involved an iterative process of coding and recoding.In multiple meetings with the research group, codes were sorted into categories based on how different codes were related and linked to each other.Subsequently, themes were formulated by organizing and grouping categories into meaningful clusters (Boeije & Bleijenbergh, 2019;Hsieh & Shannon, 2005).

Process of recruitment and data collection
In the Netherlands, 23 PPTs requested more information after receiving the flyer via the regional networks (n ¼ 10) or a symposium (n ¼ 13).Ten therapists were interested in participating.However, three had no opportunity to video-taping a treatment session, Table 1.Description of labels assigned to video segments.

Label Description IML
The therapist used an implicit motor learning approach to teach the child a motor task.

IML > EML
The therapist used a combination of implicit and explicit motor learning approaches to teach the child a motor task.However, the approach was more implicit than explicit.EML > IML The therapist used a combination of implicit and explicit motor learning approaches to teach the child a motor task.However, the approach was more explicit than implicit.

EML
The therapist used an explicit motor learning approach to teach the child a motor task.

NML
The child practiced a task.However, the therapist's approach used little or no motor learning variables.

Others
No tasks were practiced.Other activities like social talking or organizing the practice situation occurred.
leaving seven Dutch participants.In Belgium, 18 PTs requested more information after receiving the flyer via the University of Hasselt's clinical internship network (n ¼ 16) or via one of the four symposia (n ¼ 2).Eight therapists were interested in participating.However, three had no opportunity to video-taping a treatment session, leaving five Flemish participants.One therapist recorded task performances without instructions and feedback, so that video was excluded, leaving 11 videos available for analysis.Data saturation appeared reached after analyzing eight videos.The analyses of two extra videos resulted in no new meaningful information, confirming saturation.

Participants
All 10 therapists were women, with a median age of 52 years (range 26-63).Six worked in a primary health care facility, three others in a secondary health care facility; the remaining therapist worked in both.Experience in treating children with DCD ranged from 4 to 40 years.The median age of the 10 children was 6.5 years (range 5-9), with six being boys.Several gross and fine motor tasks were practised, with eight children practising more than one task during their treatment session.Table 2 provides an overview of the tasks practised in the segments selected for coding.

