Signal detection analysis of stimulus discrimination in normal and split-brain monkeys

Abstract

Five split-brain and 5 control monkeys were trained to perform a go, no-go discrimination between cutaneous stimulation at two arm locations, and subsequently retrained using the other arm. Cardiac and instrumental response latency data from various periods during training and retraining were analyzed in terms of signal detection theory, with the objective of determining the extent to which behavior is differentially controlled by the discriminative stimuli, uncontaminated by other factors affecting response bias. This analysis revealed that differential stimulus control was minimal early in training, and that its growth and final extent, as reflected by both cardiac and instrumental response latencies, were highly comparable between subjects. Differential stimulus control of both cardiac and instrumental responses reached maximum prior to the time each subject attained a conventional criterion of percentage correct responses; we attribute this discrepancy to the influence of response bias factors. At the beginning of retraining on the opposite body side, all control subjects and the two split-brain subjects trained on the left side and retrained on the right immediately displayed differential stimulus control of both cardiac and instrumental responses, whereas the other 3 split-brain subjects trained in the opposite sequence showed no transfer of differential stimulus control of either response. We conclude that this analysis revealed a functional asymmetry in the direction of information flow within the central nervous system which becomes apparent after callosal section.