Effective motor coordination is critical for the performance of functional movements underpinning physical activity and cardiorespiratory fitness in children and adolescents (4,12,20). It is well known that children with low motor coordination (LMC) score lower in cardiovascular endurance, balance, body composition, and movement time than their normally developing peers (12,20). Children with LMC also suffer from deficits in motor programming and attentional control (30,31), but much less is known about how deficits in how they mentally interact with the world underlay their physical deficits. Although studies have documented perceptual differences between LMC and “typically developing” children (8,19,23,37), these have been carried out using laboratory tasks where there has been no attempt to couple the child’s visual perception of the task environment with their movements as a real-world task is performed.
Gaze registration techniques provide an insight into how external visual information is used to guide and control goal-directed motor actions (17). Research has shown that children with impaired motor coordination use less effective gaze strategies in controlled laboratory reaction time (9), visual tracking (19), and cued reach-to-grasp (37) tasks. Although laboratory tasks provide strong internal control, they therefore provide limited transferability to the sort of dynamic, interceptive tasks relevant to sport and physical activity. For this reason, there has been a call from researchers to extend the gaze analysis paradigm to more ecologically valid, real-life tasks (19,37).
The extensive body of literature in the sporting domain that has revealed a perceptual–cognitive advantage for expert performers (21) provides a useful departure point for research examining motor coordination in children in more ecologically valid settings. Expert performers direct high acuity foveal vision to the right place at the right time to provide accurate and timely information to the neural systems controlling goal-directed movements (17,33). This strategy has been termed the quiet eye (QE), defined as the final fixation or tracking gaze to a target before the initiation of a planned motor response (32). The QE has been proposed to reflect a critical period of cognitive processing during which the parameters of a motor skill, such as force, direction, and velocity, are fine-tuned and programmed (21,32). A growing body of research has revealed that there are consistent expert–novices differences with respect to the timing and duration of QE in both self-paced far-aiming tasks and interceptive tasks (for reviews, see 21,33,36).
The current study seeks to be the first to translate the knowledge about proficiency-related differences in QE in adults, to children of varying levels of motor coordination ability. Children with poor motor skills struggle particularly with the high degree of accurate coordination required to effectively perform dynamic interceptive tasks (2,3). We have therefore chosen to examine a throwing and catching task, as it not only successfully differentiates motor ability but also is a critical component of many sports and playground games. We hypothesize that more coordinated children will have a superior visuomotor strategy on both the targeting and tracking phases of a throw-and-catch task than children with low coordination. Specifically, children with high motor coordination (HMC) ability will reveal earlier and longer targeting QE periods during the prethrow phase and earlier and longer tracking QE periods on the ball before the catch attempt compared with children with LMC ability. As this strategy will provide advanced target information by which to accurately plan the catching action, coordinated children should also make more successful catches.
Fifty-seven children (29 girls and 28 boys) were recruited from year 5 classes in two primary schools in the southwest of England (mean ± SD age = 10.4 ± 0.47 yr). Before commencing the study, ethical approval was gained from a local ethics committee, and informed written parental and participant consent was provided. Participants attended individually and were tested in a classroom provided for the duration of the research.
The Movement Assessment Battery for Children, Second Edition (MABC-2), was used to determine a motor coordination score for each participant (14). The test was designed to identify and describe impairments in the motor function of children and is one of the most used measures in both clinical and research settings (28). The MABC-2 consists of eight tasks designed for three age bands (3–6, 7–10, and 11–16 yr old), incorporating manual dexterity, aiming and catching, and balance elements. The child’s performance on each of these tasks (either a score for accuracy or completion time) are age adjusted and converted to standardized scores. An overall score is then computed, which can be converted to a population percentile to aid diagnosis (14).
