Motivational Biases in Go and Nogo Learning From Immediate Versus Delayed Feedback

Motivational biases in decision-making describe the phenomenon that individuals are more likely to perform an action to receive a reward and to withhold an action to avoid punishment. These biases likely originate from reward-related modulations of the striatum, presumably reflecting biased action selection and/or learning, the latter with facilitated learning from rewarded actions and impaired learning from punished inactions. As striatal engagement during learning depends on whether feedback is presented immediately or delayed, the present study examined whether motivational biases are similarly sensitive to feedback timing. Moreover, based on the idea that motivational biases arise through biased learning, we tested whether they manifest as biased feedback processing for action over inaction. Fifty-six healthy participants performed an orthogonalized go/nogo task learning whether to execute or withhold a response in both reward obtainment and loss-avoidance contexts-either receiving immediate (n = 28) or delayed feedback (n = 28). Applying computational modeling of choice behavior and using electroencephalography (EEG), we assessed biased instrumental learning and its neural signature in feedback processing, focusing on the feedback-related negativity/reward positivity and the P3 in the event-related potential (ERP). Behavioral results demonstrated robust motivational biases across learning from immediate and delayed feedback. Importantly, motivational biases were best explained in terms of biased action selection rather than biased learning, aligning with our ERP findings that did not provide evidence for action-biased feedback processing. Taken together, these results highlight motivational biases as a robust hallmark of decision-making, relying on common mechanisms involved in both learning from immediate and delayed feedback.