Agentic Cognitive Profiling: Realigning Automated Alzheimer's Disease Detection with Clinical Construct Validity

TL;DR

This paper introduces Agentic Cognitive Profiling, a novel framework for Alzheimer's screening that aligns with clinical protocols, improves interpretability, and enhances predictive accuracy by decomposing assessments into verifiable cognitive primitives.

Contribution

The paper proposes a new agentic framework that restores construct validity in automated AD detection by decomposing assessments into atomic tasks and using deterministic functions for measurement.

Findings

Achieves 90.5% score match rate in task examination

Attains 85.3% accuracy in AD prediction

Outperforms popular baselines in interpretability and accuracy

Abstract

Automated Alzheimer's Disease (AD) screening has predominantly followed the inductive paradigm of pattern recognition, which directly maps the input signal to the outcome label. This paradigm sacrifices construct validity of clinical protocol for statistical shortcuts. This paper proposes Agentic Cognitive Profiling (ACP), an agentic framework that realigns automated screening with clinical protocol logic across multiple cognitive domains. Rather than learning opaque mappings from transcripts to labels, the framework decomposes standardized assessments into atomic cognitive tasks and orchestrates specialized LLM agents to extract verifiable scoring primitives. Central to our design is decoupling semantic understanding from measurement by delegating all quantification to deterministic function calling, thereby mitigating hallucination and restoring construct validity. Unlike popular datasets that typically comprise around a hundred participants under a single task, we evaluate on a clinically-annotated corpus of 402 participants across eight structured cognitive tasks spanning multiple cognitive domains. The framework achieves 90.5% score match rate in task examination and 85.3% accuracy in AD prediction, surpassing popular baselines while generating interpretable cognitive profiles grounded in behavioral evidence. This work demonstrates that construct validity and predictive performance need not be traded off, charting a path toward AD screening systems that explain rather than merely predict.