Research to date suggests there are three mechanisms by which child abuse may be transmitted from generation to generation: by teaching children that aggressive behavior is appropriate, by failing to teach children an internalized morality, and by transferring a particular self or interpersonal perspective from parent to child. Although these ideas are drawn from related research, few of the ideas have been tested for their relevance to abuse. They are presented to stimulate empirical investigation into the role of social cognitive correlates of abuse. In designing research to examine these hypotheses, it is important to recall that the abused person will be exposed to other discipline styles, alternative views on morality, and varied interpersonal treatment as he/she matures. While the importance of early experience as a mold for adult personality is not denied, the effects of early experience are not immutable; therefore, research should examine the circumstances that reinforce or alter the learning experiences of the home environment. Straus (1978) has indicated that the cross-generational transmission of abuse accounts for a small proportion of abuse cases. Given that many factors are probably responsible for the cross-generational tendency, it can be expected that the influence of any given factor will be observed in only a small percentage of cases. It would therefore be wise to test the contributions of various factors in analog studies, in addition to investigations with victims and perpetrators of abuse. Finally, the social cognitive perspective is merely one approach and should be supplemented with investigation from such areas as social learning and biological and sociological points of view. Three notes and 43 references are provided. (Author summary modified)