Behavioral microanalysis of spatial delayed alternation performance: rehearsal through overt behavior, and effects of...

Rats were trained in an operant spatial delayed alternation task utilizing retention intervals from 2 to 32 s. In addition to response accuracy, operations of the levers during the retention intervals were recorded and analyzed. Animals were tested following the administration of the muscarinic antagonists scopolamine hydrobromide and methylbromide, and the benzodiazepine receptor agonist chlordiazepoxide. In vehicle-treated animals, the relative number of correct responses and correct rehearsal operations (operation of the forthcoming correct lever during retention intervals) varied with the length of the retention intervals, and these measures were correlated. The response rate for rehearsal operations increased with the length of the retention intervals. It is speculated that the delay-dependent increase in response rate reflects an effect of delayed reward that was also associated with a delay-dependent increase in the tendency to alternate between levers. The effects of delay on the accuracy of rehearsal operations may have contributed to the delay-dependent correct responding. Scopolamine hydrobromide (0.01, 0.03, 0.1, 0.3 mg/kg) and methylbromide (0.1, 0.3 mg/kg) impaired correct responding, but did not seem to interfere with the relative number of correct rehearsal operations. As only the presentation of the panel light indicated trial onset, it is speculated that the cholinergic receptor blockade resulted in an increase in the probability of a repositioning response that was triggered by light onset. Chlordiazepoxide (1, 3, 5, 10 mg/kg) did not affect behavioral performance. These results suggest that in tasks that allow the development of rehearsal operations, delay-dependent response accuracy does not represent a sufficient condition for conclusions on task demands on memory.(ABSTRACT TRUNCATED AT 250 WORDS)