Each participant was fitted with an Applied Science Laboratories Mobile Eye gaze registration system (ASL, Bedford, MA), which measures momentary point of gaze at 30 Hz. The system incorporates a pair of lightweight (78 g) glasses fitted with eye and scene cameras and a portable recording device—a modified digital videocassette recorder. Gaze data were collected wirelessly to digital videotape and an experimenter held the videocassette recorder behind the participant to ensure that relevant objects were within the field of view. All testing equipment was provided with the MABC-2 assessment pack, and standard testing procedures were followed.
Participants completed all eight tasks from the MABC-2 following the instructions outlined in the manual (14). The eye tracker was calibrated at the outset of the testing period and at the start of each new task, as the scene camera sometimes had to be adjusted to ensure that the field of view included the objects of interest. Although gaze data were collected for all tasks, we were most interested in the throwing and catching task (task 4): Not only is this interception skill highly relevant for sport and playground game participation, but specific predictions could be based on other targeting and interceptive tasks studied in sport (33). Participants stood behind a line marked 2 m from a blank wall and were instructed to throw a tennis ball against the wall and attempt to catch it cleanly in their hands. They were instructed to use only their hands to catch the ball (not to gather it against their chests) and to not allow it to bounce on the floor before it reached them. They were allowed to step forward to catch the ball once they had thrown it. The task was first explained to the participant by a researcher and then demonstrated once. Participants were then given five practice trials to reduce practice effects, before completing 10 experimental trials with the outcome of each being recorded (catch/no catch).
Catching performance was indexed by both an absolute score out of ten, expressed as a percentage (number caught cleanly × 100 / 10), and a standardized score—accounting for age differences—taken from tables in the MABC-2 manual (range 5–15) (14).
Ball Flight Times
Ball flight time (ms) was recorded as a proxy measure of how the throw and the catch were performed. Two specific phases were identified. 1) Ball flight 1 (throw: hand–wall) was defined as the time from ball release to wall contact and reflects the speed and trajectory of the throw (time “E” in Fig. 1). 2) Ball flight 2 (rebound: wall–hand) was defined as the time from wall contact until the ball was either caught, struck the participant’s body or another surface, or passed the initial throw line (time “F” in Fig. 1). Ball flight 2 is dependent on both the initial throw parameters, and the catching technique used, for example, how early the participant attempts to intercept the ball. Total ball flight time (hand–hand) was calculated by summing the two subcomponents. Ball flight times (ms) were recorded from the gaze registration system’s scene camera and analyzed in a frame-by-frame manner for each attempt.
The gaze data were downloaded from digital tapes to a computer (Lenovo Thinkpad R500) using the Eyevision software (ASL). The location and the duration of gaze were then analyzed in a frame-by-frame manner for each throw, using Quiet Eye Solutions vision-in-action software (www.QuietEyeSolutions.com). The QE is the final fixation or tracking gaze directed to a single location or object in the performance space within 3° of visual angle for a minimum of 100 ms (33). This generic definition is operationalized for each task in relation to three consistent components: its onset, offset, and duration (time from onset to offset). Earlier and longer QE periods are indicative of more expert-like performance, whether they are fixations to a stationary target or a tracking gaze on a moving object (33). Figure 1 provides a schematic representation of how the QE variables were operationally defined with respect to the key actions and outcomes of the throw and catch task. All trials where a QE onset and offset could be determined were included to help calculate a mean value for each participant, to be used in subsequent analyses (see Results section).
The onset of the targeting QE was defined as the start of the final fixation (within a 3° area on the wall) before the critical targeting action (33,36)—the release of the ball. As with other QE research for throwing tasks (32,38), the onset is reported relative to a standardized preparation phase—set at 2000 ms before the ball release (time “A” in Fig. 1). Offset occurred when the gaze deviated off the fixated location (by 3° or more) for more than three frames (100 ms). The targeting QE duration was therefore defined as the duration between targeting QE onset and offset (ms; time “B” in Fig. 1).
