Reproduce “Free Energy in a Circumplex Model of Emotion” on OpenResearch

Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.

Universal app for Apple silicon and Intel. Requires macOS 11 or later.

About this paper

Candice Pattisapu, Tim Verbelen, Riddhi J. Pitliya, Alex B. Kiefer, Mahault Albarracin

This research from VERSES Research Lab presents a computational model of emotion within the active inference framework, mapping both valence and arousal to components of free energy. Simulations demonstrate that the model can generate intuitive emotional trajectories in response to prior beliefs and environmental outcomes, offering a unified account of emotional dynamics.

Key insights

  • Emotional states, including both valence and arousal, can be mathematically grounded within the active inference framework by mapping them to specific components of free energy.
  • Prior beliefs about the environment significantly influence an agent's initial emotional state and subsequent emotional trajectories.
  • The model demonstrates an inherent capacity for emotional regulation, as agents consistently moved towards low arousal and neutral valence states over time.

Summary by alphaXiv.

Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.