We examined the relationship between endogenous rhythms, auditory and visual cues, and body movement in the temporal coordination of duet singers. Sixteen pairs of experienced vocalists sang a familiar melody in Solo and two Duet conditions. Vocalists sang together in Unison (simultaneously producing identical pitches) and Round Duet conditions (one vocalist, the Follower, producing pitches at an eight-tone delay from their partner, the Leader) while facing Inward (full visual cues) and Outward (reduced visual cues). Larger tempo differences in partners’ spontaneous (temporally unconstrained) Solo performances were associated with larger asynchrony in Duet performances, consistent with coupling predictions for oscillators with similar natural frequencies. Vocalists were slightly but consistently more synchronous in Duets when facing their partner (Inward) than when facing Outward; Unison and Round performances were equally synchronous. The greater difficulty of Rounds production was evidenced in vocalists’ slower performance rates and more variable head movements; Followers directed their head gaze away from their partner and used bobbing head movements to mark the musical beat. The strength of Followers’ head movements corresponded to the amount of tone onset asynchrony with their partners, indicating a strong association between timing and movement under increased attentional and working memory demands in music performance.

SENSORY INFORMATION AVAILABLE WHEN MUSICIANS'fingers arrive on instrument keys contributes to temporal accuracy in piano performance (Goebl & Palmer, 2008). The hypothesis that timing accuracy is related to sensory (tactile) information available at finger-key contact was extended to clarinetists' finger movements during key depressions and releases that, together with breathing, determine the timing of tone onsets. Skilled clarinetists performed melodies at different tempi in a synchronization task while their movements were recorded with motion capture. Finger accelerations indicated consistent kinematic landmarks when fingers made initial contact with or release from the key surface. Performances that contained more kinematic landmarks had reduced timing error. The magnitude of finger accelerations on key contact and release was positively correlated with increased temporal accuracy during the subsequent keystroke. These findings suggest that sensory information available at finger-key contact enhances the temporal accuracy of music performance. W HAT ROLE DOES TACTILE AND proprioceptive feedback play in the production of fast finger sequences typical of music performance? Instruments such as piano, strings, and wind instruments require fine control of finger movements in order to generate performances with high temporal precision and accuracy. The arrival of fingers on instrumental surfaces can provide important sensory information to guide performance. The precision of finger movements is attributed in part to the receptors distributed in the skin of the fingertips; these tactile sensory inputs permit high sensitivity and subsequent response to changes in force loads (Flanagan & Johansson, 2002). Tactile feedback may be more salient than proprioceptive sensations, due to the high density of mechanoreceptor innervation in the finger-tip skin and the large cortical areas involved in processing tactile information from the fingers (Jones, 1996).Although there are few studies of sensory feedback from finger-key contact in music performance, several studies of sensory feedback during tapping tasks indicate that tactile information plays an important role in the control of timing. Aschersleben and colleagues proposed the Sensory Accumulator Model (SAM) to explain how people synchronize finger tapping with an auditory pacing sequence, and how this synchronization is influenced by tactile information (Aschersleben, Gehrke, & Prinz, 2001). When participants tap in synchrony with an auditory pacing sequence, their taps precede the pacing sequence, termed a negative asynchrony (Aschersleben, 2002). According to SAM, the brain registers the time of a tap when afferent signals associated with the tap accumulate past a certain threshold, which is compared with the time at which the brain receives information from the metronome's onset. Based on the assumption that tactile information from the finger accumulates more slowly than auditory information from the metronome, participants must tap before the metronome's onset to ...

No abstract