An automated learning and memory model in mice: pharmacological and behavioral evaluation of an autoshaped response.

The purpose of the present experiments was to develop and validate pharmacologically an automated, relatively rapid, and reproducible behavioral model of learning and memory using an autoshaping procedure in mice. Nose-poke responses into a recessed area were differentiated by response-dependent reinforcement during two identical consecutive daily sessions. Performance during the first session was considered to be a measure of acquisition and that during the second session a measure of retention. Sensitivity to procedural manipulation, as well as an index of learning under these conditions, was demonstrated, for example, by a decrease in response rate when nose-poke responses did not produce a reinforcer. The sensitivity of the paradigm to pharmacological intervention was examined after drug administration before the first session. Scopolamine (0.1-10.0 mg/kg) had no effect on acquisition but caused a significant dose-related impairment of retention. Dizocilpine (0.01-1.0 mg/kg) impaired both acquisition and retention performance. 8-Hydroxy-2-(di-n-propylamino)tetralin (8-OH-DPAT; 0.1-1.0 mg/kg) disrupted behavior in general, but failed to have a selective effect on acquisition or retention. Linopirdine (0.1-1.0 mg/kg) showed only a weak enhancement of acquisition, whereas 4-aminopyridine (4-AP; 0.1-1.0 mg/kg) significantly facilitated acquisition. This paradigm offers the potential for a rapid, objective, and reliable indication of whether a drug will affect the acquisition or retention of a positively reinforced response in mice and could be a useful supplement to existing procedures.