Findings of the analyses of the treatment sessions
In The analyses of Phase 2 resulted in three themes providing insight into therapists' use of instructions and feedback: (1) therapists' intention with the instructions and feedback; (2) therapists' teaching style; and (3) parameters to shape specific instructions and feedback.The following paragraphs will elaborate on the separate themes.Theme 1: therapists' intention with the instructions and feedback With their instructions and feedback, therapists intended to motivate the child or to provide the child with specific information about the task performance.To motivate the child, all therapists used verbal and non-verbal encouragement before, during, and/or after the execution of the task.Most encouragements were verbal comments after the performance, like "Well done!" or "Good job!", sometimes accompanied by non-verbal actions like thumbs up or high fives.Additionally, some therapists promised the child a reward or gave pep talks before the start of the performance.For instance, one therapist promised the child stickers for each forward roll made, while other therapists encouraged the child by saying "Try again, I have seen that you can do it" or "You already did a great job".During task performances, therapists gave children confidence by holding the child's hand as support when the task seemed challenging, for instance, while walking on a balance beam or jumping from a height.Additionally, they made comments like "Wow!" or "Go on!" to motivate them.Throughout treatment sessions, the majority of therapists used many different types of instructions and feedback to provide the child with specific information about task performance (further described as specific instructions and feedback).Therapists' approaches varied more when the child encountered complex challenges in task performance.For example, one therapist attempted to teach a 9-year-old child rope skipping but the child did not know how to position the handles of the skipping rope.To improve this, the therapist demonstrated how to position the handles.When the child kept struggling, she told the child that the end of the handles should point to the wall, and eventually she even placed the handles in the child's hands, positioned the hands and said that thumbs should point outwards.In general, it was observed that therapists used instructions more than specific feedback.After the performance, they often complimented the child without providing insight into what went well.In case of errors in execution, therapists frequently repeated the initial instructions or altered the modality of the instructions, for instance by changing verbal instructions into a demonstration or by adding tactile guidance to the verbal instructions.Theme 3 further elaborates on how these specific instructions and feedback were shaped.
Theme 2: therapists' teaching style Most therapists used a combination of (1) asking the child questions to guide it to the correct motor performance and/or movement solution, and (2) instructing the child what to do to improve the task performance.Although therapists frequently used both, they tended to prefer one.So two types of teaching styles could be recognized: the indirect and the direct styles.Therapists using the indirect teaching style asked the child many questions to enhance motor learning.For instance, one therapist wanted to improve catching in a 7-year-old child.Before performing the task, the therapist discussed the Goal-Plan-Do-Check strategy with the child (Missiuna et al., 2001).She drew attention to specific points of interest with questions like "What do you need to do with your hands?" or "Do you remember what was important?"She also simulated situations by asking "What do your arms need to do if the beanbag ends up here?"while holding the beanbag in the air in different positions.Furthermore, she attempted to increase insight by asking questions like "Do you think that the beanbag will go faster or slower when I will stand further away?".After the performance, she asked questions like "What was the reason why you missed two?What did you forget?" or "How did it go?"Therapists using the direct teaching style informed the child directly what to do, and/or what went well or wrong when providing feedback.For example, one therapist aimed to improve the standing long jump in a 5-year-old child.The therapist demonstrated the jump while telling the child exactly what to do with his feet, knees, hips and arms.Subsequently, the therapist performed the jump simultaneously with the child, while giving short cues like "Bent knees!" and "Push!".
Theme 3: parameters to shape specific instructions and feedback Five different relevant parameters were identified: (1) focus of attention; (2) modality; (3) information content; (4) timing; and (5) frequency.Each specific instruction or feedback was shaped by its focus and modality.For focus of attention, EF was observed the most.Other observed focuses were IF and focus on the strategy needed to perform the task (e.g., the sequence of subsidiary steps).Knowledge of Results (information about the learner's success in meeting the environmental goal) and Knowledge of Performance (information about the learner's own movements) were considered subtypes of EF and IF respectively because both provide information about the results or the movement performance as a basis for error corrections in the next trial (Salmoni et al., 1984).Analogies were classified as EF, because the child was attempting to reproduce a metaphor (Wulf & Lewthwaite, 2016).As for modality, most instructions and feedback were provided verbally.Therapists also used visual and, occasionally, tactile modalities.Specifically, for more complex tasks (e.g., writing, rope skipping, forward rolls, and tying shoe laces), therapists combined several focuses and/or modalities in their instructions and feedback.For instance, one child had an incorrect pencil grip while writing.The therapist demonstrated the correct pencil grip while emphasizing that the pencil had to stay in contact with the hand (visual and verbal modalities with IF).To improve forward rolls, the therapist demonstrated the subsidiary steps while asking questions to the child about what to do with specific body parts in each step (visual and verbal modalities, with IF and focus on strategy).During the performance, the therapist manually guided the movement and told the child to put their hands on the green dots (tactile and verbal modalities with both IF and EF).More examples can be found in the modality-focus matrix in Table 3.
The third parameter identified was the information content of individual instructions and feedback, which varied from minimal to very extensive.For instance, some therapists used short instructions with minimal information, like "Throw 10 times" or "Bend your knees" (in jumping), while others used more extensive instructions by saying "Bend your knees, hips and trunk in preparation for the jump.Swing your arms backwards and then forwards while pushing off with your feet".The fourth identified parameter was the timing of instructions and feedback.It was observed that instructions and feedback were provided frequently on therapists' initiative.Occasionally, however, the therapist asked whether the child wanted instructions or feedback.For instance, one therapist asked "Shall I explain it [tying shoe laces] from the beginning?"So, in some

Visual
The therapist demonstrated a bench jump.While hopping a pattern between cones on one leg, the therapist pointed in the direction the child had to go.While writing letter-like patterns, the therapist pointed at the child's wrist to draw attention to its position.Visual instructions and feedback with focus on strategy were always combined with verbal guidance.Visual instructions and feedback with multiple focuses were always combined with verbal guidance.

Tactile
Not possible: when the therapist touches the child, the attention is always drawn to the body.Therefore, the combination of tactile instructions and feedback with a merely EF is not possible.Before single leg hopping, the therapist positioned the child's non-jumping leg.
The child stepped over an obstacle while walking a balance beam.The therapist corrected balance by grasping the child's hand when it lost balance.Not possible: when the therapist touches the child, the attention is always drawn to the body.Therefore, the combination of a tactile form with a focus only on strategy is not possible.Tactile instructions and feedback with IF and focus on strategy were always combined with verbal guidance (continued) cases, the timing of instructions and feedback was determined together with the child.The fifth identified parameter was frequency.It was observed that therapists' reactions to the child's execution of the task did not always comprise specific feedback.Therapists more often gave compliments, repeated initial instructions, and/or provided new instructions with another specific element of interest as well.Furthermore, if the various segments selected per therapists were compared, it was observed that some therapists provided little specific instructions and feedback and others did more, suggesting that the frequency varied.