In interceptive tasks such as catching, pursuit tracking on the object occurs before the hands contact the ball (1,25,27). Tracking QE onset (ms) was the first gaze on the ball as it traveled toward the participant (time “C” in Fig. 1). Offset occurred when the gaze deviated off the ball by more than 3° for three frames (100 ms) as it traveled toward the participant, or when the trial ended (end of ball flight time 2). Tracking QE duration was defined as the duration between tracking QE onset and offset (ms; time “D” in Fig. 1). To control for differences in throwing and catching strategies, we also calculated tracking QE onset and duration relative to ball flight time 2 (time “F” in Fig. 1) (6,7). The relative QE measures were therefore calculated as: (QE × 100) / ball flight time 2.
The MABC-2 performance data were recorded using a standardized answer booklet and scored in accordance with the test protocol, including age corrections and standardization procedures (14). A tertiary split was then performed on the MABC-2 percentile scores for the sample, creating an HMC group, a median motor coordination (MMC) group, and an LMC group. One-way ANOVA (Statistical Package for the Social Sciences, version 19; SPSS Inc., Chicago, IL) were computed to compare differences in MABC-2 score, catching performance, ball flight, and QE measures between these three groups. Effect sizes were calculated using partial eta squared (ηp2) for omnibus comparisons, and LSD post hoc tests were used to interrogate significant main effects.
As significant group differences in the visuomotor variables of interest may be due in part to functional differences between “catch” and “no-catch” attempts (LMC will have fewer successful catches than HMC participants), we also ran the ANOVAs on caught trials only. Mediation analyses were finally computed to determine whether any QE measures mediated between-group differences in catching performance, using the MEDIATE SPSS custom dialog (13).This process determines the total, direct, and indirect effect of group on catching performance, through a series of proposed mediators, allowing inferences to be made about the indirect effects using percentile bootstrap confidence intervals.
The gaze data of some participants were of poor quality and could not be accurately coded. In order for a participant to be included in the analyses, a minimum criterion of 3 of 10 codable trials was set for each QE variable (see degrees of freedom for each analysis). A second analyst blindly scored 10% of the codable trials (one from each participant), and interrater reliability was assessed using the interobserver agreement method (29). This analysis revealed a satisfactory level of agreement at 92.5% (24).
Movement Assessment Battery for Children, Second Edition
Motor coordination ability varied across the sample of 57 children (mean ± SD MABC-2 percentile rank = 51.05 ± 26.38, range = 97.90). Four participants were classified as “highly likely” to have a clinical movement disorder (developmental coordination disorder [DCD]), scoring below the 5th percentile of a population norm (14). A further four children were found to be at risk of having DCD as they scored below the 16th percentile. At the high end of the range, two children scored at or above the 95th percentile, demonstrating excellent movement coordination, and a further 10 children scored at or above the 84th percentile.
A tertiary split of the sample population was performed based on MABC-2 percentile rankings. The LMC group contained 16 participants (6 females and 10 males) with an MABC-2 score of 64.06 ± 13.12 and a percentile rank of 18.76 ± 8.58 (mean ± SD). The MMC group contained 25 participants (10 males and 15 females) with an MABC-2 score of 79.24 ± 3.96 and a percentile rank of 50.52 ± 10.92 (mean ± SD). The HMC group was made up of 16 participants (8 females and 8 males) with an MABC-2 score of 91.13 ± 3.61 and a percentile rank of 84.19 ± 7.02 (mean ± SD). The ANOVA yielded a significant effect of group on MABC-2 score (F2,54 = 50.49, P < 0.001, ηp2 = 0.65) and percentile rank (F2,54 = 196.41, P < 0.001, ηp2 = 0.88). The LSD comparisons revealed significant differences in movement coordination score and percentile rank between all three groups (all P values < 0.001). Age was not significantly correlated with percentile rank (r = −0.16, P = 0.242) or MABC-2 score (r = −0.18, P = 0.182), and independent t-tests showed there was no significant difference between sexes in percentile rank (t55 = 0.93, P = 0.358) or MABC-2 score (t55 = 1.30, P = 0.200). The MABC-2 data are presented in Table 1.
