Bayesian Inference of Psychometric Variables From Brain and Behavior in Implicit Association Tests

2026-03-17

AI Summary

This work proposes a sparse hierarchical Bayesian model that extends the conventional D-score in Implicit Association Tests (IATs) by treating it as a learnable parameter, thereby integrating multimodal neural and behavioral data to enhance the inference of psychometric constructs linked to mental health symptoms—even in small-sample IAT settings. The approach requires no task-specific fine-tuning and demonstrates both parameter efficiency and strong cross-task generalizability. Evaluated on E-IAT and PSY-IAT tasks, the method achieves AUCs of 0.73 and 0.76, respectively. Notably, for individuals with major depressive disorder (MDD), it elevates E-IAT performance to an AUC of 0.79, substantially outperforming the original D-score (AUC: 0.50–0.53) and matching the performance of optimized baselines such as shrinkage LDA and EEGNet.

Technology Category

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

Application Category

• User Modeling, Personalization and Recommendation: Metrics for user behavior and evaluating success
• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets
• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics

Abstract

Objective. We establish a principled method for inferring mental health related psychometric variables from neural and behavioral data using the Implicit Association Test (IAT) as the data generation engine, aiming to overcome the limited predictive performance (typically under 0.7 AUC) of the gold-standard D-score method, which relies solely on reaction times. Approach. We propose a sparse hierarchical Bayesian model that leverages multi-modal data to predict experiences related to mental illness symptoms in new participants. The model is a multivariate generalization of the D-score with trainable parameters, engineered for parameter efficiency in the small-cohort regime typical of IAT studies. Data from two IAT variants were analyzed: a suicidality-related E-IAT ($n=39$) and a psychosis-related PSY-IAT ($n=34$). Main Results. Our approach overcomes a high inter-individual variability and low within-session effect size in the dataset, reaching AUCs of 0.73 (E-IAT) and 0.76 (PSY-IAT) in the best modality configurations, though corrected 95% confidence intervals are wide ($\pm 0.18$) and results are marginally significant after FDR correction ($q=0.10$). Restricting the E-IAT to MDD participants improves AUC to 0.79 $[0.62, 0.97]$ (significant at $q=0.05$). Performance is on par with the best reference methods (shrinkage LDA and EEGNet) for each task, even when the latter were adapted to the task, while the proposed method was not. Accuracy was substantially above near-chance D-scores (0.50-0.53 AUC) in both tasks, with more consistent cross-task performance than any single reference method. Significance. Our framework shows promise for enhancing IAT-based assessment of experiences related to entrapment and psychosis, and potentially other mental health conditions, though further validation on larger and independent cohorts will be needed to establish clinical utility.

Problem

• Implicit Association Test
• psychometric variables
• Bayesian inference
• mental health
• D-score

Innovation

• Bayesian inference
• Implicit Association Test
• multimodal integration
• psychometric prediction
• sparse hierarchical model