A laboratory experiment, using videotape, was designed to test the self-interest hypothesis, that is, that a subject will increase his effectiveness if he argues in favor of a position that is opposed to his best interests. Four hundred fifty-nine subjects participated in a Solomon Four-Group design for primary data acquisition. Strong opinion change main effects were obtained in support of the self-interest hypothesis, but not for source-audience similarity predictions. In terms of credibility and likability ratings, main effects were observed for source-audience similarity, and it was found that there were no significant interactions.(CH)