ANOVA yielded a significant group difference in percentage number of balls caught (F2,54 = 18.78, P < 0.001, ηp2 = 0.41) and the standardized catching score (F2,54 = 16.46, P < 0.001, ηp2 = 0.38). LSD comparisons revealed that the HMC group performed significantly better than either the MMC or the LMC groups (all P values < 0.001), and the MMC group performed significantly better that the LMC group (P = 0.002 for balls caught and P = 0.048 for score). Age was not significantly correlated with catching performance, r = −0.15, P = 0.274, although boys were significantly better at catching than girls, t(55) = −2.33, P = 0.024. The catching performance data are presented in Table 1.
ANOVA revealed no significant group differences in ball flight 1 (throw; F2,45 = 2.06, P = 0.140), ball flight 2 (rebound; F2,45 = 0.44, P = 0.645), or total ball flight time (F2,45 = 1.58, P = 0.217). The ball flight data are presented in Table 2.
ANOVA revealed a significant difference in the time to targeting QE onset between the groups (F2,44 = 8.30, P = 0.001, ηp2 = 27). LSD comparisons demonstrated that the LMC group had significantly later onsets than both the MMC (P = 0.012) and the HMC (P < 0.001) groups. Although the MMC group also had a later onset than the HMC group, this difference only approached significance (P = 0.076). Targeting QE onset data are presented in Table 3.
There were no significant differences in the offset time (F2,44 = 2.19, P = 0.124), with all groups ending their fixation on the wall at around the point of ball release (see Table 3).
There were significant differences in the duration of the targeting QE period between groups (F2,44 = 10.12, P < 0.001, ηp2 = 32). LSD comparisons revealed that the LMC group had significantly shorter QE periods than both the MMC (P = 0.003) and the HMC (P < 0.001) groups. Although the MMC group also had a shorter QE period than the HMC group, this difference only approached significance (P = 0.098). Targeting QE duration data are presented in Table 3.
ANOVA yielded an almost significant group difference in the time to tracking QE onset (F2,40 = 3.10, P = 0.056, ηp2 = 0.13). As this finding approached significance, LSD analyses were carried out. These revealed that the effect was largely driven by the LMC group having significantly later onsets than the HMC group (P = 0.018). The effect of standardizing this time with respect to ball flight 2 (relative tracking QE onset) was negligible (F2,40 = 3.14, P = 0.054, ηp2 = 0.14). Tracking QE onset data are presented in Table 4.
The ANOVA on both the absolute tracking offset data (F2,40 = 2.85, P = 0.069, ηp2 = 0.13) and the relative tracking offset data (F2,40 = 2.68, P = 0.081, ηp2 = 0.12) also only approached significance. Again, this effect was driven by the significant differences in offset between LMC and HMC groups (P values = 0.022 and 0.027, respectively). Tracking QE offset data are presented in Table 4.
ANOVA revealed a significant difference for the duration of the tracking QE duration (F2,40 = 13.66, P < 0.001, ηp2 = 0.41). LSD comparisons showed that the LMC group had significantly shorter tracking durations than both the MMC (P = 0.005) and the HMC (P < 0.001) groups. The MMC group also had significantly shorter tracking durations than the HMC group (P = 0.007). When the tracking duration was standardized to account for ball flight time 2, the between-group ANOVA remained significant (F2,40 = 12.29, P < 0.001, ηp2 = 0.38). LSD comparisons remained significant between all groups. Tracking QE duration data are presented in Table 4.
Caught Trials Only
When only the trials that resulted in a catch were subjected to the same ANOVA as described previously for all codable trials, the significant main effects for targeting QE onset and duration and tracking QE duration (absolute and relative) remained but were reduced. Table 5 provides a detailed summary of the ball flight and QE data for caught trials only.
