Elsevier

Human Movement Science

Volume 27, Issue 2, April 2008, Pages 270-285
Human Movement Science

The link between motor impairment level and motor imagery ability in children with developmental coordination disorder

https://doi.org/10.1016/j.humov.2008.02.008Get rights and content

Abstract

The present study examined motor imagery ability in children with developmental coordination disorder (DCD). Specifically, it explored whether children with varying degrees of motor impairment differed in their ability to perform motor imagery tasks. Fourty-two children scoring below the 15th percentile on the Movement ABC were split into two groups – DCD severe (DCD-S), scoring on or below the 5th percentile, and DCD mild (DCD-M), scoring from the 6th to 15th percentiles – and compared to 21 age matched controls. Participants performed two motor imagery tasks – hand (performed without and with specific imagery instructions) and whole-body rotation. The results demonstrated that children in the DCD-S group had a generalized motor imagery deficit in that they were less accurate across tasks than controls (and the DCD-M group on the hand task) and showed little benefit when given specific imagery instructions. The DCD-M group appeared capable of performing simpler motor imagery transformations, but were less successful as task complexity increased. Unlike the DCD-S group, the DCD-M group did show some benefit from specific imagery instructions with increases in accuracy on the hand task. These results suggest that a motor imagery deficit does exist in many children with DCD but that its presentation can vary – factors such as the individual child’s level of motor impairment and task complexity appear to be linked to the profile of deficits observed, which could explain the inconsistent findings of previous studies. Although this study lends support to the theory that a deficit in internal modeling is an underlying problem for children with DCD, still more research is required to develop the theory further.

Introduction

The internal modeling deficit (IMD) hypothesis was originally formulated following a series of studies designed to explore visuospatial processing in children with developmental coordination disorder (DCD) (Katschmarsky et al., 2001, Maruff et al., 1999, Wilson and Maruff, 1999, Wilson et al., 2004, Wilson et al., 2001, Wilson et al., 1997). According to the IMD hypothesis, children with DCD have a deficit in their ability to accurately utilize internal models of motor control. Because internal models are hypothesized to provide stability to the motor system by predicting the outcome of movements before slow, sensori-motor feedback becomes available (Wolpert, 1997), such a deficit would severely reduce the motor control and learning capabilities of a child with DCD.

So far, the IMD hypothesis has been supported in studies using covert orienting of attention (Wilson et al., 1997), a double-step saccade task (Katschmarsky et al., 2001) and motor imagery (Maruff et al., 1999, Wilson et al., 2004, Wilson et al., 2001). Motor imagery provides an interesting avenue for the exploration of the IMD hypothesis due to the extensive range of paradigms that are available. Motor imagery refers to the mental simulation of any motor act without motor execution (Decety, 1996) and it shares many of the same neural and physiological components as its real counterpart, including increases in muscular activity, spinal excitability, autonomic system activity, and an increase in neural activity in areas associated with movement, such as the supplementary motor area (SMA), cerebellum, and basal ganglia (Jeannerod & Decety, 1995). Further, behavioral data show that in healthy populations, the time to perform an imagined movement corresponds closely to the time taken to actually perform that same movement (Parsons, 1994, Sirigu et al., 1996). These similarities between real and imagined movements have led researchers to suggest that motor imagery is actually the efference copy of a movement (or internal model), which has come to consciousness only because the actual movement has been inhibited (Crammond, 1997). Thus, motor imagery ability is thought to be reflective of one’s ability to accurately form internal models of motor control (e.g., Skoura, Papaxanthis, Vinter, & Pozzo, 2005).

Children with DCD have generally demonstrated atypical performance of motor imagery tasks when compared to age-matched controls, though this has not been unequivocal (Katschmarsky et al., 2001, Lust et al., 2006, Maruff et al., 1999, Williams et al., 2006, Wilson et al., 2004, Wilson et al., 2001). Studies using the visually guided pointing task (VGPT) have shown that most children with DCD do not conform to the Fitts’ law speed-accuracy trade-off typically seen in age-matched controls (Katschmarsky et al., 2001, Maruff et al., 1999, Wilson et al., 2001) and in adults elsewhere (Sirigu et al., 1996). Although group effects have generally been evident, individual analysis has demonstrated that not all children in the DCD group perform atypically when compared with controls (e.g., Wilson et al., 2001). Mental rotation studies have been less consistent than those using the VGPT. Children with DCD have displayed atypical response time (RT) patterns (Wilson et al., 2004) or significantly reduced accuracy (Williams et al., 2006) in some studies, whereas another reported no differences between children with DCD and age-matched controls for either RT or accuracy (Lust et al., 2006).

