Probability Learning by Aphasic Subjects

Nine aphasic and nine nonaphasic patients participated in a two-choice probability learning experiment in which they attempted to turn on a set of red “reinforcement” lights by pressing push buttons. During the first 50 trials, responses on one push button turned on the lights 70 times as often as responses on the other. During the next 50 trials, the reinforcement ratio was 5/1 in favor of the push button that first delivered maximum reinforcement. During the following five 50-trial blocks the ratio was changed successively to 2/1, 1/1, 1/2, 1/5, 1/70. Most aphasic subjects changed their response patterns to accord with the changing reinforcement ratios. Between-subject variability was greater for aphasic subjects than for nonaphasic subjects, and seven aphasic subjects exhibited “perseverative” response patterns in early reinforcement ratios. The performance of two aphasic subjects did not appear to be influenced by the reinforcement ratios in the first session. However, further observation and experimental treatment of these two subjects resulted in appropriate changes in their performance. The results of this study suggest that behavior-shaping techniques involving changing reinforcement schedules can be used in clinical treatment of aphasic patients.