Abstract

Psychotherapy asks clients to change the way they think about their difficulties, including attributions of causes. Sometimes clients come to surprising, even ill-founded, conclusions. What kind of reasoning do people use in arriving at beliefs concerning the causes of personal problems? The reasoning literature suggests there are many weaknesses in human reasoning in other domains (e.g., social and physical). The literature suggests skill in reasoning is acquired slowly and as a result of repeated engagement in reasoning challenges. Yet, even when skill improves, poorer quality reasoning strategies remain in the reasoner's repertoire and are sometimes used. An individual's reasoning performance is therefore likely to fluctuate with changes in the demands of the reasoning task and the cues available in the situation. The present study examined (a) the informal reasoning skills demonstrated by participants when talking about the causes of personal problems and (b) the effect of two different therapeutic approaches on type and quality of reasoning processes. To test the effect of two different styles of therapy 90 participants were randomly assigned to one of 3 groups: a control group, a cognitive group, or a childhood group. The control group received no intervention, the cognitive group received an analogue of a cognitive therapy session, and the childhood group received an intervention designed to encourage the participant to instantiate the theory that a current personal problem was caused by a pattern of experience that began in childhood. Following the intervention, argumentive and epistemological reasoning were elicited from all participants during semi-structured interviews modeled on interviews used by Kuhn (1991). Interview protocols were coded for degree of success in the skills of argument (developing theory, finding supporting evidence, developing opposing theories, finding counterevidence, and rebutting counterarguments) and for epistemology. Reasoning about the causes of personal problems showed many of the characteristics of reasoning in other domains. However, more participants than in other studies expressed a high degree of confidence in their theories. Participants often used unsuccessful or poor quality evidence to support their theories. Type of analogue therapeutic intervention was nonsignificantly related to argumentive reasoning quality.