Reinstatement after human feature-positive discrimination learning

Mathijs Franssen; Nathalie Claes; Bram Vervliet; Tom Beckers; Dirk Hermans; Frank Baeyens

doi:10.1016/j.beproc.2017.01.007

Reinstatement after human feature-positive discrimination learning

Behav Processes. 2017 Apr:137:73-83. doi: 10.1016/j.beproc.2017.01.007. Epub 2017 Jan 17.

Authors

Mathijs Franssen¹, Nathalie Claes², Bram Vervliet³, Tom Beckers³, Dirk Hermans³, Frank Baeyens³

Affiliations

¹ Center for the Psychology of Learning and Experimental Psychopathology, University of Leuven, Leuven, Belgium; Health Psychology, University of Leuven, Leuven, Belgium. Electronic address: mathijs.franssen@kuleuven.be.
² Health Psychology, University of Leuven, Leuven, Belgium; Department of Experimental-Clinical and Health Psychology, University of Ghent, Ghent, Belgium.
³ Center for the Psychology of Learning and Experimental Psychopathology, University of Leuven, Leuven, Belgium.

PMID: 28108345
DOI: 10.1016/j.beproc.2017.01.007

Abstract

In two experiments, using an online conditioned suppression task, we investigated the possibility of reinstatement of extinguished feature-target compound presentations after sequential feature-positive discrimination training in humans. Furthermore, given a hierarchical account of Pavlovian modulation (e.g., Bonardi, 1998; Bonardi and Jennings, 2009), we predicted A-US reinstatement to be stronger than US-only reinstatement. In Experiment 1, participants learned a sequential feature-positive discrimination (X→A⁺|A^-), which was subsequently extinguished (X→A^-). During the following reinstatement phase, group US-only received US-only presentations (not signalled), group A-US received A-US presentations, and the Control group received exposure to the context, but no CSs or USs, for an equal amount of time. Reinstatement of differential X→A/A responding was observed in the US-only group but not in the Control or A-US groups. Although differential X→A/A responding was not significant in group A-US, responding to the X→A compound was significantly stronger compared to that in group US-only. Hence, it could be the case the group A-US showed stronger reinstatement, but that differential responding was abolished due to excitation gained by A. Experiment 2 was set up to circumvent the acquired excitation of A by testing transfer of the feature after A-US reinstatement to a different target, B. Participants acquired two discriminations, X→A/A and Y→B/B, of which X→A was then extinguished. Subsequently, group A-US received reinforced presentations of A during a reinstatement phase while group Control received exposure to the context. Final testing of the novel X→B compound was hypothesized to show higher responding in group A-US than in group Control, but findings of this approach were limited due to acquired equivalence and/or perceptual factors causing a secondary extinction effect. We conclude to have obtained clear evidence in favour of reinstatement of differential responding after human Feature-Positive discrimination training and subsequent compound extinction, but no evidence in favour of A-US presentations being a stronger trigger for reinstatement than are US-only presentations.

Keywords: Feature positive discrimination; Human; Modulation; Occasion setting; Reinstatement.

Publication types

Review

MeSH terms

Adolescent
Animals
Association Learning*
Avoidance Learning
Conditioning, Classical*
Conditioning, Operant
Cues*
Discrimination Learning*
Extinction, Psychological*
Female
Humans
Male
Mental Recall*
Pattern Recognition, Visual
Software
Transfer, Psychology*
Video Games
Young Adult