Graph-Based Representation Learning of Neuronal Dynamics and Behavior

2024-10-01

AI Summary

A core challenge in neuroscience is modeling the time-varying neural connectivity structure at high spatiotemporal resolution and establishing its causal relationship with behavioral outputs in dynamic, high-dimensional environments. To address this, we propose TAVRNN—the first dynamic graph neural network integrating probabilistic graphical learning with temporal attention mechanisms, enabling unified single-neuron-resolution modeling and population-level interpretable representation. Our method synergistically combines variational graph neural networks, dynamic graph representation learning, and cross-modal data alignment. Evaluated on three heterogeneous neural datasets—freely behaving, task-evoked, and neuromorphic platforms—TAVRNN achieves state-of-the-art performance in classification and clustering. Critically, it is the first framework to systematically uncover the dynamic coupling between behavioral adaptability and evolving network topology. Moreover, TAVRNN demonstrates robust generalization across species and experimental platforms.

Technology Category

• Cognitive Modeling & Cognitive Systems: Neural Spike Coding
• Machine Learning: Probabilistic Circuits and Graphical Models
• Reasoning under Uncertainty: Relational Probabilistic Models

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Search and Retrieval-Augmented AI: Web query analysis, representation and understanding
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Understanding how neuronal networks reorganize in response to external stimuli and give rise to behavior is a central challenge in neuroscience and artificial intelligence. However, existing methods often fail to capture the evolving structure of neural connectivity in ways that capture its relationship to behavior, especially in dynamic, uncertain, or high-dimensional settings with sufficient resolution or interpretability. We introduce the Temporal Attention-enhanced Variational Graph Recurrent Neural Network (TAVRNN), a novel framework that models time-varying neuronal connectivity by integrating probabilistic graph learning with temporal attention mechanisms. TAVRNN learns latent dynamics at the single-unit level while maintaining interpretable population-level representations, to identify key connectivity patterns linked to behavior. TAVRNN generalizes across diverse neural systems and modalities, demonstrating state-of-the-art classification and clustering performance. We validate TAVRNN on three diverse datasets: (1) electrophysiological data from a freely behaving rat, (2) primate somatosensory cortex recordings during a reaching task, and (3) biological neurons in the DishBrain platform interacting with a virtual game environment. Our method outperforms state-of-the-art dynamic embedding techniques, revealing previously unreported relationships between adaptive behavior and the evolving topological organization of neural networks. These findings demonstrate that TAVRNN offers a powerful and generalizable approach for modeling neural dynamics across experimental and synthetic biological systems. Its architecture is modality-agnostic and scalable, making it applicable across a wide range of neural recording platforms and behavioral paradigms.

Problem

• Model time-varying neuronal connectivity and behavior relationships
• Capture evolving neural structure in dynamic, uncertain settings
• Generalize across diverse neural systems and modalities

Innovation

• Integrates probabilistic graph learning with temporal attention
• Learns latent dynamics at single-unit level interpretably
• Generalizes across diverse neural systems and modalities