Waves of control: An ER(S)P study to explore the neural dynamics of proactive and reactive stability.

Abstract

Cognitive stability, the ability to maintain task focus while facing distractions, is a core component of cognitive control. According to the Dual Mechanisms of Control model, stability can be implemented proactively, through anticipatory goal maintenance, or reactively, through transient, conflict-triggered adjustments. However, their neural dynamics remain debated, mainly due to the limited temporal resolution of fMRI and to EEG designs that fail to disentangle different forms of stability. Here, we recorded EEG while 40 participants performed a spatial Stroop task to dissociate predictive proactive stability, induced by list-wide proportion congruency (LWPC), and predictive reactive stability, induced by item-specific proportion congruency (ISPC). Moreover, using response-locked analyses, we exploratively probed the late, corrective form of reactive stability engaged after conflict detection. Using a state-of-the-art analytical approach combining trial-level Bayesian estimates of the manipulated variables with multilevel modeling, we isolated these predictive mechanisms from low-level confounds like contingency. The analyses of event-related potentials and time-frequency activity revealed neural signatures of proactive stability, including anticipatory pre-stimulus modulations and a systematic reduction of post-stimulus conflict-related activity. In contrast, no reliable neural evidence for predictive reactive stability was observed once contingency was controlled. Importantly, response-locked analyses revealed distinct late neural markers compatible with a corrective reactive stability mechanism engaged to resolve conflict when predictive mechanisms failed. Together, these findings provide clear electrophysiological evidence for predictive proactive stability neural mechanisms, while failing to reveal those of ISPC-dependent predictive reactive stability. Moreover, exploratory analyses provided insights into late corrective reactive stability neural mechanisms, engaged when predictive mechanisms failed.