Purchase this article with an account.
Jennifer L. O'Brien, Helena J.V. Rutherford, Jane E. Raymond; Can value learning modulate low-level visual object recognition? An ERP study. Journal of Vision 2008;8(6):36. doi: 10.1167/8.6.36.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Experience of interacting with visual objects allows us to acquire neural codes that predict both the value of interacting with them (i.e., in terms of reward or punishment) and the likelihood of obtaining that outcome. Acquisition and storage of these expected value (EV) codes involves pre-frontal and limbic system circuits that have ample opportunity to interface with visual object processing networks. Here we ask whether object recognition is modulated by previously acquired EV codes; would prior experience with reward or punishment facilitate object recognition. To test this, we used face stimuli with learned values and measured both behavioral recognition and the N170 ERP component. The N170 is widely seen as an automatically elicited, face-specific response reflecting pre-categorical structural encoding. Modulating the N170 by presenting stimuli with different EVs would indicate an influence of value learning at a relatively early stage of processing. We first had participants engage in a simple choice task in which they gained or lost money. Through this experience, they acquired specific (and quantified) EVs for 12 different face stimuli (O'Brien & Raymond, VSS 2007). We then measured recognition (old/new judgement) for these and novel faces when briefly presented. ERPs were simultaneously recorded from 64 scalp electrodes. Recognition was significantly higher and faster for faces associated with high positive (gain) EV than for similarly exposed faces with low positive, negative (loss), or zero EV. Preliminary results indicate that the N170 amplitude during recognition may also depend on EV, demonstrating a top-down influence on ‘early’ face processing and explaining conflicting data on N170 modulation by face identity, familiarity, and top-down information such as emotional context.
This PDF is available to Subscribers Only