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

2026-03-18

AI Summary

Current automated screening approaches for Alzheimer’s disease rely heavily on statistical pattern recognition, often compromising construct validity and lacking interpretability and clinical alignment. This work proposes Agentic Cognitive Profiling, a novel framework that aligns multi-agent large language models with standardized cognitive assessment protocols for the first time. By atomically decomposing cognitive tasks, specialized agents extract verifiable scoring primitives, which are then quantified through deterministic functions, effectively decoupling semantic understanding from measurement. Evaluated on clinical data from 402 participants across eight cognitive tasks, the method achieves a 90.5% task-level scoring agreement and an 85.3% accuracy in disease prediction—significantly outperforming baseline models—while simultaneously preserving predictive performance, clinical logical consistency, and interpretability.

Technology Category

• Cognitive Modeling & Cognitive Systems: Agent Architectures
• Multiagent Systems: Adversarial Agents
• Machine Learning: Calibration & Uncertainty Quantification

Application Category

• Search and Retrieval-Augmented AI: Agentic search
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Responsible Web: Machine-in-the-loop, human agency and autonomy

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.

Problem

• Alzheimer's Disease
• construct validity
• automated screening
• cognitive assessment
• clinical protocol

Innovation

• Agentic Cognitive Profiling
• construct validity
• deterministic function calling
• cognitive task decomposition
• interpretable AI