Answers are grounded in this paper and cite their sources.
We studied synchronization with a metronome that smoothly changes tempo, from slow to fast (accelerando) or from fast to slow (ritardando). During the transition phase, systematic alternations of underadjustment and overadjustment of period and phase were observed. We analyzed the synchronization error (�asynchrony�) sequences in terms of two models that both assume linear period and phase correction mechanisms but differ in terms of how the timekeeper period is adjusted to the tempo change. In the interval-based model, period corrections are based on comparisons between timekeeper and metronome intervals, whereas in the asynchrony-based model, period corrections are based on the deviations of taps from metronome events. The qualitative data pattern is more compatible with an asynchrony-based model than with an interval-based model. Additional mechanisms that switch on and off period adjustment seem to be needed, however, for a quantitative fit of this model to our data.
No takes yet. Share an insight, caveat, or question.
Schulze et al. (2005) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: