Reappraising cognitive control: normal reactive adjustments following conflict processing are abolished by proactive emotion regulation.

The congruency sequence effect (CSE) reflected by the influence of the congruency of the previous trial on the current one translates improved cognitive control (CC). Yet, it remains debated whether reactive or proactive control processes mostly contribute to this effect. To address this question, we administered a Stroop task controlling for effects of feature repetition and contingency learning to a large group of participants, where we manipulated the frequency of incongruent trials in a block-wise fashion to induce either proactive (high-conflict frequency) or reactive (low-conflict frequency) control. Moreover, as the presentation of trial-by-trial evaluative feedback could influence control processes operating at a local level, we compared effect of evaluative vs. neutral feedback on the CSE, for each control mode separately. We tested the prediction that CSE should be influenced by conflict frequency and feedback type concurrently. Results showed that when evaluative feedback was used, the CSE was increased if conflict frequency was low, confirming that the CSE stemmed from reactive control mainly. If conflict frequency was high, a different sequence effect was observed. The use of neutral feedback abolished the modulation of the CSE by conflict frequency. Moreover, correlation results showed that reappraisal, corresponding to a proactive emotion regulation strategy, was negatively related to the CSE in this condition, suggesting that proactive control can alleviate the reactive dominance of the CSE. Altogether, these results suggest that CC is flexible, and its expression depends on the subtle balance between proactive and reactive control processes.