Forty-eight middle-income children, equally divided as to sex, looked at a sequence of animated films in which a pair of male figures underwent a change in facial affect. Three questions were posed: (1) Could a qualitative-structural analysis be applied to the kinds of explanations given by the children to account for changes in facial affect? (2) Is there a relationship between age and the kinds of explanations given? and (3) is there a relationship between a child's level of verbal skill and the nature or social complexity of the explanation offered? The verbal skill measure used was the Peabody Picture Vocabulary Test (PPVT). The results indicated that four basic categories of explanation emerged: (1) mere description, no explanation given; (2) impersonal events or situations cause expression change; (3) third person directly affects the dyad and thus causes facial expression change, and (4) interpersonal relations within the dyad account for expression change. Six- and eight-year olds gave significantly more category 1 responses, and the ten- and twelve-year olds gave significantly more category 4 explanations. Significant age effects were found for the PPVT; however, when the PPVT scores were covaried in a multivariate analysis of variance for age differences in category of explanation offered, significant age differences were still maintained on type of explanation given. Intercorrelations of PPVT and category scores were also undertaken with zero to moderate relationships the result, depending on the age group. No sex differences were found in any of the data. (Author)