Examining the Modulating Effects of Reinforcement Learning and Explicit Instructions on Cognitive Stability and Flexibility

In goal-directed action, individuals must adjust their cognitive stability and flexibility levels to meet varying situational demands and minimize the costs associated with these cognitive control functions. Identifying factors that reliably modulate stability and flexibility is essential for studying their underlying mechanisms and holds potential to support individuals in reducing associated costs. The present study investigated two potential control modulators and tested whether reinforcement-based learning and explicit instructions can induce shifts in cognitive stability and flexibility. Furthermore, it aimed to contribute to the ongoing debate on whether cognitive stability and flexibility represent extreme points on a unidimensional continuum and therefore underlie an inherent tradeoff or rather depend on independent mechanisms. To this end, we conducted two online task-switching experiments. In Experiment 1, we examined whether selectively rewarding participants for stable or flexible behavior would promote the respective control states. In Experiment 2, we aimed to replicate previous findings of instruction-based control adaptations. We did not find coherent evidence for reinforcement- and instruction-based control adjustments. However, across both experiments, shifts in cognitive stability and flexibility were negatively correlated at the level of individual participants. Results thus indicated at least a strategic tradeoff between cognitive stability and flexibility in the applied paradigms.