Trait concept formation and change as a function of the validity of behavioral information

A search of the literature revealed that there are very few studies which have investigated two topics: (a) trait concept formation and change; and (b) trait attribution as a function of the amount and type of behavioral information which is used as stimuli. The purpose of the present investigation is to study some parameters of these processes. First, two aspects of trait concept formation are studied: (a) the effect of differentially-valid behavioral information on subjects' ability to form trait concepts; and (b) the resistance to change of trait concepts which are formed by reading either high-or low-valid behavioral information. Second, two aspects of trait concept change are investigated: (a) the influence of differentially-valid behavioral information on trait concept adjustment; and (b) the influence of two within-trait concept manipulations on trait concept adjustment. One manipulation was the reversal of one trait dimension and the other was the replacement of a second trait dimension with an entirely new trait dimension. Third, two aspects of trait attribution are studied: (a) the effect of the validity of the behavioral information which was presented in the trait concept formation and the trait concept adjustment parts of the study; and (b) the effect of the amount and the consistency of the behavioral information. Each of 96 introductory psychology students formed one of several conjunctive trait concepts by reading either high- or low-valid behavioral information. Then subjects adjusted their trait concept to accommodate new and contradictory behavioral information of either high- or low-validity. Finally, after they had successfully adjusted their concepts, subjects gave their overall trait impressions. The behavioral information was obtained in three stages. In the first stage, 185 introductory psychology students generated one behavioral statement for each of eight trait terms which were used as stimuli. In the second stage, two major ratings were obtained from independent samples of introductory psychology students: (a) the probability of the trait given that the behavior was performed (113 introductory psychology students gave this rating); and (b) the probability of the trait given that the behavior was not performed (63 introductory psychology students gave this rating). For each behavioral item, a validity index (Fishbein and Ajzen, 1974) was obtained by subtracting the second rating from the first one. On the basis of the validity indices, two pools of behavioral statements (the high- or low-valid pools) were obtained and were used in the experimental part of the study. The principal results from analyses of variance are as follows: (a) trait concept formation did not vary significantly as a function of differentially valid behavioral information; (b) trait concepts which were formed by reading high-valid behavioral information were found to be more resistant to change than are those which were formed by reading low-valid behavioral information; (c) trait concept adjustment varied as a function of differentially-valid behavioral information but not as expected; (d) the reversed component of the trait concept was harder to adjust than was either the unchanged or the new component; (e) trait attribution on the reversed component of the adjusted trait concept did not vary significantly as a function of the validity of the behavioral information; and (f) trait attribution reflected the amount and the consistency of the behavioral information that subjects received throughout the entire study. The generalizability of some of these findings is limited because the contextual arrangement of the trait dimensions was found to influence both trait concept formation and change and trait attribution. A discussion of the results centered on: (a) the curvilinear hypothesis of Jones and Goethals ( 1972); (b) some content- and context-related effects on impression formation; (c) some aspects of trait attribution and implicit personality theory; (d) the summary view of traits (Wiggins, 1974); and (e) the Fishbein and Azjen (1974) approach to behavioral validity.Arts, Faculty ofPsychology, Department ofGraduat