Acquisition and extinction across multiple virtual reality contexts

Abstract

This experiment examined the effects on recovery following a human contingency learning task across multiple virtual contexts. Specifically, if recovery of excitatory responding can be attenuated by experimental extinction in a single context or across multiple contexts when acquisition was also conducted in a single context or across multiple contexts. Sixty-one participants received conditioned stimulus (CS) and unconditioned stimulus (US) pairings in one or three virtual contexts. Orthogonal to this, participants received experimental extinction (CS presentations alone) in one or three different virtual contexts. Participants were then tested in a novel virtual context and the differences in expectation levels of the US in the presence of the CS were compared. Participants that experienced acquisition in one context and extinction in multiple contexts showed lower expectation of the US relative to participants that experienced acquisition and extinction in one context. Participants that experienced acquisition across multiple contexts, however, had higher expectation of the US regardless of whether extinction was also conducted across multiple contexts. This study demonstrated that learning across multiple contexts can negatively impact subsequent extinction; it is the first to do so using interactive VR-generated contexts, and it is the first to demonstrate this effect in humans and only the second study to demonstrate this effect at all. These findings highlight the importance of learning history and the likelihood that first-learned associations acquired across multiple contexts may interfere with subsequent extinction treatment. Availability of data and materialsThe dataset generated during and/or analysed during the current study is available on the Open Science Framework at https://osf.io/64tzw/?view_only= 20aaa34f9fd44f108a8391ca426a7031.