Assessment of cognitive self-statements during marital problem solving: A comparison of two methods

Twenty maritally distressed couples (DC) and 20 nondistressed couples (NDC) were recruited and asked to undertake 10 minutes of problem-solving discussions, which were videotaped. Each individual partner's cognitive self-statements during the interaction were assessed using two methods: videoassisted recall (VR) and thought listing (TL). Reported cognitions from each method were content-analyzed and classified into five categories: partnerreferent positive, partner-referent negative, self-referent positive, self-referent negative, and other. Proportions of reported cognitions falling into each category were analyzed in two separate two-way MANOVAs (marital distress/nondistress ×sex) for the VR and TL measures. Results of each MANOVA indicated a highly significant effect of marital distress on cognitions, and a significant effect of sex on the VR but not the TL measure. Discriminant analyses showed that the VR and TL methods both discriminated between DC and NDC groups. Post hoc univariate ANOVAs indicated that DC had significantly higher proportions of negative partner-referent cognitions, and lower proportions of positive partner-referent cognitions, than NDC while problem solving. The relative merits of each cognitive assessment method, and their potential use in increasing marital therapy effectiveness, are discussed.