How People (Fail to) Control the Influence of Affective Stimuli on Attitudes

People’s attitudes toward almost any stimulus (e.g., brands, people, food items) can change in line with the valence of co-occurring stimuli (e.g., images, messages, other people), a phenomenon known as the evaluative conditioning (EC) effect. Recent research has shown that EC effects are not always controlled, which is problematic in many circumstances (e.g., advertising, misinformation). We examined conditions under which uncontrolled EC effects are more likely to reflect retrieval failures or uncontrolled encoding processes. To provide an experimental test of this question, we propose that people can either integrate or add validity information to the stimulus valence. Specifically, we propose that controlled processes can integrate validity information into the stored valence representations mostly when validity information is provided at the time of exposure to the evaluative information. Control attempts taking place later are more likely to add than to integrate the validity information to the stored representation. Moreover, if validity information is merely added to the stimulus valence as compared to integrated, forgetting this information potentially inflates indices of uncontrolled processes. Our findings demonstrate important boundary conditions for the interpretation of measures of uncontrolled encoding processes. Nevertheless, they provide further evidence that uncontrolled encoding processes can contribute to EC effects. We discuss implications for theories of attitude change and for protection from misinformation.