Reducing Racial Prejudice by Verbal Operant Conditioning

Abstract

Free associations of adjectives to 10 pictures of black persons in various situations by two groups of 10 prejudiced white Southern student nurses in a verbal operant conditioning procedure was employed to reduce prejudice toward black people. Prejudice was measured by the California Ethnocentrism Scale, a 29-scale semantic differential of bipolar evaluative adjectives, and by a 5-min. sample of freely emitted adjectives. The 10 subjects in the two experimental groups and in the control group were first evaluated for prejudice in a group setting and then in the individual conditioning session, to determine their base level. During conditioning the two experimental groups were reinforced with “good” for every favorable adjective emitted, while the control group received none. As expected, positive adjectives increased and negative ones decreased, and scores on the semantic differential and the Ethnocentrism scale were consistent. Prejudiced subjects were ready to respond with a highly unfavorable attitude with reference peers and to give more favorable responses individually to the experimenter. After the individual conditioning of the positive concept, the tendency was to return to the original level of unfavorable responsiveness or to retain the original prejudiced attitude. This effect was called the “rho phenomenon” of concept change. The influence of social status and suggestibility on the “rho phenomenon” needs further research.