Type III error in research on interpersonal models of depression.

Reviews an interactional model of depression proposed by J. C. Coyne (see PA, Vols 56:2455 and 61:1146), empirical evaluations of this model, and related research. It is contended that laboratory studies designed to test these interactional models have examined stangers' reactions to depressives and therefore have little relevance to these models, which focus on the interactional processes between a depressed individual and his/her family and friends. It is argued that relevant empirical research is necessary before theoretical conjectures and treatment strategies based on an interactional model are accepted. (27 ref) (PsycINFO Database Record (c) 2012 APA, all rights reserved)