708 GAWRONSKI AND BODENHAUSEN Influence Figure&. n propos d b ing。i rrows reflect a lack of infl reflect a c as open arrows reflect a lack of influ doxical when decisional diss lici eshould affect only explicit but not im tion.As s n b PE regarded as highly due to an r consideration involve sfer of self-e The eighth and final mplex pattern of direct and indirect on both affect ning (see Gree d which.turn.m 1005 and Becker (in press)demonstr ated that simply ing to eate an a atio bet affect ass elf. object.Given that most p consistent with the licit 20N imply n evalu of the dy hy 2004).Pat ts were ted with pic ures of men whe as ch ional attitude changes may be due to a direct inf esults negative a evalua men who (EC)Ac to the APe model uch patte (2004)studie nO I given that molester des tha nism there is one would pot Case 7. The seventh involves a dir influen n as iative ce o (see 9).This ponding Influence 个↓ Even thoughh possible from a me atoria nvolve a rejection of associative evaluations as a valid basis fo Case 8-Direct inflpostdecisional dissonance should affect only explicit but not im￾plicit attitudes (see Case 3). In addition to cognitive dissonance, however, we previously noted that postdecisional attitude changes can also be due to an associative transfer of self-evaluations to objects that are associated with the self. This particular variant of EC may directly influence implicit attitudes without requiring processes of propositional reasoning (see also Greenwald & Ba￾naji, 1995). Consistent with this assumption, Gawronski, Boden￾hausen, and Becker (in press) demonstrated that simply choosing an object was sufficient to create an association between the object and the self, thus leading to an associative transfer of implicit self-evaluations to the chosen object. Given that most people’s implicit self-evaluation is highly positive (Bosson et al., 2000; Greenwald & Farnham, 2000; Koole et al., 2001), this process of associative self-anchoring can lead to postdecisional changes of implicit attitudes without requiring higher order propositional pro￾cesses (see also Lieberman, Ochsner, Gilbert, & Schacter, 2001). It is important to note that, because dissonance reduction and EC represent two independent mechanisms of attitude change, post￾decisional attitude changes may be due to a direct influence on evaluative judgments (cognitive dissonance) and an indirect influ￾ence on evaluative judgments that is mediated by associative evaluations (EC). According to the APE model, such patterns should emerge as long as there is no reason to discount the propositional implications of these associative evaluations. How￾ever, given that the EC mechanism produces an outcome that is fully consonant with the propositional dissonance reduction mech￾anism, there is very little reason why people would not rely on their automatic affective reactions as well as their propositional conclusions. Case 7. The seventh potential case involves a direct influence on associative evaluations and an additional indirect influence on associative evaluations mediated by processes of propositional reasoning (see Figure 9). This pattern implies corresponding changes in implicit and explicit attitudes, with changes in implicit attitudes being partially mediated by changes in explicit attitudes. Even though this case seems possible from a mere combinatorial perspective, it is highly unlikely from the perspective of the APE model. The particular mediation pattern implied in this case would involve a rejection of associative evaluations as a valid basis for evaluative judgments even though processes of propositional rea￾soning influence associative evaluations in the same direction. However, a rejection of associative evaluations seems quite para￾doxical when processes of propositional reasoning lead to exactly the same evaluation. As such, this case can be regarded as highly unlikely from the perspective of the APE model. Case 8. The eighth and final case under consideration involves a complex pattern of direct and indirect effects on both associative and propositional processes (see Figure 10). That is, a given factor may directly affect associative evaluations, which, in turn, may indirectly affect evaluative judgments. Moreover, the same factor may directly affect processes of propositional reasoning, which, in turn, may indirectly affect associative evaluations. According to the APE model, such complex patterns should emerge when (a) a given factor leads to changes in pattern activation or associative structure, (b) associative evaluations are consistent with the mo￾mentarily considered set of subjectively valid propositions, and (c) the same factor provides a basis for new propositions that directly imply an evaluation of the same valence. This case can be illustrated with a recent study by Castelli et al. (2004). Participants were first presented with pictures of men who were described either as child molesters or as child counselors and then completed a measure of associative evaluations for these individuals. Results indicated more negative associative evalua￾tions for men who had been described as child molesters as compared with men who had been described as child counselors. Even though Castelli et al.’s (2004) studies did not include a measure of explicit attitudes, it seems quite likely that the child molester description had two independent but synergetic effects on participants’ attitudes. First, because the category child molester can be assumed to have a clear negative associative valence, Figure 8. Case 6: Direct influence on propositional reasoning, with additional indirect influence mediated by associative evaluations, leading to corresponding changes in explicit and implicit attitudes. Solid arrows reflect a causal influence, whereas open arrows reflect a lack of influence. Figure 9. Case 7: Direct influence on associative evaluations, with ad￾ditional indirect influence mediated by propositional reasoning, leading to corresponding changes in explicit and implicit attitudes. Solid arrows reflect a causal influence, whereas open arrows reflect a lack of influence. Figure 10. Case 8: Direct influence on associative evaluations and prop￾ositional reasoning with mutual indirect influences, leading to correspond￾ing changes in explicit and implicit attitudes. Solid arrows reflect a causal influence. 708 GAWRONSKI AND BODENHAUSEN