Once in contact always in contact: Evaluative conditioning is resistant to extinction

The present study aimed at obtaining some further support for the hypothesis of a distinction between two basically different kinds of learning in a Pavlovian conditioning preparation: signal-learning and affective-evaluative learning (Baeyens et al., 1988a, Baeyens et al., 1988b; Levey and Martin, 1987, Martin and Levey, 1987). In this respect, we conducted an experiment to verify the Levey and Martin, 1983, Levey and Martin, 1987, Martin and Levey, 1987 hypothesis that, unlike signal-learning, evaluative conditioning should be resistant to extinction. Mere contingent presentation of neutral with (dis)liked stimuli was sufficient to change the affective—evaluative tone of the originally neutral stimuli in a (negative) positive direction (p<0.0001). A subsequent extinction procedure did not have any influence on the acquired evaluative value of the originally neutral stimuli (p<0.0001). A follow-up study demonstrated that the evaluative discriminations were still present two months after the acquisition and extinction manipulations (p<0.0001). These findings provide full support for the resistance to extinction hypothesis. At a theoretical level, this is considered to be further evidence for the hypothesis that evaluative conditioning is not mediated by the acquisition of propositional—declarative knowledge about stimulus contingencies. Finally, we suggest an intriguing analogy between the evaluative conditioning phenomenon and the ‘laws of sympathetic magic’ (Rozin et al., 1986).