Abstract
The feedback-related negativity (FRN) is an event-related potential (ERP) component associated with processing of performance feedback, with more negative amplitudes for losses relative to wins. The amplitude of the FRN following near misses, i.e. the experience of coming close to winning, is between the amplitude elicited by losses and wins. In gambling, however, outcome value may not always be obvious since initially placed bets need to be taken into account when evaluating wins or losses. It is still unclear if initial bet size is reflected in the FRN or the later P300 component. The present study applied a virtual card gambling task to investigate the sensitivity of FRN and P300 to the manipulation of outcome magnitude as implemented through the presence or absence of initial bets, resulting in wins, losses or ambivalent outcomes, with the latter representing losses with and wins without bets. The FRN was larger for trials with bets compared to trials without bets. Wins were associated with a smaller FRN than losses or ambivalent outcomes, while losses and ambivalent outcomes did not differ. P300 amplitudes were larger for trials without bets, and wins were associated with a larger P300 than losses or ambivalent outcomes. Crucially, P300 amplitudes were also smaller for ambivalent outcomes compared to losses. Thus, the different dimensions determining outcome value appear to be integrated in early and late stages of feedback processing. However, only at later stages reflected in the P300 were ambivalent outcomes with and without bet clearly distinguished from other outcomes.
Highlights
Feedback is essential for informing us about the outcome of our actions and about the success or failure of strategic behaviour and behavioural adaptations
More recent studies which took into account subjective reward expectations, did report that the feedbackrelated negativity (FRN) reflected gradual deviations of outcomes from reward expectations, this effect being most pronounced when feedback could be used for learning action-outcome contingencies [10,14,15,16]
Post-hoc tests revealed that the FRN was larger for clear losses (3.37 mV64.05) compared to clear wins (3.88 mV64.44; t(27) = 2.222, p = .017), and for ambivalent outcomes (3.30 mV64.37) compared to clear wins (t(27) = 2.257, p = .016)
Summary
Feedback is essential for informing us about the outcome of our actions and about the success or failure of strategic behaviour and behavioural adaptations. Negative feedback usually signals failure and decreases the occurrence rate of preceding actions, while positive feedback commonly indicates success, increasing occurrence rate. The FRN has been shown to be larger when outcomes are unfavourable compared to favourable [5,6,7], and is thought to reflect phasic decreases in dopaminergic signals conveyed from the basal ganglia to the anterior cingulate cortex [2,8,9,10]. Feedback indicating an ‘‘even’’ draw, i.e. neither a win nor loss, has been shown to elicit an FRN comparable to that elicited by negative feedback [11,12]. More recent studies which took into account subjective reward expectations, did report that the FRN reflected gradual deviations of (negative) outcomes from reward expectations, this effect being most pronounced when feedback could be used for learning action-outcome contingencies [10,14,15,16]
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
