Flexibility of temporal expectations for triple subdivision of a beat.
Bottom Line: The present study asked whether latent expectancies at 1/3 and 2/3 of the IBI can be induced by a global experimental context of triple subdivision, and whether a local context of consistently phase-shifted triple subdivisions can induce different expectancies.These results suggest that temporal referents between beats, which typically are linked to simple ratios of time spans, are flexible and context-dependent.In addition, we show that the PCR, a response to expectancy violation, is independent of and sometimes contrary to the simultaneous phase adaptation required by a change in subdivision timing.
Affiliation: Haskins Laboratories, New Haven, Connecticut.
When tapping in synchrony with an isochronous sequence of beats, participants respond automatically to an unexpectedly early or late beat by shifting their next tap; this is termed the phase correction response (PCR). A PCR has also been observed in response to unexpected perturbations of metrical subdivisions of a beat, which suggests that participants have temporal expectancies for subdivisions to occur at particular time points. It has been demonstrated that a latent temporal expectancy at 1/2 of the inter-beat interval (IBI) exists even in the absence of explicit duple subdivision in previous IBIs of a sequence. The present study asked whether latent expectancies at 1/3 and 2/3 of the IBI can be induced by a global experimental context of triple subdivision, and whether a local context of consistently phase-shifted triple subdivisions can induce different expectancies. Using the PCR as the dependent variable, we find weak evidence for latent expectancies but strong evidence for context-induced shifts in expectancies. These results suggest that temporal referents between beats, which typically are linked to simple ratios of time spans, are flexible and context-dependent. In addition, we show that the PCR, a response to expectancy violation, is independent of and sometimes contrary to the simultaneous phase adaptation required by a change in subdivision timing.
No MeSH data available.
Related in: MedlinePlus
Mentions: Figure 8 shows all the S12 conditions, including the ones with an empty A-pattern. In three of the conditions the A- and B-patterns are the same, so there is neither a PCR nor phase adaptation. In four conditions (A on time, B early; A late, B early; A on time, B late; A early, B late), the PCR is clearly distinct from the phase adaptation, going in the opposite direction. In the conditions with empty A, there is a clear negative PCR that deviates from the rather minimal phase adaptation in two cases (B early, B on time) and seems to form part of a large phase adaptation in the third case (B late). The remaining two conditions (A early, B on time; A late, B on time) show a different pattern: There is no PCR, only a rapid phase adaptation at a delay of one tap (indistinguishable from a delayed PCR). Thus, it seems that an on-time B-pattern elicited a PCR only when the A-pattern was empty (cf. Figure 4), which suggests maintenance of on-time expectations for S12 in the face of a phase-shifted A-pattern. However, there is another way of interpreting these data. Suppose the phase adaptation was not delayed (and why should it be?) but started with Tap 11. A conservative estimate of the phase adaptation on Tap 11 could be obtained by interpolating between the asynchronies of taps 10 and 12; if phase adaptation were immediate, that would make the argument only stronger. Viewed against this predicted asynchrony, the actual asynchrony deviates in the direction the PCR would have been expected to go (i.e., positive for A early, B on time; negative for A late, B on time). Thus the apparent absence of a PCR can be understood as resulting from the cancellation of the PCR by simultaneous phase adaptation in the opposite direction. It need not be concluded, therefore, that participants’ expectations were not changed by shifted subdivisions in some conditions.
No MeSH data available.