The free-energy principle: a unified brain theory?

• Published in: Nature reviews. Neuroscience Vol. 11; no. 2; pp. 127 - 138
• Main Author: Friston, Karl
• Format: Journal Article
• Language: English
• Published: London Nature Publishing Group UK 01.02.2010; Nature Publishing Group
• Subjects: 631/378/116/2395; 631/553/2702; Accuracy; Animal Genetics and Genomics; Animals; Behavioral Sciences; Biological and medical sciences; Biological Techniques; Biomedical and Life Sciences; Biomedicine; Brain; Brain - physiology; Central nervous system; Cognition - physiology; Control theory; Electrophysiology; Energy; Entropy; Expected utility; Fundamental and applied biological sciences. Psychology; General aspects. Models. Methods; Gibbs' free energy; Humans; Information theory; Learning - physiology; Nerve Net - physiology; Neurobiology; Neurosciences; Perception - physiology; Physiological aspects; Physiology; Probability distribution; Psychological Theory; review-article; Vertebrates: nervous system and sense organs; United Kingdom; Learning; Costs; Acquisition process; Central nervous system; Prediction; Perception; Review; Reward; Optimization; Encephalon; Information theory
• ISSN: 1471-003X; 1471-0048; 1471-0048; 1469-3178
• DOI: 10.1038/nrn2787

Key Points Adaptive agents must occupy a limited repertoire of states and therefore minimize the long-term average of surprise associated with sensory exchanges with the world. Minimizing surprise enables them to resist a natural tendency to disorder. Surprise rests on predictions about sensations, which depend on an internal generative model of the world. Although surprise cannot be measured directly, a free-energy bound on surprise can be, suggesting that agents minimize free energy by changing their predictions (perception) or by changing the predicted sensory inputs (action). Perception optimizes predictions by minimizing free energy with respect to synaptic activity (perceptual inference), efficacy (learning and memory) and gain (attention and salience). This furnishes Bayes-optimal (probabilistic) representations of what caused sensations (providing a link to the Bayesian brain hypothesis). Bayes-optimal perception is mathematically equivalent to predictive coding and maximizing the mutual information between sensations and the representations of their causes. This is a probabilistic generalization of the principle of efficient coding (the infomax principle) or the minimum-redundancy principle. Learning under the free-energy principle can be formulated in terms of optimizing the connection strengths in hierarchical models of the sensorium. This rests on associative plasticity to encode causal regularities and appeals to the same synaptic mechanisms as those underlying cell assembly formation. Action under the free-energy principle reduces to suppressing sensory prediction errors that depend on predicted (expected or desired) movement trajectories. This provides a simple account of motor control, in which action is enslaved by perceptual (proprioceptive) predictions. Perceptual predictions rest on prior expectations about the trajectory or movement through the agent's state space. These priors can be acquired (as empirical priors during hierarchical inference) or they can be innate (epigenetic) and therefore subject to selective pressure. Predicted motion or state transitions realized by action correspond to policies in optimal control theory and reinforcement learning. In this context, value is inversely proportional to surprise (and implicitly free energy), and rewards correspond to innate priors that constrain policies. Karl Friston shows that different global brain theories all describe principles by which the brain optimizes value and surprise. He discusses how these brain theories fit into the free-energy framework, suggesting that this framework might provide a unified account of brain function. A free-energy principle has been proposed recently that accounts for action, perception and learning. This Review looks at some key brain theories in the biological (for example, neural Darwinism) and physical (for example, information theory and optimal control theory) sciences from the free-energy perspective. Crucially, one key theme runs through each of these theories — optimization. Furthermore, if we look closely at what is optimized, the same quantity keeps emerging, namely value (expected reward, expected utility) or its complement, surprise (prediction error, expected cost). This is the quantity that is optimized under the free-energy principle, which suggests that several global brain theories might be unified within a free-energy framework.