To check whether catching performance had been mediated by any of the gaze variables, group (coded as 1 = HMC, 2 = MMC, 3 = LMC) was entered as the independent variable, catching performance score as the dependent variable, and the significant QE measures from the ANOVA individually entered as mediators. Results from bootstrapping (based on 10,000 sampling rate) indicated that there was only a significant indirect effect for tracking QE duration (95% confidence interval = 2.80–24.00). When caught trials only were considered, no QE variables mediated the significant group performance differences.
This is the first study to measure the QE in children, providing a novel examination of processes underpinning differences in children’s motor coordination ability. The strength of the study was that it used an ecologically valid interception task (throwing and catching), that not only has relevance to sport and playground games but also has been shown to have predictive validity in many studies (28). We hypothesized that children with HMC ability would reveal a perceptual–cognitive advantage over less coordinated children. Specifically, we predicted that they would demonstrate earlier and longer targeting QE fixations (prethrow) and earlier and longer tracking QE gaze (precatch). We also performed additional mediation analyses to better understand which (if any) of these gaze differences mediates catching ability. This final step is seldom performed in the QE or motor expertise literature (24) and is necessary to avoid overinflating the importance of “matching” group effects across variables of interest.
There was a wide performance range across the eight MABC-2 tasks, and it was possible to classify three distinct groups of movement ability (Table 1). There were significant differences in catching ability between the three groups (Table 1) and significant differences between the low and HMC ability children in nearly all the QE measures (Tables 3 and 4). Interestingly, there were no significant differences in any of the ball flight measures, suggesting that the HMC group’s performance advantage was not (solely) due to differences in the way the task was performed (e.g., speed and trajectory of throw and position of catch; Table 2). Rather, this advantage was underpinned by differences in visuomotor control during both the prethrow and the precatch phase of the task.
In the prethrow phase of the task, HMC and MMC participants revealed earlier and longer targeting QE durations. Indeed, the HMC group’s QE duration was nearly twice as long as that of the LMC group (500 vs 260 ms; Table 3). This finding mirrors the research examining far aiming performance in adults, where the longer preparatory fixation is postulated to provide a quiet period to plan the ensuing motor response (36). As there was no specific target to throw to (unlike with most far aiming, targeting skills), it is interesting that ability-related differences in the QE duration were still evident, despite the lack of precision required. We propose that the HMC participants used this processing time to help to predict in advance where the ball would bounce and thus provide more time to track the ball’s final flight to their hands (18).
In the precatch phase of the task, there were differences between the HMC and the LMC participants in all the tracking QE measures examined (Table 4). The fact that the precatch gaze behaviors were more discriminatory is not surprising, given the increased precision required in this interceptive catching element, compared with the throwing element, of the overall task. Most notably, near-significant differences in both the onset and offset of the tracking gaze led to significant differences in the duration of the tracking QE between all three groups: 30%, 41%, and 51% of rebound time (flight time 2) for the LMC, MMC, and HMC groups, respectively. Not only was this the only process measure to reveal similar significant effects as the catching performance data between all three groups (Table 1), but also the formal tests of mediation also revealed tracking QE duration to be the only significant predictor of group differences in catching.
Encouragingly, the significant group differences in both targeting and tracking QE durations remained even when unsuccessful catching attempts were removed before running the ANOVA (Table 5). Not surprisingly, this reanalysis had the largest effect on the values of the LMC group, who caught the fewest attempts. They improved on nearly all the QE measures when catches only were considered, most notably increasing their relative QE duration by 10% (from 29% to 39%). This finding was perhaps unsurprising as studies with adult participants have also found significant intraindividual effects in QE, in addition to interindividual effects, with successful attempts categorized by longer QE than unsuccessful attempts (21,33,36). Although the significant effects in the ANOVA remained, the mediation analysis was no longer significant.
