Kevin Ortego¹, Douglas Addleman², Viola Stoermer¹; ¹Dartmouth College, ²Gonzaga University

Statistical regularities in the environment, such as target or distractor items appearing more frequently in particular locations or colors, influence the brain’s processing and prioritization of information, even when we are not explicitly aware of these regularities. Previous research has suggested that the behavioral benefits from incidental learning rely on modulations of both early attentional processes and later decision-related processes (Ortego et al., 2025; Wang et al., 2023). In two experiments, we recorded EEG while participants performed a visual search task in which color probabilities were manipulated such that either the target (a left/right gap Landolt C among top/bottom gap distractors) or distractors occurred more frequently in a particular color (Exp 1 target learning, N=32; Exp 2 distractor learning, N=54). In both experiments, trials in which the target/distractors matched the most likely color (valid) showed faster response times than trials where this association was reversed (invalid). These behavioral benefits were accompanied by latency shifts of the N2pc ERP component, a neural index of attentional selection, as well as amplitude modulations of the LPC, an ERP component often interpreted as reflecting later decision-making stages, suggesting that the color regularities impacted both early selection and later decision-making processes. However, when controlling for response time differences across conditions (both by subsampling trials to equate RT distributions before computing stimulus-locked ERPs, and by analyzing response-locked rather than stimulus-locked ERPs) we found that the amplitude differences in the LPC disappeared, suggesting these differences are a downstream consequence of earlier processing changes and not a unique contributor to the behavioral benefits. We further demonstrate that latency between attentional selection (as indexed by the N2pc) and response execution was similar for valid and invalid trials, suggesting that modulations early in the processing chain, rather than at later decision-making stages, are the key driver of incidental learning benefits.