The purposes of this study were to investigate the effects of variability of training on the acquisition of motor skill of closed loop type tracking task using Rotary Pursuit, and to determine if there was a bilateral transfer effect to the non-dominant hand following practice with the dominant hand. Twelve healthy volunteer students (5 males and 7 females, aged 25 to 37) were randomly divided into a constant practice group and a variable practice group. A photoelectric rotary pursuit apparatus with stop clock and repeat cycle timer by Lafayette Instrumentation Co. was used for this study. Rotary pursuit is a closed loop task in which a subject attempts to keep a photoelectric stylus on a lighted target in motion. Subjects performed the clockwise circular pursuit task while standing. Experimental procedure was divided into three sessions, namely, pre-test, training, and post-test. The constant group practiced all 60 trials at 30 rpm. Variable practice group did a varied practice session with 15 trials at speeds of 20 rpm, 26 rpm, 34 rpm, and 46 rpm. No one in either group practiced with their non-dominant arm. A Mann-Whitney test and a Wilcoxon Signed Ranks test were used for statistical analyses. The results of this study showed no different training effect between groups on the post-test with the dominant hand. However, bilateral transfer effect of rotary pursuit task between hands was demonstrated. Possible mechanisms are discussed.

This study investigated the effects of different sequence structures on the bilateral transfer of learning from one sequence to another. Subjects were asked to respond to visual stimulus of "x" mark appeared in a particular sequence on the CRT at one of four locations arranged horizontally. They were instructed to respond to the visual signal by pressing the corresponding keys with their four fingers. After practicing one serial pattern for a substantial period of time with one hand, the subjects were randomly assigned to one of two experimental conditions and a control condition [ hierarchy (n=6) , vs , linear (n=6) , control (n=6) ] . Reaction time and latency profile constituted the dependent variables and learning of the sequence was evaluated by measuring the amount of facilitation of the performance on the transfer tasks. The findings support the notion that the internal representation of a sequence organiges in hierarchically, rather than linearly and the executionin of movement sequence is controlled hieararchically even though different effector(different hand in this study) . In addition, the results provided strong evidence for the notion that the subunits of movement can be interchangeable once subjects establish interval structure of the sequence.