The question remains, why do the shorter QE periods of LMC children affect their performance in this way? The QE is postulated to provide the external spatial information needed by the brain (in conjunction with previous knowledge) to decide what it is going to do and how it is going to do it (33). In effect, the QE aids a prediction function in visuomotor control—helping performers to process novel transformations relating actions to their sensory consequences (10,11). The catching component of the task is simplified if a consistent relationship between the throw and the rebound can be established. We postulate that the HMC group’s superior prediction is assisted by the extended information processing time facilitated by the longer QE durations during both targeting and tracking. This postulation is supported by QE research in other interceptive tasks with adults, such as returning serve in volleyball (1) and table tennis (27) as well as shotgun shooting (6) and hockey goal tending (25).
Further support for this expectation is provided by research demonstrating that this ability to predict and calibrate movements based on sensory feedback may be impaired in children with movement coordination difficulties (2,3,16,23). Indeed, Wilmut and Wann (37) have demonstrated in a relatively abstract desktop task that children with DCD are slower in parameterizing a movement on the basis of predictive motion than typically developing children. In relation to the current study, LMC children may have greater difficulty in determining the consequences of using a particular level of force when throwing. We suggest that this may be not due to limitations in physiology and/or biomechanical characteristics but due to deficits in identifying relevant targets in space, allocating sufficient visual attention to that location to be successful, and predicting the consequences of the ensuing action. These children therefore base their catching movement on inaccurate cues, formulate inaccurate motor plans, and gain inappropriate feedback due to inhibited perception and sensory feedback—driven in the main by their shorter QE periods.
There are some caveats to the findings presented, which are reflective of limitations in the study design. Although using a standardized task with existing normative comparisons was a strength of the study, the MABC-2 protocol also added some constraints. First, the low number of trials meant that it was difficult to examine intraindividual differences in visuomotor control in participants who were at either end of the ability spectrum: 13 participants caught all 10 attempts, whereas 7 participants caught none. The power of the analyses was reduced when participants had to be omitted for having insufficient successful trials to analyze (Table 5). Second, the scoring system is rather imprecise and fails to distinguish between better and poorer attempts, where the result was still a failed catch. The imprecision of the dependent variable in the mediation analyses may therefore partially explain why more potential mediators were not found. Future studies could seek to apply more precise qualitative judgments of catching performance, which may be more sensitive to differences in visuomotor strategy (26). A third limitation of the study, reflected in the findings of the mediation analysis for the caught trials, was that other unconsidered variables are clearly important for the successful completion of the task. Although we found no differences in our proxy measure of how the task was performed (ball flight times), this is a rather crude measure. Future research should look to perform more detailed movement kinematic analyses of the participants during the task to further our understanding of the processes underpinning successful interception skill in children (2,3,22).
Although the results of this first study investigating the QE in children need to be replicated for other tasks, they suggest that children with high movement coordination are better able to predict ball flight for the interceptive task of throwing and catching a ball. This interpretation is supported by previous QE research in interceptive tasks with adults (1,6,25,27), and by research examining more abstract tasks in adults (10,11), and in children with DCD (2,3,16,37). The findings also suggest that interventions designed to improve such prediction may be useful to support children with conditions like DCD. Indeed, a systematic review of DCD interventions found that those focusing on supporting perceptual motor training displayed the most positive benefits (15). There may therefore be utility in designing QE training interventions for basic interceptive tasks like catching, which are important building blocks to increased physical activity. Previous research has supported the efficacy of such training interventions in other interceptive tasks with skilled adults (1,7) and for targeting tasks with novice performers (24,34,35). Although such interventions will need to be specifically tailored to the needs of children with motor coordination difficulties, there is evidence to suggest that QE training may have additional benefits for psychological constructs related to control and beliefs about success (39). It is recognized that children with DCD have lower beliefs about their ability to be successful in performing movement skills (5) and may therefore especially benefit from QE training.