DCD groups in these previous studies generally scored below the 15th percentile on the Movement ABC (Henderson & Sugden, 1992), so performance differences across the studies might reflect the varying levels of motor impairment within the groups. When a DCD group was divided into those children who scored on or below the 5th percentile and those who scored between the 6th and 15th percentiles, there were notable differences in the performance profiles of the two subgroups on a motor imagery task (Williams et al., 2006). Small and unequal group sizes prevented further investigation, but this result suggested that the level of children’s motor impairment could be linked to their motor imagery ability.

Thus, the aim of the current study was to determine whether there were differences in the motor imagery performances of children with severe versus mild DCD and whether these performances differed from those of age-matched controls. It is important to know whether the type and severity of any deficits vary among children with DCD so that appropriate interventions can be developed. We have again utilized the mental rotation paradigm in this study, with children required to complete both hand and whole-body rotations, as in our previous study (Williams et al., 2006). These tasks are suggested to elicit the use of implicit motor imagery, either through the imagination of movements of the body in the hand task (see for example, Parsons & Fox, 1998) or through egocentric perspective transformations in the case of the whole-body task (Zacks, Ollinger, Sheridan, & Tversky, 2002). Young children, in particular, have been shown to rely heavily on motor processes when solving hand rotation tasks (Funk, Brugger, & Wilkening, 2005). In that study, a child’s hand posture at the time of the task had a significant impact on their performance, far more than that observed in the adult group, indicating that the children were implicitly trying to move their hands into the position of the hand on the screen – a form of motor imagery.

We previously suggested that children with DCD might not use this implicit motor imagery when solving a hand rotation task, given their atypical response time pattern (Wilson et al., 2004). Instead they may be treating the hands as objects rather than as body parts. As in Williams et al. (2006), we have addressed this issue by administering the hand rotation task twice. In the first presentation of the task, we provided instructions that only asked the children to decide which hand was presented on the screen. In the second presentation of the task, we instructed them to imagine their own hand in the position of the presented stimulus to assist in their decision making. We expected these explicit imagery instructions to reduce the likelihood of children in the study utilizing techniques other than motor imagery to perform the task, and investigated whether these instructions brought about any change in the performance profiles of the DCD groups.

In summary, the aim of the current study was to determine whether motor imagery performance was linked to a child’s motor impairment level. We expected that children with severe DCD would exhibit greater levels of impairment in their ability to perform motor imagery tasks than either children with mild DCD or age-matched controls. This might be evident in terms of an atypical RT pattern or decreased accuracy. It was further expected that children with mild DCD would also be impaired in their ability to perform motor imagery tasks when compared to controls, but that this impairment would not be as great as for those children with severe DCD.

Section snippets

Participants

Children between the ages of 7 and 11 years were selected using the same method adopted in previous studies and described below (Williams et al., 2006, Wilson et al., 2004). Teachers from two local standard primary schools, one of which had participated in a study conducted 12 months earlier, identified children in grades 2–6 whose motor coordination they believed to be below an age-appropriate level. These children were assessed using the Movement ABC, which was chosen due to its reliability

Response time – regression estimates

Table 1 shows mean regression estimates for RT on angle for each condition of the hand rotation task. For the HR-NI condition, the first multivariate planned contrast compared the combined DCD groups to the controls on slope, intercept, and r2. This revealed a significant multivariate effect, Wilks’ Λ = .832, F(3, 58) = 3.92, p = .013, η2 = .168, and a significant difference between the groups for intercept, F(1, 60) = 11.85, p = .001, η2 = .165. There were no significant differences for slope (p = .18), or r2 (p

Discussion

Dividing the DCD group into mild and severe DCD participants has enabled the identification of different motor imagery performance patterns among children with varying levels of motor impairment. The results provided qualified support for the existence of a motor imagery deficit in children with DCD, indicating that a deficit in the ability to accurately utilize internal models of motor control may be an underlying problem in some children with DCD. This discussion will focus first on the

Conclusion

This study showed that the motor imagery ability of children with DCD varied according to their level of motor impairment. Children with severe DCD were generally less accurate than age-matched controls on measures of motor imagery, whereas children with mild DCD were capable of performing simple imagery transformations, but were not as successful as these became more complex. The children with mild DCD also demonstrated some benefit from imagery instructions, but the severely motor impaired

References (37)

Cited by (91)

  • Combined action observation and motor imagery: An intervention to combat the neural and behavioural deficits associated with developmental coordination disorder

    2021, Neuroscience and Biobehavioral Reviews
    Citation Excerpt :

    To date, deficits have been reported in children with DCD on this task both when implicit MI is used (Barhoun et al., 2019; Wilson et al., 2004), and when explicit imagery instructions are provided (Williams et al., 2006, 2008). When explicit MI was instructed, Williams et al. (2008) reported that children at or below the 5th percentile on the MABC-2 performed significantly worse than both TD children and children between the 6th and 15th percentile. Similar findings were also reported for a whole-body mental rotation task (Williams et al., 2006, 2008).

View all citing articles on Scopus
View full text