Effects of practice variability and distribuion of practice on musicians' performance of a procedural skill



Journal Title

Journal ISSN

Volume Title



I designed three experiments to determine how procedural memory consolidation in a music task is affected by practice under different conditions of speed regulation and different time intervals between practice sessions. Ninety-two nonpianist musicians practiced a 9-note sequence with their nondominant hand on a digital piano in three sessions, each of which comprised 3 blocks of 15 performance trials. In Experiment 1 (n= 31), participants were instructed to perform as quickly and accurately as possible but determined their own tempos in each trial. In Experiment 2 (n = 31), three defined practice tempos (M. M. = 52, 72, and 92) were externally regulated in a stable practice procedure in which tempo changed between, not within, blocks. In Experiment 3 (n =30), the same three tempos were externally regulated in a variable practice procedure in which practice tempo changed from trial to trial within each block. In each experiment, three different groups' practice sessions were separated by either 5 min, 6 hr, or 24 hr. Consistent with previous descriptions of procedural memory consolidation, the results of Experiment 1 show that note accuracy improved significantly between Sessions 1 and 2 only when the sessions were separated by a 24-hr interval that included sleep; performance speed improved in all groups between Sessions 1 and 2, and between Sessions 2 and 3 when sessions were separated by 6 or 24 hr. In Experiment 2 (stable practice) there were significant improvements in note and tempo accuracy between Sessions 1 and 2 when those sessions were separated by 5 min or 6 hr, but not when the sessions were separated by 24 hr. In Experiment 3 (variable practice), note accuracy improved between Sessions 1 and 2 only when the sessions were separated by a 24-hour interval that included sleep; there were no significant improvements in tempo accuracy, perhaps due to the high physical demands of matching varying target tempos in successive trials. These results demonstrate that motor skill learning in music is affected by the time interval between practice sessions, and that the effects of distributed practice are dependent upon practice conditions.