To conclude, the current study was the first to examine the QE phenomenon in children and answers the call from researchers to examine the processes underpinning movement coordination difficulties in real-life tasks (19,37). Children with LMC ability demonstrated impaired visuomotor control and performance in a throwing and catching task, which were related to an inability to accurately track the ball as it rebounded off the wall. These results need to be replicated with other tasks, but there appears to be utility in exploring the application of QE training to populations outside of adult sport performers. Such interventions may help children with LMC to break the negative cycle linking low motor skill competence with low levels of physical activity and cardiorespiratory fitness.
The authors declare no conflict of interest.
The results of the present study do not constitute endorsement by the American College of Sports Medicine.
1. Adolphe RM, Vickers JN, LaPlante G. The effects of training visual attention on gaze behaviour and accuracy: a pilot study. Int J Sports Vis. 1997; 4: 28–33.
2. Astill S, Utley A. Two-handed catching
in children with developmental coordination disorder. Motor Contr. 2006; 10: 109–24.
3. Astill S, Utley A. Coupling of the reach and grasp phase during catching
in children with developmental coordination disorder. J Mot Behav. 2008; 40: 315–23.
4. Barnett LM, van Beurden E, Morgan PJ, Brooks LO, Beard JR. Does childhood motor skill proficiency predict adolescent fitness. Med Sci Sports Exerc. 2008; 40 (12): 2137–44.
5. Cairney J, Hay JA, Faught BE, Wade TJ, Corna L, Flouris A. Developmental coordination disorder, generalized self-efficacy toward physical activity, and participation in organized and free play activities. J Pediatr. 2005; 147: 515–20.
6. Causer J, Bennett SJ, Holmes PS, Janelle CM, Williams AM. Quiet eye duration and gun motion in elite shotgun shooting. Med Sci Sports Exerc. 2010; 42 (8): 1599–1608.
7. Causer J, Holmes PS, Williams AM. Quiet eye training in a visuomotor control
task. Med Sci Sports Exerc. 2011; 43 (6): 1042–9.
8. de Oliveira RF, Wann JP. Driving skills of young adults with developmental coordination disorder: regulating speed and coping with distraction. Res Dev Disabil. 2008; 32 (4): 1301–1308.
9. Emes CG, Vickers JN, Livingston L. Gaze control of children with high versus low motor proficiency. In: Yabe K, Kusano K, Nakata H, editors. Adapted Physical Activity. Tokyo: Springer-Verlag; 1994. p. 147–54.
10. Flanagan JR, Bowman MC, Johansson RS. Control strategies in object manipulation tasks. Curr Opin Neurobiol. 2006; 16: 650–9.
11. Flanagan JR, Vetter P, Johansson RS, Wolpert DM. Prediction precedes control in motor learning. Curr Biol. 2003; 13: 146–50.
12. Hands B. Changes in motor skill and fitness measures among children with high and low motor competence: a five-year longitudinal study. J Sci Med Sport. 2008; 11: 155–62.
13. Hayes AF, Preacher KJ. Statistical mediation analysis with a multicategorical independent variable. 2012 [cited 2012 Dec 10]. Available from: http://www.afhayes.com/public/mediate.pdf
14. Henderson SE, Sugden DA, Barnett AL. Movement Assessment Battery for Children—Second Edition (Movement ABC-2). London (UK): Pearson Education Inc; 2007. 125 p.
15. Hillier S. Intervention for children with developmental coordination disorder: a systematic review. Internet J Allied Health Sci Pract. 2007; 5 (3): 1–11.
16. Hyde C, Wilson PH. Dissecting online control in developmental coordination disorder: a kinematic analysis of double-step reaching. Brain Cogn. 2011; 75 (3): 232–41.
17. Land MF. Vision, eye movements, and natural behaviour. Visual Neurosci. 2009; 26: 51–62.
18. Land MF, McLeod P. From eye movements to actions: how batsmen hit the ball. Nat Neurosci. 2000; 3 (12): 1340–5.
19. Langaas T, Mon-Williams MA, Wann JP, Pascal E, Thompson C. Eye movements, prematurity and developmental co-ordination disorder. Vision Res. 1998; 38: 1817–26.
