Back

Hedonic foraging: From reward to action

Clemente, A.; Penacchio, O.

2025-05-06 animal behavior and cognition
10.1101/2025.05.04.652128 bioRxiv
Show abstract

Short abstractAssigning hedonic value shapes behaviour across organisms, yet how these values translate into motivated behaviour remains unclear. We introduce hedonic foraging as a mechanism by which hedonic evaluation guides behaviour via iterative reward maximisation. This framework integrates hedonic evaluation with active inference, whereby expected free energy encodes wanting - the motivational drive toward rewarding outcomes - while variational free energy corresponds to liking - the hedonic impact of the outcome. Hedonic foraging unifies the mechanisms underlying behaviours ranging from allostasis to art appreciation, casting cognition and motivated behaviour as purposive engagement with the environment and its affordances. Long abstractHedonic evaluation shapes behaviour. Yet, the mechanisms that transform hedonic values into motivated behaviour remain poorly understood, and key questions remain unanswered. How do habits, wanting and liking interact to generate behaviour? Do common principles govern behaviour across species, cognitive systems and behavioural complexity? We propose a unifying framework for understanding motivated behaviour that integrates hedonic evaluation with active inference, a theory in which perception, action and cognition emerge from the imperative to minimise free energy - a proxy for the discrepancy between actual and preferred states. We argue that expected free energy encodes wanting - the motivational drive toward rewarding outcomes - while variational free energy provides a measure of liking and disliking of resulting outcomes. This synthesis, which we term hedonic foraging, offers a principled explanation for how organisms navigate their environments based on a common mechanism of pleasure-driven learning and decision-making. It posits that motivated behaviour emerges from the iterative pursuit of maximising reward, with hedonic evaluation motivating and monitoring actions through wanting and liking mechanisms. We demonstrate how hedonic foraging captures a wide range of behaviours, from allostatic processes to sophisticated cultural endeavours such as art appreciation. Our integrative perspective enables testing the neurobiological implementation of the core constructs of active inference as the main components of hedonic evaluation, and fosters mutual enrichment between these fundamental approaches to behaviour. Ultimately, hedonic foraging reinforces the view that cognition is essentially about purposive engagement in adaptive interactions with environmental affordances and offers a unified framework for investigating cognition and behaviour.

Matching journals

The top 7 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.