You do not currently have permission to view the full text for this document. Please sign in to an account with an active subscription or contact support.
Dr. Stoller's study of the development of gender identity has benefited us in several ways. First, we are now much better informed about the problem itself and some of the conditions that bring it about. Second, his work testifies to the value of a research strategy that selects a particular problem, gathers empirical data about it, generalizes cautiously, interprets the findings in as parasimonious a manner as possible and, when necessary, reformulates theory to fit the new findings. Third, his findings are an important contribution to the movement within psychoanalysis that recognizes that personality development, whether healthy or pathological, cannot be understood except in terms of the interactions that occur between an individual and his family of rearing, and consequently that the study of these interactions is as necessary a part of psychoanalytic science as is study of intrapersonal processes.
In my own thinking about personality development and functioning (1969), I have found it useful to postulate that during childhood each of us builds up two working models, one of self and another of environment (especially of the significant people in it), on the basis of which we plan and
act.1 Information used in the construction of an individual's working models is derived from three sources: personal experience, parents, and others. Information from each source may be more or less valid and more or less compatible with information from other sources. A matter of much interest is to discover how influential the information from each source may be in comparison with that from other sources.
Stoller's findings show unmistakably that, in some cases at least, the information reaching a child from his parents can have great influence on the development of his self-model, even when the information is demonstrably false and incompatible with information from other sources. In addition, Stoller's findings support the hypothesis that a self-model, once developed, tends to