Relationships between associative and non-associative inhibition in feature negative and extinction preparations

Abstract

Associative learning comprises of a wide range of mechanisms through which an organism can gain an evolutionary advantage by learning about the surrounding environment and adapting to changes. Associative learning is highly flexible allowing for learnt associations to be changed if there are no longer advantageous or become redundant, and two ways of changing the meaning of previously learnt associations is through conditioned inhibition and extinction. The latter is of particular interest for changing the maladaptive mechanisms developed as a result of addiction for example. Conditioned inhibition and extinction rely on inhibition, individual differences in inhibition having major potential implications, however inhibition is not a purely associative construct and has been defined in many different ways in the wider field of Psychology. The current thesis aimed to assess the link between associative and non-associative inhibition, which are assumed to be independent subtypes of inhibition, by using various inhibition measures. For associative inhibition, the speed of feature negative discrimination learning, conditioned inhibition, speed of extinction, and context inhibition were used. For non-associative inhibition, following the structure of inhibition proposed by Bari and Robins (2013), measures of cognitive inhibition, delay discounting, and response inhibition were employed. It was also aimed to assess the effectiveness of using compound extinction techniques in the forms of super-extinction and deepened extinction compared to cue alone extinction. The final aim was to compare which of three formal models of associative learning Rescorla-Wagner, configural Rescorla-Wagner, and Pearce configural model, is best at predicting the observed data in the extinction study where cue alone, super-extinction, and deepened extinction were compared. For the first aim no evidence of a relationship between associative and non-associative inhibition was found, with the potential exception of a link with the Behavioural Inhibition System of the Behavioural Inhibition System/Behavioural Activation System. As a result, it was concluded that associative inhibition is an independent inhibitory construct, unrelated to non-associative inhibition, and therefore should be included as such in future inhibitory models. Regarding the extinction methodology comparisons, it was found that neither super-extinction or deepened extinction resulted in a more stable extinction, in fact super-extinction was found to be more unstable compared to cue alone and deepened extinction. It was concluded that based on the current results, compound extinction was not a reliable method of enhancing extinction. When assessing the predictions of the three formal models of associative learning, the Pearce configural model was found to be the best overall model, however this model was not the best for every participant.

More information

Identifiers

ORCID for Steven Glautier: orcid.org/0000-0001-8852-3268

Catalogue record

Export record