Transcranial Direct Current Stimulation (tDCS) of the Anterior Prefrontal Cortex (aPFC) Modulates Reinforcement Learning and Decision-Making Under Uncertainty: a Double-Blind Crossover Study

Abstract

Reinforcement learning refers to the ability to acquire information from the outcomes of prior choices (i.e. positive and negative) in order to make predictions on the effect of future decision and adapt the behaviour basing on past experiences. The anterior prefrontal cortex (aPFC) is considered to play a key role in the representation of event value, reinforcement learning and decision-making. However, a causal evidence of the involvement of this area in these processes has not been provided yet. The aim of the study was to test the role of the orbitofrontal cortex in feedback processing, reinforcement learning and decision-making under uncertainly. Eighteen healthy individuals underwent three sessions of tDCS over the prefrontal pole (anodal, cathodal, sham) during a probabilistic learning (PL) task. In the PL task, participants were invited to learn the covert probabilistic stimulus-outcome association from positive and negative feedbacks in order to choose the best option. Afterwards, a probabilistic selection (PS) task was delivered to assess decisions based on the stimulus-reward associations acquired in the PL task. During cathodal tDCS, accuracy in the PL task was reduced and participants were less prone to maintain their choice after positive feedback or to change it after a negative one (i.e., win-stay and lose-shift behavior). In addition, anodal tDCS affected the subsequent PS task by reducing the ability to choose the best alternative during hard probabilistic decisions. In conclusion, the present study suggests a causal role of aPFC in feedback trial-by-trial behavioral adaptation and decision-making under uncertainty.