Discrimination learning performance under different first-trial procedures.

Abstract

Four groups of naive rhesus monkeys were tested on a series of six-trial object-quality discrimination problems. All animals were treated alike on trials 2 through 6 of each problem. The groups differed, however, in terms of what procedures prevailed on the first trial of each problem. The noncorrection procedure was used. The results showed the two groups trained to perseverate initial choices eventually attained a high percentage of correct responses on the second trials of the problems. Those groups which had to reverse a response continued to have difficulty throughout the experiment. Group differences in respect to response to correct and to incorrect initial choices were maintained in a series of 50 additional problems, although the groups were all treated alike.