Black and white Ss were given recognition training for black and white faces. One experiment asked black and white Ss to describe black or white faces verbally, to recognize faces from verbal descriptions, and to describe similarities and differences in triads of faces. While verbal training did affect verbal usage and verbal recognition performance, there was no indication, either before or after training, of verbal usage or performance patterns which parallel visual recognition performance. Black Ss yielded higher recognition scores for black faces than for white; white Ss yielded higher recognition scores for white faces than for black. The second experiment trained white Ss for recognition of black or white faces by feedback trials, using a four-alternative forced-choice procedure. The initial difference in performance for white and black faces rN > B) was not present after 100 training trials. Results are related to an earlier study of face recognition by Malpass and Kravitz (1969).

The culture assimilator is a procedure designed to train persons from one culture to improve their interpersonal behavior with persons from another culture. The approach is based on the idea that interpersonal effectiveness is optimized when members of two cultures make similar attributions concerning the causes of a person's behavior. A laboratory test of an assimilator for training whites to interact with blacks is reported. A multimethod assessment of the effects of this training shows a number of the hypothesized changes in attributional tendencies and cognitions. However, these changes did not lead to differences on behavioral measures. Possible causes and implications of these findings are discussed.

Conducted a study of samples of blacks and whites, males and females, young and old, hardcore unemployed, and working- and middle-class individuals to investigate "ecosystem distrust." The term is defined as a distrust of people, things, and institutions in one's environment. The components include less trust in people, suspicion of the motives of others, rejection of authority figures and institutions of the establishment, and seeing the environment as malevolent. Consistency was obtained across a heterogeneous set of questions, suggesting the presence of ecosystem distrust among the black hard-core. The implications of these findings for training the hard-to-employ and their supervisors are discussed. (22 ref)

We investigated the attitudinal congeniality hypothesis (the assumption that people learn material congenial to their attitudes more easily than uncongenial material) in a 2 X 2 design: Instruction Set (learn vs. judge) X Essay Bias (pro vs. con), with attitude toward student activism as the focal ex post facto variable. Verbal skills, quantitative skills, and overlap of prior knowledge structure with essay content were treated as covariates. In addition, a number of variables related to quality of essay content and demand characteristics were controlled and/or measured to achieve the maximum possible control over recall variance. Results indicated that greater recall was associated with greater intellectual skills, greater overlap of prior knowledge, more positive attitudes toward the experimental setting, instructions to learn the essay, and the attitudinal congeniality effect (indexed by the Attitude X Essay Bias interaction). Interpretation is based on the effect of each variable on the perceived utility of the essay's content.

Fishbein's model was extended to female occupational choice. Normative beliefs and the attitudinal components of the model, as applied to female career choices, were expected to predict subjects' behavioral intentions better than would attitude alone. Eighty-eight sophomore women filled out and returned field survey questionnaires consisting of scales to measure the necessary components of the model. Additional measures were also taken. The normative beliefs component of the model was found to be the major predictor of behavioral intentions. The attitudinal component was relatively unimportant. Other findings were consistent with the model. Implications of the results are discussed.