Coincidence detectors and two-pulse visual temporal integration: new theoretical results and comparison with data

Abstract

Exact predictions for two-pulse visual temporal integration data are derived from the Bouman-van der Velden quantum coincidence model for threshold vision. The predictions of the model start with complete summation for superposed pulses, then pass to a transition zone of partial integration, and finally reach the level of probability summation for pulses presented with large interstimulus intervals. From our results we can clearly reject the assumption of constant integration times with the basic model. We thus generalize the coincidence model to allow for variable integration times, derive the corresponding predictions for two-pulse integration data, and compare these predictions to published data currently available. It is shown that detectors of low order of coincidence generally underestimate the actual reduction of threshold intensity (or equivalently the corresponding increase of the detection probability) for two pulses as compared to the simple-pulse performance.

On the relationship between the redundant signals effect and temporal order judgments: Parametric data and a new model.

“…The principal inferential logic behind the time-course paradigm has in fact been used in many classical psychophysical studies. For example, the so-called two-pulse paradigm to study visual temporal integration (cf., Bouman & van den Brink, 1952; for review, see Schwarz, 1993) differs from the RSE paradigm essentially simply in using detection probability, not RT, as a dependent variable.…”

Better to take one long or two quick looks? Independent processing of evidence from discrete visual presentations.

“…Such models generally predict Bloch’s law: The summation of two neural counting processes of half duration ( t /2) each is equivalent to having one uninterrupted counting process of duration t . More advanced “leaky” integration models capable of dealing with partial summation (e.g., Schwarz, 1993a) generally predict equal or better performance for a single, uninterrupted presentation of duration t , relative to two separated presentations of duration t /2 each. Thus, standard integration models predict either no Presentation × SOA interaction at all or else they predict better performance for uninterrupted single presentations.…”

“…Another line of research relevant in the present context is related to studies of the integrative capacity of the visual system. For example, in a classical psychophysical paradigm the task is to detect either a single pulse of 20 ms or two separated pulses of 10 ms each (e.g., Bouman & van den Brink, 1952; for review, see Schwarz, 1993a). These studies are similar in spirit to the general design described above in that for a given task the total presentation time of the stimuli is fixed: Uninterrupted pulses are systematically compared with separate pulses of equal overall duration.…”