Goal‐directed EEG activity evoked by discriminative stimuli in reinforcement learning

Abstract

In reinforcement learning (RL), discriminative stimuli (S) allow agents to anticipate the value of a future outcome, and the response that will produce that outcome. We examined this processing by recording EEG locked to S during RL. Incentive value of outcomes and predictive value of S were manipulated, allowing us to discriminate between outcome-related and response-related activity. S predicting the correct response differed from nonpredictive S in the P2. S paired with high-value outcomes differed from those paired with low-value outcomes in a frontocentral positivity and in the P3b. A slow negativity then distinguished between predictive and nonpredictive S. These results suggest that, first, attention prioritizes detection of informative S. Activation of mental representations of these informative S then retrieves representations of outcomes, which in turn retrieve representations of responses that previously produced those outcomes.

Testing the controllability of contextual cuing of visual search

“…These results, jointly with the current data, are compatible with the hypothesis that contextual cuing of visual search is not supported by uncontrollable attentional orientation towards the target in repeated trials. This idea will benefit from future research examining the controllability of CC, for instance by using alternative techniques for measuring attentional orientation, such as event-related potentials24, fMRI or/and eye-tracking25.…”

The effects of reward magnitude on reward processing: An averaged and single trial event-related potential study

“…Indeed, a few recent studies have employed such tasks and observed valence by magnitude interactions (Bellebaum, Polezzi, & Daum, 2010;Gu et al, 2011;Kreussel et al, 2012;Luque, Morís, Rushby, & Le Pelley, 2015;Weinberg, Riesel, & Proudfit, 2014), although an earlier study by Sato et al (2005) did not reveal a valence by magnitude interaction.…”

Ambiguity produces attention shifts in category learning

“…Our results confirm that learning interfering information does produce a change in attention, so that part of the attention usually allocated to predictive cues diverts to nonpredictive and nonsalient cues analogous to the contexts of associative learning experiments (in particular the type used by . Ideally, future research should follow up our results using alternative procedures to measure attentional processes (e.g., Livesey et al 2009;Wills et al 2014;Glautier and Shih 2015;Luque et al 2015).…”