Predictive coding is presented as the neural instantiation of Bayesian inference in the paper's account of autism.
First-pass extracted concept
predictive coding
Extracted Explainers
Evidence Snippets
Predictive coding theories of sensory brain function interpret the hierarchical construction of the cerebral cortex as a Bayesian, generative model capable of predicting the sensory data consistent with any given percept.
when Bayesian inference is grounded in its neural instantiation-namely, predictive coding-many features of autistic perception can be attributed to aberrant precision
Supporting Sources
Linked Claims
The review argues that predictive-coding computational architecture can be combined with the canonical microcircuit and laminar extrinsic connectivity to produce a template schematic for cortical implementation.
The review states that early predictive-coding formulations inferred separate source populations for forward and backward pathways, and notes that non-bifurcating directional projection has only recently been confirmed.
The review identifies rapid advances in transgenic neural engineering in the mouse as enabling further examination of predictive-coding microcircuit hypotheses.
The review presents interneuron diversity and pyramidal-neuron connectomic specificity as recurring themes that can drive further theoretical development of predictive-coding microcircuits.
The review describes predictive coding as a hierarchical Bayesian generative model in which backward predictions and forward prediction errors iteratively optimize perceptual representations.
Many features of autistic perception can be attributed to aberrant precision or beliefs about precision when Bayesian inference is grounded in predictive coding.
when Bayesian inference is grounded in its neural instantiation-namely, predictive coding-many features of autistic perception can be attributed to aberrant precision (or beliefs about precision)