When are support and opposition not opposites? Depth of processing as a moderator of the valence-framing effect

Prior research has demonstrated the valence-framing effect, in which leading people to frame a preference negatively (e.g., 'I oppose Romney') yields stronger attitudes than does leading people to frame that same preference positively (e.g., 'I support Obama'). Three studies tested whether or not depth of processing (as operationalized by manipulations of motivation and ability to cognitively process) moderate the effect. The valence-framing effect was replicated, such that opposers manifested stronger attitudes than did supporters, but only when attitudes were relevant to the participants (Experiments 1 and 3), and when participants were not under cognitive load (Experiment 2). Our results thus identify depth of processing as an important moderator of the valence-framing effect and provide potential insight into the effect's mechanism.