More practice does not necessarily enhance transfer of learning: evidence and interpretations

The purpose of this study was to investigate what was learned and transferred of a criterion task using a transfer task and a self-paced procedure. Based on the results of two previous studies, five durations of practice were specifically selected to examine what can be transferred to a new task after different amounts of practice (40, 75, 100, 125, and 150 acquisition trials) on a criterion task. The criterion task required subjects to practice finding an 8-in. line with no fixed starting and ending points from left to right. The transfer task was different from the criterion task in two dimensions, distance (10-in. line) and direction (right to left). 50 volunteers were randomly assigned into 5 groups. Appropriate 5 x 2 (groups by blocks of practice) analyses of variance and a priori contrasts were conducted for all dependent variables (absolute error, variable error, movement time, KR-delay, post-KR interval, intertrial interval, interstimulus interval, and KR-delay by post-KR interval ratio) to examine the effects of terminal acquisition of the criterion task on the initial performance of the transfer task. Analysis indicated that, during practice, participants developed a functional interaction between performance characteristics and chronological profile of KR-delivery of the criterion task that was transferable to the transfer task. Also, practice beyond a certain point was detrimental to the ability to transfer what was learned.