Discussion
This qualitative study aimed to explore therapists' use of instructions and feedback when teaching motor tasks to children with DCD.The video-taped treatment sessions showed that therapists used a lot of encouragement.Furthermore, they used numerous specific implicit and/or explicit instructions and feedback to enhance children's motor learning.They preferred either a direct or indirect style in which instructions and feedback were shaped by the parameters focus of attention, modality, information content, timing and frequency.Therapists' intentions with instructions and feedback were to motivate children to learn or to provide them with information about the performance of the task (Theme 1).Motivation is considered important in motor learning (Simpson et al., 2020;Wulf & Lewthwaite, 2016): according to the Self-Determination Theory (SDT), intrinsic motivation will be advanced by promoting competence and autonomy (Ryan & Deci, 2000).Encouragement can improve the child's feeling of competence (Ryan & Deci, 2000;Simpson et al., 2020), while providing the child with choice enhances autonomy (Ryan & Deci, 2000;Simpson et al., 2020;Wulf & Lewthwaite, 2016).The therapist can let the child choose when or in what modality the child wants to receive instructions or feedback (Lemos et al., 2017;Simpson et al., 2020;van der Veer et al., 2022).Published research has demonstrated that both motor performance and the child's perceived competence showed greater improvements when children decided when they wanted these so-called self-controlled instructions and feedback (Lemos et al., 2017;Simpson et al., 2020;van der Veer et al., 2022;Wulf et al., 2014).Because most children with DCD have lower levels of perceived athletic competence and self-esteem (Noordstar et al., 2014;Noordstar & Volman, 2020), motivational and self-controlled instructions and feedback are considered relevant.In this study, all therapists observed used encouragement but only two occasionally asked whether the child would like to have instructions, and none asked whether the child preferred a specific modality.With specific instructions and feedback, it was observed that therapists used relatively little feedback to provide the child with information about the movement performance and/or results of the task, somewhat surprisingly, given that specific feedback is considered fundamental to enhance motor learning in evidence-based interventions for children with DCD (Levac et al., 2009;Missiuna et al., 2001;Schoemaker et al., 2003).Furthermore, a meta-analysis in educational learning investigating the effectiveness of feedback on several outcome measures, including physical performance, showed that feedback with information about performance and process was more effective than feedback without that information (Wisniewski et al., 2020).Besides the informational purpose of specific feedback, it can improve competence as well.Studies have shown that children who received feedback after good trials showed greater improvements in motor task performances, and were more motivated, than children who received feedback after poor trials (Simpson et al., 2020).Because of the reported beneficial effects of self-controlled conditions and specific feedback, it would be interesting to further explore what choices therapists make in using feedback to teach motor skills to children with DCD.
The therapists used either a more direct or a more indirect teaching style to enhance motor learning in children with DCD (Theme 2).The international DCD guideline (Blank et al., 2019) recommends evidence-based interventions such as CO-OP and NTT (Smits-Engelsman et al., 2018).CO-OP strongly promotes an indirect teaching style: the therapist questions the child in order to enhance their problem-solving abilities to develop alternative solutions for the current movement problem (Missiuna et al., 2001).In NTT, both direct instructions and feedback and indirect questioning are used to enhance motor learning (Niemeijer et al., 2003).In physical education (PE), the Spectrum of Teaching Styles (STS) is a commonly used framework (Mosston & Ashworth, 2008).This describes 11 styles that PE teachers can use to teach children motor skills.It assumes that teachers will shift between styles to adapt to child characteristics (e.g., motivation, cooperation and cognitive skills) and to the task being practiced (Mosston & Ashworth, 2008).As both therapists and PE teachers teach children motor skills by adapting instructions and feedback to child and task, it would be interesting to explore which teaching styles of the STS would be preferred in children with DCD, and how these styles relate to CO-OP and NTT.
Therapists shaped their instructions and feedback using various focuses of attention and modalities, with different information content, timing and frequency (Theme 3).While labeling the video segments, the combined labels (IML > EML and EML > IML) were used most frequently (n ¼ 72).However, in general, more implicit (IML and IML > EML) (n ¼ 82) than explicit (EML and EML > IML) (n ¼ 58) labels were assigned to the segments.For the focus of attention, literature showed inconclusive results on the effectiveness of EF versus IF in children with DCD and in typically developing children, so further study is warranted (Jarus et al., 2015;Khatab et al., 2018;Li et al., 2019;Psotta et al., 2020;van Abswoude et al., 2021;van Cappellen-van Maldegem et al., 2018).As it is suggested that the effectiveness depends on the child's characteristics (Khatab et al., 2018;van Abswoude et al., 2021;van Cappellen-van Maldegem et al., 2018), it would be interesting to explore therapists' arguments for choosing a specific focus of attention.Each instance of instruction or feedback was shaped by its focus on attention and modality.In complex tasks specifically, therapists combined multiple focuses and modalities, seemingly in line with research findings suggesting that the type and amount of information needed to learn new skills depends on the level of difficulty of the task (Guadagnoli & Lee, 2004;Wulf & Shea, 2002).Furthermore, it was observed that therapists changed the focus of attention and/or modality when the child encountered complex challenges in performing a task.These findings support previous research showing that characteristics of the individual and of the task influence therapists' use of motor learning strategies (Kleynen et al., 2015;MacWilliam et al., 2021), which is considered important in interventions with children with DCD (Wilson et al., 2017).
However, more research is necessary to gain a better understanding of how child and task characteristics guide or should guide therapists' use of the instructions and feedback.
Few studies assessed therapists' use of motor learning strategies in children (MacWilliam et al., 2021;Niemeijer et al., 2006).This study expands previous studies since the use of instructions and feedback was explored more comprehensively.A strength of this study was that it used videotaped observations which provided the opportunity to review them repeatedly from different points of view (Pearce et al., 2014).Because existing observation tools score frequencies of predefined items, they provided insufficient insight into the full spectrum of instructions and feedback used, and into whether therapists adapted instructions and feedback (Levac et al., 2011;Niemeijer et al., 2003).Therefore, we developed this new comprehensive video-based analysis plan to investigate more exploratory research questions.Frequent discussions within the research group throughout the analysis process advanced the understanding of: (1) how instructions and feedback were shaped; (2) the implementation of implicit and explicit motor learning approaches; (3) interactions between instructions and feedback; and (4) the adaptation of motor learning strategies to child and task.With these insights, we were able to answer our research question.In order to investigate future research questions that explore whether characteristics of therapist, child, and/or task influence therapists' use of instructions and feedback, the results of our study can serve as a basis for an analysis plan developing predefined codes within the population of interest.A mixed-methods design can then be used to calculate frequencies if preferred.
This study also has some limitations.Firstly, the therapists selected which treatment session was videotaped and shared.It is possible that they chose to videotape and share a session in which they felt more competent in their use of motor learning strategies.Secondly, there is a possibility that the behavior of the therapist and/or child was influenced by the knowledge that they were recorded.In order to reduce this influence, the treatment session was videotaped by the parent or unmanned with the camera tripodmounted, because the presence of an unknown person (e.g., researcher) increases the risk for behavioral changes (Asan & Montague, 2014).