20. Lubans DR, Morgan PJ, Cliff DP, Barnett LM, Okely AD. Fundamental movement skills in children and adolescents. Sports Med. 2010; 40 (12): 1019–35.
21. Mann DTY, Williams AM, Ward P, Janelle CM. Perceptual–cognitive expertise in sport: a meta-analysis. J Sport Exerc Psychol. 2007; 29: 457–78.
22. Mazyn L, Savelsbergh GJP, Montagne G, Lenoir M. Planning and on-line control of catching
as a function of perceptual–motor constraints. Acta Psychol. 2007; 126: 59–78.
23. Mon-Williams M, Tresilian JR, Bell VE, Coppard VL, Nixdorf M, Carson RG. The preparation of reach-to-grasp movements in adults, children, and children with movement problems. Q J Exp Psychol. 2005; 58A (7): 1249–63.
24. Moore LJ, Vine SJ, Cooke A, Ring C, Wilson MR. Quiet eye training expedites motor learning and aids performance under heightened anxiety: the roles of response programming and external attention. Psychophysiology. 2012; 49: 1005–15.
25. Panchuk D, Vickers JN. Using spatial occlusion to explore the control strategies used in rapid interceptive actions: predictive or prospective control? J Sports Sci. 2009; 27 (12): 1249–60.
26. Przysucha EP, Maraj BKV. Movement coordination in ball catching
: comparison between boys with and without developmental coordination disorder. Res Q Exerc Sport. 2010; 81: 152–61.
27. Rodrigues ST, Vickers JN, Williams AM. Head, eye and arm coordination in table tennis. J Sports Sci. 2002; 20: 187–200.
28. Schulz J, Henderson SE, Sugden DA, Barnett AL. Structural validity of the Movement ABC-2 test: factor structure comparisons across three age groups. Res Dev Disabil. 2011; 32: 1361–9.
29. Thomas JR, Nelson JK, Silverman S. Research Methods in Physical Activity. 6th ed. Champaign (IL): Human Kinetics; 2011. p. 472.
30. Tsai CL. The effectiveness of exercise intervention on inhibitory control in children with developmental coordination disorder: using a visuospatial attention paradigm as a model. Res Dev Disabil. 2009; 30: 1268–80.
31. Tsai CL, Pan CY, Chang YK, Wang CH, Tseng KD. Deficits of visuospatial attention with reflexive orienting induced by eye-gazed cues in children with developmental coordination disorder in the lower extremities: an event-related potential study. Res Dev Disabil. 2010; 31 (3): 642–55.
32. Vickers JN. Visual control when aiming at a far target. J Exp Psychol Hum Percept Perform. 1996; 22 (2): 342–54.
33. Vickers JN. Perception, Cognition and Decision Making: The Quiet Eye in Action. Champaign: Human Kinetics; 2007. p. 280
34. Vine SJ, Wilson MR. Quiet eye training: effects on learning and performance under pressure. J Appl Sport Psychol. 2010; 22: 361–76.
35. Vine SJ, Wilson MR. The influence of quiet eye training and pressure on attention and visuo-motor control. Acta Psychol. 2011; 136 (3): 340–6.
36. Vine SJ, Moore LJ, Wilson MR. Quiet eye training: the acquisition, refinement and resilient performance of targeting skills. Eur J Sport Sci. doi:10.1080/17461391.2012.683815.
37. Wilmut K, Wann JP. The use of predictive information is impaired in the actions of children and young adults with developmental coordination disorder. Exp Brain Res. 2008; 191: 403–18.
38. Wilson MR, Vine SJ, Wood G. The influence of anxiety on visual attentional control in basketball free throw shooting. J Sport Exerc Psychol. 2009; 31: 152–68.
39. Wood G, Wilson MR (2012). Quiet-eye training, perceived control and performing under pressure. Psychol Sport Exerc. 2012; 13: 721–8.