Changes in dynamic balance control over time in children with and without Developmental Coodination Disorder

Changes in dynamic balance control over time in children with and without Developmental Coordination Disorder L.D. Jelsma1, B.C.M. Smits-Engelsman2 & R.H. Geuze1 1Clinical and Developmental Neuropsychology, University of Groningen, Grote Kruisstraat 2-1, 9712 TS Groningen, the Netherlands. l.d.jelsma@rug.nl; 2Faculty of Kinesiology and Rehabilitation Sciences, KU Leuven, Heverlee Belgium. Aim: Many studies report balance problems in children with Developmental Coordination Disorder (DCD). Dynamic balance, when evaluated as the control of Centre of Pressure (CoP), differs from typical developing to children with DCD. The latter group shows an increase of variability mainly during tasks under complex or changed circumstances. How children with DCD learn to control dynamic balance in dynamic conditions is unknown. This study examines changes over time in a task requiring high level of dynamic balance control (the Wii-Fit ski slalom game). Our hypotheses are 1) children with DCD will show larger and more variable CoP excursions than control children; 2) balance will improve after repetition as measured by a reduction of the variability of the Centre of Pressure (CoP); 3) differences in CoP excursions reflect differences in movement strategies. Method: Twenty eight children with DCD between the age of 6-12 year, participated in this study. Twenty one typically developing children (TD) matched for age and with a score >16th percentile on the Movement Assessment Battery for Children 2 (MABC2) were included as the control group. The Wii balance board was placed on an AMTI force plate. The children played the Wii Fit ski slalom game for ten consecutive runs before (T0) and after (T1) a period of six weeks. The force plate data were analysed for CoP variability and total path length. Results: The control group outperformed the DCD group with an average of 4.5 gates versus 10.5 gates missed. Preliminary analyses of the CoP reveal differences between groups at T0 on variability of the lateral CoP (F=7.19, p=.011) pathlength (F=10.35, p=<.01) and counts of change of direction in the anterior-posterior (F=15.47, p=<.001) and lateral (F=4.28, p=.047) CoP. Remarkably, the pathlength was longer in the TD group (M=.51, SD 15.8) compared to the DCD group (M=.35, SD 12.3). After six weeks (T1) both groups had further improved (controls missed 3.4 gates; DCD missed 8.8 gates) and pathlength had become near equal (TD: M=.46 (12.7) and DCD: M=.40, SD 11.5) (see left Figure). Repeated measures ANOVA revealed a difference between groups (F(5,28)=2.9, p=.03), with a significant interaction effect between group and time for the variability of lateral CoP (p=.048), the counted changes in anterior-posterior direction (p=.043), and a trend for pathlength (p=.066). Discussion: The TD group seems to gain efficiency over time in reducing path length; the BP group seems less active at first, gaining more efficiency over time by increasing path length. As the change of group differences at T0 and T1 does not match the change in performance (# missed gates) we conclude that inaccurate timing may be a better predictor of performance than control of dynamic balance during the first phase of learning. Keywords: Balance control; DCD; Force plate; Centre of Pressure (CoP)