TEACHING SOCIAL LANGUAGE TO MODERATELY HANDICAPPED STUDENTS

Three students with moderate handicaps were taught to initiate and expand on conversational topics. The teaching procedure used stimuli generated from actual conversations with nonhandicapped peers. Generalization was assessed by audiotaping conversations between the handicapped students and their peers in natural school contexts without adult supervision. Results indicated that training generalized to natural contexts. These results were socially validated by undergraduate special education students, who rated tapes of two of the students' conversations during training phases as more socially competent than during baseline. Results are discussed in terms of the evaluation of complex social behavior as multioperant behaviors.     

Of beasties and butterflies: Evidence for the stability and domain-specificity of individual differences in categorization

Previous research has failed to demonstrate unambiguously the relative stability across time and consistency across tasks of individual differences in categorization, or categorization breadth The present study of categorizing behavior in college students assesses breadth at each of four points in time over a six-week period Results are consistent with an interpretation of breadth as a highly stable individual difference variable, but one that is constrained by the nature of the stimulus sets employed High stability coefficients were observed for each of two redundant tasks In contrast, cross-task consistency indices were generally low Personality correlates of breadth were similarly weak and inconsistent Several variables are suggested that may attenuate the generality of categorization breadth The existence of multiple styles of categorization is discussed     

Behavioral genetic methods

Traditional behavioral genetic methods involve the use primarily of family, twin, and adoption correlations to estimate the relative contributions of genetic and environmental influences in the etiology of individual differences. These methods and representative results for personality are described. However, newer methods are emphasized: structural models and model-fitting, multivariate analysis, genetic change and continuity in development, shared and non-shared components of environmental variance, and genetic components of "environmental" variation. Because most applications of these behavioral genetic methods to the study of personality involve self-report omnibus questionnaires, an important direction for future research in this area is to use these methods to explore new issues and new measures that have emerged from personality theory and research during the past decade.