Mere exposure effects for affectively valenced stimuli

Zajonc (1968) established that exposure to a neutral stimulus can increase liking for that stimulus, a phenomenon he labeled the mere exposure effect (MEE). Extensive subsequent research has shown the MEE to be robust for a wide variety of neutral stimuli. However, few studies have examined the MEE for stimuli with intrinsic positive or negative affective valence. Prior research in this area has been characterized by methodological difficulties, producing an inconsistent and inconclusive pattern of results.;The effects of mere exposure were examined in two experiments. Experiment 1 employed a standard MEE paradigm to examine whether an MEE would be found for affectively positive, negative, and neutral stimuli. Participants were pre-exposed to photographs of positive, negative and neutral faces. In a subsequent forced-choice test they were presented with pairs of positive, negative, and neutral faces, matched on affective valence; each pair included one pre-exposed face and one novel face. Participants were instructed to choose which face they preferred. An MEE was found for the positive faces and for the data from the negative and neutral faces combined. However, no MEE was found for either the negative or neutral faces when analyzed separately.;Because previous research has characterized the MEE for neutral stimuli as a robust phenomenon, a follow-up experiment was conducted to investigate why no effect was obtained for the neutral stimuli in Experiment 1. Experiment 2 replicated Experiment 1 using only the neutral stimuli. Results showed a significant MEE for these stimuli, suggesting that the affectively mixed context Experiment 1 may have decreased the MEE for the neutral stimuli.;These results suggest that affective valence of stimuli and affective context influence the MEE. The MEE is not restricted to neutral stimuli, and can add to existing positive affect. This suggests that the MEE has more ecological validity and generality than has previously been appreciated. While the mechanisms underlying the interaction between the MEE and affective valence remain unclear, these results are a starting point for further investigation of the interaction between stimulus affect and the MEE. Results are discussed with respect to existing theoretical models of the effect.