Marc–Lluís Vives

The unique value of zero prediction errors in reinforcement learning

The paper shows that zero prediction errors in reinforcement learning lead to unique psychological and neural responses, challenging the assumption that only unexpected outcomes drive belief updates.

Replay shapes abstract cognitive maps for efficient social navigation

The paper shows that people build and refine cognitive maps of social networks through learning and rest, improving social navigation accuracy.

Uncertainty aversion predicts the neural expansion of semantic representations

This study shows that uncertainty aversion leads to more differentiated semantic representations, reflected in neural activity patterns and behavioral outcomes.