Abstract

A theoretical extension of the dual process model (e.g., elaboration likelihood model, ELM) of attitude change was attempted to include the negatively biased processing of strong attitudes. It was assumed that the default level of involvement and arousal (i.e., integral affect) was high in strong attitudes, which might hinder the objective processing of strong attitudes. Instead of ELM'S "high involvement → high elaboration" assumption, the extended model postulated a process of "medium involvement → high elaboration → objective processing." This implies that not only low involvement but also extremely high involvement may impede the objective processing. Thus, strong attitudes' resistance to change should be attributed to a highly motivated processing caused by a very high motivation to keep one's original position, rather than to an absent-minded peripheral processing assumed by ELM. Efforts were also made to show that the extended "three-level attitude change framework" could effectively explain various concepts related to strong attitudes such as cognitive elaboration, emotional commitment, motivation, and ego-involvement, as well as encompass the stereotype concept which might be considered as a kind of group-relevant strong attitudes. After the comparison of the new framework with the traditional social judgment theory, a final discussion on the measurement of the strengths of attitudes was added for empirical validation of the extended framework in future.