On the relations among different measures of visible and informational persistence.

We report research designed to accomplish two goals. We first consider the question, raised by Coltheart (1980) and others, of whether three measures of visible and informational persistence--performance in temporally integrating two successively presented stimuli, subjective rating of the degree to which two successively presented stimuli appear to constitute a single or a dual temporal event, and partial-report performance--all measure the same underlying mental entity. We answer this question using a superset of dissociation logic called state-trace analysis (Bamber, 1979), and within the context of a systematic empirical foundation consisting of seven closely related experiments. Our second goal is to extend and apply a theory to data acquired from our seven experiments and also to data reported by other investigators. This theory, which has been confirmed in a variety of paradigms (see Busey & Loftus, 1994) assumes that (1) the initial stages of the visual system act as a low-pass linear filter which operates on a stimulus temporal waveform to produce a sensory response; (2) instantaneous rate of acquiring information from the stimulus is jointly proportional to sensory-response magnitude and proportion of as-yet-to-be-acquired stimulus information; (3) partial-report performance is determined by total amount of acquired information; (4) the probability that two events are perceived as contemporaneous is determined by the temporal correlation of their respective information-acquisition rate functions (which is similar to a suggestion by Dixon & Di Lollo, 1994); and (5) temporal integration is successful to the degree that the two temporal events are perceived as contemporaneous. This theory was highly successful in accounting for our and other investigators' temporal-integration and completeness-rating data, and was moderately successful in accounting for partial-report data. We discuss the degree to which our three persistence measures can be united within the context of our theory; we comment on the distinction between objective and subjective measures of visible persistence; and we address the decades-old question: "What is persistence good for?"