Conclusions
As a first step in developing practical recommendations on the use of instructions and feedback to enhance motor learning in children with DCD, this study explored their current use in videotaped treatment sessions.Therapists motivated the child to learn and used numerous specific instructions and feedback in a direct or indirect manner to provide the child with information to enhance task performance.In this study, it was observed that therapists used the parameters focus of attention, modality, information content, timing and frequency to shape instructions and feedback.They often combined multiple focuses and/or modalities, especially in more complex tasks.Furthermore, they changed the focus of attention, modality and information content frequently when a child encountered challenges in performing a task.It was observed that therapists used relatively little self-controlled timing and more specific instructions than feedback.Therefore, as a next step, interviews might gain more insights into therapists' clinical decision-making process regarding their use of instructions and feedback, and how characteristics of the child, task, and themselves may influence their choices.Future research should also focus on exploring whether the used instructions and feedback met the therapist's expectations of immediate improvement of performance using a thinkaloud procedure and investigating the effectiveness and success rates of instructions and feedback in children with DCD in a quantitative study.This study showed that instructions and feedback were frequently shaped by multiple focuses and modalities, which researchers should take into consideration when designing future studies.
Katrijn Klingels, PT, PhD is associate professor in Rehabilitation Sciences at University Hasselt and guest professor at KU Leuven, Belgium, specialized in pediatric rehabilitation.Her expertise lies in studying assessment and intensive treatment models of sensorimotor function from a clinical, biomechanical and neurological approach in pediatric populations.Current research projects focus on assessment and intensive intervention of upper limb function and postural balance, and motor learning paradigms in children with diverse disabilities.

Table 2 .
The tasks practiced in the segments selected for coding.
IF: internal focus; EF: external focus.