The impact of previously learned feature-relevance on generalisation of conditioned fear in humans

Abstract

Background and objectivesExcessive generalisation of fear learning has recently been implicated in the development and maintenance of anxiety disorders such as Posttraumatic Stress Disorder. Evidence is growing to suggest that cognitive processes such as rule-abstraction may be involved in fear generalization. In a study by Vervliet et al. (2010), verbal instructions regarding the relevance of stimulus colour or shape for predicting the delivery of an electric shock led to greater generalisation of fear to novel stimuli which contained the instruction-relevant feature. Here, we examined the impact of pre-training rather than instructing feature relevance. MethodsA pre-training phase was used to allow learning of the predictive relevance of either stimulus colour or shape. This was followed by the same conditioning and generalization phases used in the previous study. ResultsThere was greater generalisation of fear to the test stimulus that contained the pre-training relevant feature (either colour or shape), and this effect was stronger in participants who correctly reported the training rule. This pattern was statistically significant on the expectancy measure but not on skin conductance. LimitationsHigh levels of variability on the skin conductance measure reduced the power to detect a significant difference on test. ConclusionsThe results demonstrate the potential for prior experiences and the beliefs derived from them to impact on generalisation of fear. They also add to the evidence for an involvement of higher-order cognitive processes in how fear learning spreads and how such a process may be addressed clinically.