The basic procedure
A neutral item, such as an unfamiliar face or brand, is repeatedly presented with a liked or disliked stimulus. Liking for the neutral item is then compared with a control. Levey and Martin described early demonstrations using pictures in the 1970s, and a 2010 meta-analysis by Hofmann and colleagues found a reliable average effect of modest size.
How it compares with other conditioning
Evaluative conditioning was once thought to differ from classical conditioning in important respects. Many researchers proposed it was resistant to extinction and did not require awareness. These claims have been challenged.
- Some studies find that effects fade or differ depending on the measure
- The meta-analysis found larger effects among people aware of the pairings
- Effects can arise from verbal instructions alone, suggesting a role for propositions and inference
Debates about mechanism
One view holds that pairings build simple mental associations automatically. Another, associated with Jan De Houwer and colleagues, argues that people form beliefs about how things are related. There are also methodological concerns, including demand effects, in which participants report what they assume the study is looking for, and the possibility that an item picked for pairing was not truly neutral. Because of these concerns, the practical size of the effect in everyday settings remains uncertain.
Further reading
- Levey, A. B., & Martin, I. (1975). Classical conditioning of human evaluative responses. Behaviour Research and Therapy, 13(4).
- De Houwer, J., Thomas, S., & Baeyens, F. (2001). Associative learning of likes and dislikes: A review of 25 years of research on human evaluative conditioning. Psychological Bulletin, 127(6).
- Hofmann, W., De Houwer, J., Perugini, M., Baeyens, F., & Crombez, G. (2010). Evaluative conditioning in humans: A meta-analysis. Psychological Bulletin, 136(3).