Cognitive Neuroscience
The Free-Energy Principle: A Unified Brain Theory?
Karl Friston · 2010 · Nature Reviews Neuroscience
Proposed that the brain minimizes a single quantity - variational free energy - across perception, action, and learning.
Research objective
Provide a unifying mathematical framework linking perception, action, and neural plasticity.
Methodology
Theoretical synthesis combining Bayesian inference, statistical physics, and predictive coding into a single variational principle.
Key findings
- Perception, learning, and action can all be cast as minimizing prediction error.
- The framework generalizes Helmholtzian inference and active inference into one formalism.
- Many cortical phenomena fit the predictive-coding template.
Strengths
- Mathematically elegant and broadly generative of testable hypotheses.
- Bridges neuroscience, machine learning, and theoretical biology.
Limitations
- Highly abstract; direct empirical falsification is contested.
- Implementation in real neural circuits remains an active question.
Practical implications
- Influences computational psychiatry, BCI design, and theories of consciousness.
- Connects neuroscience to modern generative modeling in AI.