Effects of stimulus discriminability on the latency distribution of identification responses.

Abstract S s had to decide whether a briefly exposed point was on the left or right of the centre of a field. The distance of the point from the centre varied from trial to trial and S s performed under both Accuracy and Speed instructions. The response probability, mean latency, latency of the fastest response and slowest response, and the standard deviation, skew, kurtosis and kappa of the latency distribution for each response to each stimulus were calculated. As the point appeared nearer to the centre, response probability decreased, as did skew and kurtosis, while the other measures increased. The trends all carried over to the responses made erroneously to stimuli on the other side of the centre. The results were compared with predictions from a deadline model of stimulus identification (Swensson 1972), which was rejected, and with predictions from a number of incremental models of the process (Vickers, 1971). None of these provided an adequate fit to the data and attempts to resolve the differences were only partially successful. It is concluded that none of the models examined is acceptable.