Probabilistic feedback: what do we do with it?

Our ability to optimize future behaviour based on information we received about the result of our previous behaviour allows us to navigate flexible through an ever-changing world. Especially stimuli in our environment that we associate with rewarding outcomes shape our behaviour. Research on value-based attentional capture, for example, consistently showed that reward-associated stimulus features can modulate attention voluntarily or involuntarily, dependent on their task relevance. This is manifested behaviourally through faster response times (RTs) in selecting target items that carry reward-associated features, and slower RTs when distractor items do. In a series of studies we investigated the neurocognitive dynamics underlying probabilistic reward learning, with a specific focus on feedback processing, and related modulations of attention in subsequent encounters with these reward-related stimuli. I will present our findings which provide insights into the temporal order of distinct cognitive stages of feedback processing in the brain, and how this in turn leads to changes in the sensory cortices in order to prepare for upcoming behaviour.