Resistance to extinction of fear-relevant stimuli: Preparedness or selective sensitization?

Abstract

Human conditioning research has revealed an apparent resistance to extinction of aversive conditioning to pictures of fear-relevant stimuli such as snakes and spiders, supporting M. E. P. Seligman's (1971) preparedness theory of fears and phobias. This article examines an alternative account based on activation of preexisting response tendencies under threat (selective sensitization). Two experiments demonstrate that selective sensitization of electrodermal responses is attenuated when a fear-relevant stimulus serves as a negative conditioned stimulus (CS-), but is maintained when it serves as a positive conditioned stimulus (CS+). Previous extinction results may therefore be due to preservation of initial responding to CS+ but not CS-. Selective sensitization offers a model for the nonassociative activation of fears and phobias to prepotent stimuli under conditions of stress or threat. Possible genetic and cognitive mechanisms are discussed.

Can prepared fear conditioning result from verbal instructions?

“…This bias is translated into stronger conditioned reactions on other measures, such as SCRs (Davey, 1992;Lovibond, Siddle, & Bond, 1993). In our Experiment 2, selective learning effects were obtained on US expectancy ratings, replicating the US expectancy bias effect of Davey (1992) and Lovibond et al (1993).…”

“…This bias is translated into stronger conditioned reactions on other measures, such as SCRs (Davey, 1992;Lovibond, Siddle, & Bond, 1993). In our Experiment 2, selective learning effects were obtained on US expectancy ratings, replicating the US expectancy bias effect of Davey (1992) and Lovibond et al (1993). Arguably, the noreinforcement instructions rapidly abolished the expectancy bias for the fear-relevant CSwhereas the threat instructions retained the expectancy bias for the fear-relevant CS+2, resulting in a selective learning effect on US expectancy ratings.…”

“…Even though prepared learning effects have been observed on US expectancy ratings and SCRs (e.g., Davey, 1992;Lovibond, Siddle, & Bond, 1993;Öhman, Erixon, & Löfberg, 1975), fear ratings might be a better measure for capturing the emotional components of preparedness effects, that is, components that are driven by a fear learning module (Mertens et al, 2015;Soeter & Kindt, 2012).…”

Asymmetrical stimulus generalization following differential fear conditioning

“…The second is that context conditioning has cuespecific effects. Both possibilities suppose that pre-wired circuitry supports the processing of "fear-relevant" stimuli and that this circuitry becomes enabled under conditions of high fear, anxiety or arousal (Gray, 1982;1987;Lovibond et al, 1993). One study of humans found that an aversive electrical stimulation sensitized responding to "fear-relevant" stimuli but not to "fear-irrelevant" stimuli (Öhman et al, 1975).…”

Evaluating conditioning of related and unrelated stimuli using a compound test

“…It is even more rare to use the kind of compound test procedure employed here to evaluate the possibility of initial differences in fear. Indeed, the only study of which I am aware that has used such a test (Lovibond, Siddle, & Bond, 1993) was able to detect no differences attributable to the pairing of a fear-related CS with a shock, compared with the pairing of a fear-unrelated CS with a shock. That result is consistent with the possibility that observed differences reflect preexperimental differences in fear rather than differences that develop as a result of the associative conditioning conducted in the experiment.…”