Temporal coordination between performing musicians

Abstract

Many common behaviours require people to coordinate the timing of their actions with the timing of others' actions. We examined whether representations of musicians' actions are activated in coperformers with whom they must coordinate their actions in time and whether coperformers simulate each other's actions using their own motor systems during temporal coordination. Pianists performed right-hand melodies along with simple or complex left-hand accompaniments produced by themselves or by another pianist. Individual performers' preferred performance rates were measured in solo performance of the right-hand melody. The complexity of the left-hand accompaniment influenced the temporal grouping structure of the right-hand melody in the same way when it was performed by the self or by the duet partner, providing some support for the action corepresentation hypothesis. In contrast, accompaniment complexity had little influence on temporal coordination measures (asynchronies and cross-correlations between parts). Temporal coordination measures were influenced by a priori similarities between partners' preferred rates; partners who had similar preferred rates in solo performance were better synchronized and showed mutual adaptation to each other's timing during duet performances. These findings extend previous findings of action corepresentation and action simulation to a task that requires precise temporal coordination of independent yet simultaneous actions.

Keywords:

• Temporal coordination
• Synchronization
• Joint action
• Action simulation
• Music performance

Acknowledgments

This research was supported in part by the Canada Research Chairs program and Natural Sciences and Engineering Research Council of Canada (NSERC) Grant 298173 to the second author. We thank Rowena Pillay and Melissa Trivisonno for assistance with data collection, and Peter Pfordresher, Fran Spidle, and three anonymous reviewers for helpful comments on an earlier version of this manuscript.

Notes

¹ To investigate whether the findings differed depending on which participant performed the left or right hand part, we compared the joint performance measures when the faster performer performed the right hand (and the slower performer performed the left hand) with when the faster performer performed the left hand. Mean overall asynchronies were near zero (M = –0.12, SE = 1.98) when the faster performer performed the right-hand melody, but were positive (M = 12.19; SE = 1.62) when the faster performer performed the left-hand accompaniment. This difference was significant in an ANOVA by faster performer's part (right hand, left hand) and left-hand accompaniment (simple, complex), F(1, 30) = 12.14, MSE = 199.79, p < .003. The left-hand lead that occurred in joint performances was thus driven by the faster performers playing the left-hand part. A similar effect was also evident in the mean final asynchrony (M = –2.00, SE = 2.10, and M = 11.50, SE = 1.71, when the faster performer performed the right- and left-hand parts, respectively, F(1, 30) = 13.93, MSE = 209.36, p < .002. However, the assignment of performer to part (right vs. left hand) did not affect the cross-correlation difference. Correlational analyses conducted separately for performances in which the faster participant performed the right-hand or the left-hand part yielded equivalent results; we therefore report the combined analysis.