TEMPORAL INTEGRATION AND SEGREGATION OF BRIEF VISUAL-STIMULI - PATTERNS OF CORRELATION IN TIME

Two brief sequential displays separated by a brief interstimulus inter val (ISI) are often perceived as a temporally integrated unitary confi guration. The probability of temporal integration can be decreased by increasing the ISI or (counterintuitively) by increasing stimulus dura tion. We tested three hypotheses of the relative contributions of stim ulus duration and ISI to the breakdown of temporal integration (the st orage, processing, and temporal correlation hypotheses). In the first of two experiments, stimulus duration and ISI were varied factorially, and estimates of temporal integration were obtained with a form-part integration task. The second experiment was a replication of the first at two levels of stimulus intensity. The outcomes were inconsistent w ith the storage and processing options, but confirmed predictions from the temporal correlation hypothesis. Whether two sequential stimuli a re perceived as temporally integrated or disjoint depends not on the a vailability of visible persistence, but on the emergence of a neural c ode that is based on the temporal correlation between the two visual r esponses.