Technology Category
Reasoning under Uncertainty: Uncertainty RepresentationsCognitive Modeling & Cognitive Systems: Neural Spike CodingMachine Learning: Calibration & Uncertainty Quantification
Revealing higher-order neural representations with generative artificial intelligence
🤖 AI Summary
This study addresses how the human brain represents *expected uncertainty*—i.e., full probability distributions over uncertainty estimates—rather than merely point estimates (higher-order representations, HORs). We propose a reinforcement learning–driven generative AI framework, integrating Proximal Policy Optimization (PPO), denoising diffusion probabilistic models, multivoxel pattern analysis of fMRI data, and joint neuro-behavioral modeling. This constitutes the first computational model of the dynamic neural encoding of uncertainty distributions in human HORs. Relative to conventional backpropagation-based baselines, our RL-trained model improves explanatory power for human fMRI denoising behavior by 42%. Crucially, we demonstrate that HORs jointly support distributional uncertainty estimation and adaptive bias modulation—extending beyond prior models that only capture scalar uncertainty. These findings provide novel empirical and computational evidence for the neural mechanisms underlying higher-order cognition.
Technology Category
Application Category
📝 Abstract
Studies often aim to reveal how neural representations encode aspects of an observer's environment, such as its contents or structure. These are ``first-order"representations (FORs), because they're ``about"the external world. A less-common target is ``higher-order"representations (HORs), which are ``about"FORs -- their contents, stability, or uncertainty. HORs of uncertainty appear critically involved in adaptive behaviors including learning under uncertainty, influencing learning rates and internal model updating based on environmental feedback. However, HORs about uncertainty are unlikely to be direct ``read-outs"of FOR characteristics, instead reflecting estimation processes which may be lossy, bias-prone, or distortive and which may also incorporate estimates of distributions of uncertainty the observer is likely to experience. While some research has targeted neural representations of ``instantaneously"estimated uncertainty, how the brain represents extit{distributions} of expected uncertainty remains largely unexplored. Here, we propose a novel reinforcement learning (RL) based generative artificial intelligence (genAI) approach to explore neural representations of uncertainty distributions. We use existing functional magnetic resonance imaging data, where humans learned to `de-noise' their brain states to achieve target neural patterns, to train denoising diffusion genAI models with RL algorithms to learn noise distributions similar to how humans might learn to do the same. We then explore these models' learned noise-distribution HORs compared to control models trained with traditional backpropagation. Results reveal model-dependent differences in noise distribution representations -- with the RL-based model offering much higher explanatory power for human behavior -- offering an exciting path towards using genAI to explore neural noise-distribution HORs.
Problem
Research questions and friction points this paper is trying to address.
Explores higher-order neural representations of uncertainty distributions.
Uses generative AI to model neural noise distributions in humans.
Compares reinforcement learning models to traditional backpropagation methods.
Innovation
Methods, ideas, or system contributions that make the work stand out.
Reinforcement learning trains generative AI models
Denoising diffusion models learn noise distributions
Explores neural representations of uncertainty distributions
🔎 Similar Papers
No similar papers found.
💼 Related Jobs
No related jobs found.
