Intertrial association at the visual threshold as a function of intertrial interval.

Abstract

Two kinds of intertrial association are possible. In the first, the shifts in the probability of response are some function of a factor independent of the outcome of the preceding response or responses, while in the second, the shifts are some function of the preceding response or responses. The present experiment attempts to decide between these alternatives by determining the relationship between the amount of association and intertrial interval. A Hecht-Shlaer adaptometer was used to present the visual stimulus. Six subjects were run on a single brightness method over seven intertrial intervals, 2, 3,4, 5, 9, 12, and 30 sec. in a counterbalanced design. A positively accelerated, decreasing function was obtained, which appeared to be asymptotic to both the association and time axes. Several possible functions were inferred from the two alternative types of association and, on the basis of the results, it was concluded that intertrial association is a function of the outcome of the preceding trials, the effect decaying in time.