Extracting Root-Causal Brain Activity Driving Psychopathology from Resting State fMRI

2026-02-06

AI Summary

Traditional neuroimaging studies rely on diagnostic labels or composite symptom scores, which often obscure the underlying causal mechanisms of psychiatric disorders. To address this limitation, this work proposes SOURCE, a method that constructs a two-layer structural causal model integrating inter-individual symptom architecture with intra-individual resting-state fMRI connectivity. By imposing constraints based on independent latent variables and local direct effects, SOURCE identifies brain regions whose activity causally drives specific symptom dimensions. This approach is the first to extract anatomically localized BOLD perturbation maps from resting-state fMRI that selectively associate with distinct symptom dimensions, demonstrating superior anatomical specificity and causal interpretability compared to existing methods. Moreover, it successfully recapitulates focal brain activity patterns consistent with the root-cause hypothesis of psychiatric symptoms.

Technology Category

• Machine Learning: Causal Learning
• Cognitive Modeling & Cognitive Systems: Symbolic Representations
• Reasoning under Uncertainty: Causality

Application Category

• User Modeling, Personalization and Recommendation: Psychology-informed user models and recommender systems
• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs
• Web Mining and Content Analysis: Robustness and generalizability of Web mining methods

Abstract

Neuroimaging studies of psychiatric disorders often correlate imaging patterns with diagnostic labels or composite symptom scores, yielding diffuse associations that obscure underlying mechanisms. We instead seek to identify root-causal maps -- localized BOLD disturbances that initiate pathological cascades -- and to link them selectively to symptom dimensions. We introduce a bilevel structural causal model that connects between-subject symptom structure to within-subject resting-state fMRI via independent latent sources with localized direct effects. Based on this model, we develop SOURCE (Symptom-Oriented Uncovering of Root-Causal Elements), a procedure that links interpretable symptom axes to a parsimonious set of localized drivers. Experiments show that SOURCE recovers localized maps consistent with root-causal BOLD drivers and increases interpretability and anatomical specificity relative to existing comparators.

Problem

• root-causal brain activity
• psychopathology
• resting-state fMRI
• symptom dimensions
• BOLD disturbances

Innovation

• root-causal inference
• resting-state fMRI
• structural causal model
• symptom dimensions
• localized BOLD drivers