Abstract
The processing of a target event is facilitated if its timing is known, e.g. if it appears in phase with a preceding temporally regular input. This effect has been attributed to automatic entrainment of internal oscillators to the input frequency and phase (exogenous expectation). However, the timing of an event can also be expected based on voluntary shifting of attention to an explicitly memorized interval, and the involve- ment of such endogenous factors in facilitation following rhythmic input was not examined. Here, we presented a visual target following regularly flickering stimuli, and examined the facilitative effect of the rhythm (faster responses for target in phase vs. target out of phase) in three block types: 'exogenous', in which rhythm phase was not predictive of target timing; 'endogenous entrainment', in which rhythm phase validly predicted target timing; 'endogenous-exogenous conflict', in which subjects were endogenously cued to attend to a mem- orized interval, while rhythm phase was not predictive. We found a facilitative effect for the rhythm in the all conditions, with similar magnitude in the exogenous and conflict blocks, but larger than both in the entrainment block. Importantly, the facilitation effect in the conflict block was independent of the interval expected based on the color cue, or its validity effect. EEG analysis revealed that the CNV, a potential assumed to reflect expected interval, was driven by the rhythm, suggest- ing that the behavioral effect reflected directional bias of temporal expectation by the rhythm, and not general conflict. In a second experiment subjects did not perform endogenous entrainment blocks, thus eliminating the possibility that the exogenous effect in the conflict block results from task con- fusion. Yet, the results for the exogenous and the endogenous- exogenous conflict blocks were replicated. In conclusion, regular rhythms bias temporal expectation even in the pres- ence of voluntarily high-level expectation.
