Coupling between statistical learning and interference suppression

In an ever-changing environment, synchronizing automatic and controlled behaviors is essential for successful adaptation. Yet, the relationship between these behaviors is often debated, partly because of the diverse approaches used to investigate them. The present study examined how statistical learning of environmental regularities, a process that supports automatic behaviors, and interference suppression, which contributes to controlled responding in the presence of distraction, interact when both operate concurrently. Participants completed a visual four-choice reaction time paradigm that combined a statistical learning task with the Eriksen flanker task. In this novel paradigm, a central target stimulus, surrounded by flanker stimuli, was either predictable or unpredictable based on the statistical regularities induced by the sequence of target stimuli. Statistical learning was indexed by the performance difference between predictable and unpredictable trials, whereas interference suppression was reflected by the flanker congruency effect (i.e., the performance difference between congruent and incongruent trials). Reaction times revealed a significant interaction: the flanker congruency effect was reduced on predictable trials, and the behavioral expression of statistical knowledge was larger in the incongruent than in the congruent condition. Together, these findings provide a proof-of-concept demonstration of a behavioral interaction between statistical learning and interference suppression within the present paradigm. Although this experimental design does not establish the causal direction of the interaction, the observed pattern indicates that learned predictability is associated with a reduced cost of conflicting distractors.