Discrimination-reversal learning of normal and septal rats.

Abstract

The purpose of this study was to assess the complex learning skills of normal and septal rats using Rumbaugh's discrimination-reversal task. Transfer Index testing was carried out by using the automated test apparatus, stimulus items, and procedure described by Rumbaugh, et al. in 1972. Even though there were no significant differences between normal and septal rats, all rodents tested showed above chance performance on the transfer index task ( M = .82, range was .67 to .92). An analysis of errors in reversal indicated that the rats had more errors involving a single wrong response and fewer strings of consecutive errors. Also, the occurrence of “win-stay” and “lose-shift” response strategies for the four rats with the highest percentages correct on the second reversal trial (High Group) were compared with those of the four rats who had the lowest percentages correct on the second reversal trial (Low Group). The animals in the High Group demonstrated both “win-stay” and “lose-shift” response strategies, while those rats in the Low Group showed only a “lose-stay” response strategy. The rodents in the Low and High groups did not differ from each other in the use of a “win-shift” response strategy. These results suggest that rats can use rule or association learning to solve discrimination-reversal problems.