Covert Rehearsal and Motor Long-Term Memory

Abstract

This was a preliminary study of the effects of covert rehearsal, i.e., mental practice, on retention of movement in motor long-term memory. Rehearsal has been recognized as one of the most important factors in experiments on human memory ( 1 ) . There have been many studies of rehearsal in motor short-term memory (2), however, most studies of motor long-term memory concerning the role of covert rehearsal have been atheoretical (3). If rehearsal is necessary for retention beyond the typical short-term memory period of a few minutes, it follows that preventing rehearsal should produce poorer retention of movement information than conditions in which rehearsal is allowed. In the present study 18 right-handed subjects participated in a study of a linear positioning task. In the first phase three movement lengths, 10, 15, 25 cm, were used. A movement length was assigned to one of three rehearsal conditions: immediate recall, recall after 20 sec. of rehearsal, recall after 20 sec. interpolated activity (backward counting by threes). Rehearsal intervals were counterbalanced between subjects, and subjects received both distance and location information cues. For each trial subjects moved the slide carriage from right to left until a stop was reached. Following the appropriate rehearsal interval the stop was removed and the subject attempted to reproduce the movement three times. This procedure was followed for all three rehearsal intervals. Following the completion of this phase subjects received a word-analogy test. After a 10-min. period subjects received a surprise recall test, in which they attempted to reproduce each movement three times in the same serial order as presented in the initial phase. Movement error was recorded in centimeters. Since the movement lengths were used simply to provide a repertoire of movements to remember, errors were collapsed over movement lengths. Errors were also collapsed over reproduction attempts within rehearsal condirions. Analysis of absolute error2 indicated there was no significant effect for rehearsal condirions in the initial phase. However, there was a significant effect for rehearsal in the second phase (F = 4.04, p < .05). The cell means were 4.5, 2.8, 4.9 cm for 0, 20, and 20 sec. filled, respectively. Subsequent analysis (Tukey HSD = 1.9) indicated a significant difference between 20 sec. filled and 20 sec. rehearsal but not between 0-sec. and 20-sec. rehearsal groups. The separate analysis was performed for each phase because recall in the first phase is sequential whereas recall in the second phase is serial. These findings provide preliminary evidence that covert rehearsal increases the retention of movement in motor long-term memory. The rehearsal condition (20 sec.) led to superlor retention of movement compared to the noor block-rehearsal conditions. The data suggest study of rehearsal in motor long-term memory is warranted.