The influence of misinformation manipulations on evaluative conditioning.

We tested the influence of misinformation on evaluative conditioning (EC) by giving false information about the contingencies between CS and US stimuli after a conditioning procedure. This was done by asking participants questions about the USs while inaccurately suggesting that some CSs had been paired with a US that had the opposite valence than the US it had actually been paired with. For CS-US pairs from other conditions, accurate suggestions or no suggestions at all were given to participants. This manipulation significantly moderated EC effects. For pairs that were combined with inaccurate information we found a reversed EC effect, while we found a standard EC effect for both pairs combined with no suggestions and pairs combined with accurate suggestions. Additional analyses showed that the misinformation manipulation also moderated memory for the pairs. These results show that misinformation manipulations cannot only influence explicit memory but also attitudes. Furthermore, and in line with some theories of EC, they support the relevance of explicit memory for EC effects.