Abstract:This paper examines neural representation under active inference. Organisms employ generative models to navigate the world adaptively, not (or not solely) to understand it. Crucially, different organisms might employ different types of generative models, ranging from action‐oriented models (predicting outcomes) to explicit models (reconstructing the external environment). Connecting meaning to agentive action to the array of generative models imparts diverse forms of understanding of an organism's ecological niche. This paper considers neural representation through the lens of active inference, a normative framework for understanding brain function. It delves into how living organisms employ generative models to minimize the discrepancy between predictions and observations (as scored with variational free energy). The ensuing analysis suggests that the brain learns generative models to navigate the world adaptively, not (or not solely) to understand it. Different living organisms may possess an array of generative models, spanning from those that support action‐perception cycles to those that underwrite planning and imagination; namely, from explicit models that entail variables for predicting concurrent sensations, like objects, faces, or people—to action‐oriented models that predict action outcomes. It then elucidates how generative models and belief dynamics might link to neural representation and the implications of different types of generative models for understanding an agent's cognitive capabilities in relation to its ecological niche. The paper concludes with open questions regarding the evolution of generative models and the development of advanced cognitive abilities—and the gradual transition from pragmatic to detached neural representations. The analysis on offer foregrounds the diverse roles that generative models play in cognitive processes and the evolution of neural representation.

Active inference as a theory of sentient behavior

Active inference on discrete state-spaces: a synthesis

Expanding the Active Inference Landscape: More Intrinsic Motivations in the Perception-Action Loop

Active Inference and Intentional Behaviour

The Anatomy of Inference: Generative Models and Brain Structure

The Active Inference Approach to Ecological Perception: General Information Dynamics for Natural and Artificial Embodied Cognition

Active Inference, Curiosity and Insight

Active Inference as a Computational Framework for Consciousness

Active Inference as a Model of Agency

How Active Inference Could Help Revolutionise Robotics

Neural representation in active inference: Using generative models to interact with—and understand—the lived world

Active Inference and Human--Computer Interaction

Neural representation in active inference: using generative models to interact with -- and understand -- the lived world

Active inference: demystified and compared

Artificial enactive inference in three-dimensional world

Neuroscience-inspired perception-action in robotics: applying active inference for state estimation, control and self-perception

A Minimal Active Inference Agent

Deconstructing deep active inference

Active Inference in Psychology and Psychiatry: Progress to Date?

Active inference body perception and action for humanoid robots

Deriving time-averaged active inference from control principles