Purchase this article with an account.
Anne Milner, Mary MacLean, Barry Giesbrecht; Can value-driven attentional capture be extinguished?. Journal of Vision 2017;17(10):1294. doi: 10.1167/17.10.1294.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Visual stimuli previously associated with reward can capture attention in the absence of reinforcement (Anderson & Yantis, 2012). We investigated whether reward-associated attentional capture would persist after repeated exposure in the absence of reinforcement or whether the effect could eventually be extinguished. During the training phase, which consisted of 800 trials, participants (n=12) learned to associate one of two target colors (red and blue) with reward, and the other with the absence of reward. One week later, during the extinction phase, participants completed 1600 trials of the same task but were no longer rewarded. Reaction times were significantly faster to reward-associated targets than no reward-associated targets. Critically, there was an interaction between reward condition and task phase such that the difference between rewarded and unrewarded trials was significantly larger during training (MD=50.32ms, SEM=10.8) than extinction (MD=25.08ms, SEM=4.64), suggesting that there was reduction of the reward-associated capture effect during extinction. However, when the analysis was broken down by block, there was an interaction between reward condition and block during training, but not during extinction. This interaction indicates the presence of learning during training, because the difference between rewarded and unrewarded trials increased over the course of the training phase, but not during extinction. These findings suggest that while the effect of value-driven attentional capture is reduced, extinction learning did not occur. However, 7 out of the 12 participants displayed some extinction learning, whereby the difference between previously reward-associated and no reward-associated targets decreased from the first to the last block of the extinction phase. Although there is evidence of a reduction in the effect of value-associated stimuli, the effect still remains statistically significant over 1600 trials, suggesting evidence for some persistence. Thus, reward learning is persistent, but individual differences may play a role in how value-driven attention is learned and extinguished.
Meeting abstract presented at VSS 2017
This PDF is available to Subscribers Only