Contextual interference effect: manipulation of generalized motor programs and parameters in serial positioning tasks

Abstract

Contextual interference effect (CIE) is only observed when the task variations require different generalized motor programs (GMPs) (Magill & Hall, 1990). However, some studies found inconsistent results when verifying this hypothesis. The present study compared the performance of random and blocked groups that varied GMPs or parameters in a novel task that required both a new structure of movement and parameter value. Forty eight participants were randomly assigned to one of four groups: random GMP, random parameters, blocked GMP, and blocked parameters. The subject's task was transporting three tennis balls among six containers in the same box. The participants of GMP groups performed three different movement sequences in one target time (2,700 ms) during the acquisition phase. The subjects of parameters groups performed one movement sequence in three different target times (2,500 ms, 2,700 and 2,900 ms). Three and 10 minutes after the acquisition were applied the transfer and retention of transfer tests that required a new sequence and target time (2,300 ms). The test analysis indicated a lower level of absolute error for the random parameters group compared to the other ones. The results did not support Magill & Hall's hypothesis. The randomly parameter variations raised the best adaptation in a new context. A possible explanation is that type of practice manipulation created an optimal interference level in the learning of a novel task.