AI Insight
This study extends the theoretical framework connecting the brain's free-energy principle to predictive coding by using exponential-family distributions instead of limiting assumptions to Gaussian distributions. The extended model better captures biological neural network properties including nonlinear responses, heterogeneous neuron types, and biologically plausible positive-only firing rates, while maintaining the theoretical correspondence between variational Bayesian inference and predictive coding up to the second cumulant. The researchers demonstrate that this broader framework can be trained using biologically plausible local learning rules.
Why it matters
This work strengthens the theoretical foundation for understanding how the brain performs perceptual inference and could inform the development of more biologically realistic artificial neural networks. The framework's ability to incorporate diverse neuron types and nonlinear dynamics while maintaining computational tractability may advance both neuroscience models and neuromorphic computing applications.
Understand the Science
Abstract: The sensory cortices of the brain perform perceptual inference efficiently through their complex networks of neurons. One of the theoretical accounts of this process is the free-energy principle (FEP), which postulates that the brain performs variational Bayesian inference. Pioneering studies have shown that FEP can correspond to the predictive coding (PC) hypothesis under the Gaussian assumption and Laplace approximation. However, PC-based implementations of FEP within such a limited Gaussian regime have failed to capture several properties of biological neural networks, such as nonlinearity and heterogeneity of input–output properties within a network, and the biological implausibility of negative firing rates. This study shows that, when a broader class of probability distributions, namely the exponential family of distributions (EFD), is assumed for the variational posterior and prior, these missing characteristics are exhibited within the network, maintaining the FEP–PC correspondence up to the second cumulant of the posterior. We also show that the proposed model can be trained by biologically plausible local plasticity rules. Our results enrich the explanatory power of FEP regarding neural dynamics involved in perception as variational inference.