Measuring Behavior 2005: Benjuya and Wegman

Variable practice effect on bi-lateral transfer during a novel skill acquisition

N. Benjuya and E. Wegman

Kaye College of Education, Department of Physical Education, Beersheba, Israel

The ability to perform a motor skill bilaterally is rudimentary in many competitive sports. Bilateral transfer phenomenon is based on a motor control theory. According to the schema theory, action is controlled by a generalized motor program mechanism (GMP) and can be produced by different muscles on the left or right side. Another mechanism responsible for the selection of muscles is the recall schema, which can be enhanced by practice variability. Lai et al. (2000) suggested that a stable GMP developed earlier by constant practice is a prerequisite for developing, at some later time, an effective recall schema by means of variable practice. The purpose of the current study was to examine the effect of three practice models (constant, variable, and combined) on bilateral transfer during the acquisition of a specific judo skill in children. 36 children (28 male, 8 females; ages 8-10) were selected by the coach as unskilled in the execution of the O GOSHI, a hip throw in judo. The subjects were then divided randomly into 3 practice groups: 1) Constant practice (C); 2) Variable practice (V); and, 3) Mixed practice (M). Each group practiced the new skill in its own distinct practice modality, using the dominant arm and side. The coach taught the skill as follows:

To the C group, using an evenly paced rhythm (1-2-3), a constant ‘grab’ height, and a single, fixed direction for approaching the opponent and executing the follow-through motion.
To the V group, using a varied, random pace (1--2-3, 1-2--3, etc.), a high and low ‘grab’ height, and forward/backward/sideways directions for approaching the opponent and executing the follow-through motion.
To the M group, using C group conditions for the first 2 weeks of trials and the V group conditions for the second 2 weeks of trials.

Each group executed 120 practice trials over 4 weeks. Each group was tested three times: a pre-test before the practice period, a post-test after the practice period and, on the day of the post-test, a transfer test on the non dominant hand of the contra-lateral limb/side. The variables measured were: behavioral: time to completion of 6 throws in sequence (T), physiological: linear envelopes of the EMGs of the bi-lateral Rectus Abdominis (AB) and Rectus Femoris (RF). The transfer test revealed a significant positive effect with the V group on the T variable. All the groups showed significant improvement in the T variable from preto post-test conditions, as well. The transfer test also showed that, with regard to the EMGs, the amount of muscle activation per muscle at any given time was more organized: more consistent, more clustered. While acquiring the described novel skill, the V practice group achieved better bilateral transfer than the C or M group.

Paper presented at Measuring Behavior 2005 , 5^th International Conference on Methods and Techniques in Behavioral Research, 30 August - 2 September 2005, Wageningen, The Netherlands.