Operant behavioral analysis of memory loss in monkeys with prefrontal lesions.

The effects of dorsolateral prefrontal cortical ablation were examined in rhesus monkeys preoperatively trained to perform two versions of an indirect spatial delayed-response task. In one task (DR I), the stimulus keys that signaled the position of the reward were accessible throughout a trial. Thus, the monkeys could orient to the positive side ('rehearse') by continuing to press the correct key during the delay. The other task, DR II, was designed to disrupt bodily orientation and to prevent 'rehearsal'. In this task the monkeys were required to press an extraneous upper key during the delay. After bilateral ablation of the prefrontal cortex, 3 out of the 4 monkeys easily regained preoperative levels of performance on the DR I task, but all subjects were markedly impaired in the DR II task on which they exhibited strong positive preferences. These deficits were ameliorated by eliminating the position biases through training with short delay intervals followed by gradual increments in delay length. Analysis of the location of key presses during each phase of delayed-response trials showed that the monkeys press the correct key in the cue period even when they subsequently made incorrect responses. A consistent finding was that operated monkeys shifted their response to the incorrect key early in the delay period regardless of its duration, whenever the correct side was the nonpreferred side. These results support the idea that delayed-response deficits in monkeys with prefrontal lesions may be due in part to a defect in registering or encoding the correct spatial position in memory rather than to a loss in storage or retrieval processes.