Choice reaction times for temporal numerosity.

Subjects watched rapidly presented visual sequences of two events until they decided which event was the more frequent. Accuracy was reduced both by increased rates of presentation and by the addition of more inconclusive information to the initial portion of the sequence. Response latencies were longest when there was no objective numerosity inbalance, and they decreased, for both correct responses and errors, when discrimination was relatively easy. The patterns of decision latencies for individual subjects could be represented on a continuum defined by the sensitivity of the latencies to the rate of event presentation and to the objective inbalance between the two events. The discussion points out formal and empirical similarities between sequential numerosity judgments and choice reaction times. The present results are discussed within the framework of a first-passage time model, to be developed formally in a subsequent paper.