Uncertainty-Driven Valuation era
Wolfram Schultz's work established dopamine signals as reward prediction errors that drive learning, a canonical teaching signal still central to uncertainty-sensitive valuation in this era. Peter Dayan and Matthew Daw advanced computational accounts showing how state and outcome uncertainty modulate learning and policy control, linking dopamine signals to uncertainty and to shifts between model-based and model-free strategies. Frank and colleagues showed that subthalamic representations gate action by integrating subjective reward with effort costs, shaping decision thresholds under uncertainty. Yoav Niv has framed cross-task, uncertainty-driven adjustments of learning and policy as a unifying mechanism, highlighting translational relevance to effort-related dysfunctions.