Modeling Multi-Dimensional Cognitive States in Large Language Models under Cognitive Crowding

2026-04-18

AI Summary

Current large language models struggle to effectively model multidimensional cognitive states—such as emotion, thinking style, stance, and intent—as defined in psychology, exhibiting significant performance bottlenecks. This work proposes the “cognitive crowding” hypothesis, revealing how limited representational capacity in Euclidean space causes interference among these cognitive dimensions. To systematically investigate this phenomenon, we introduce CognitiveBench, the first unified benchmark with multidimensional cognitive annotations. Addressing this challenge, we propose HyCoLLM, a novel approach that leverages hyperbolic geometry for its hierarchical expressiveness, incorporating Gromov δ-hyperbolicity analysis and Hyperbolic Guided Alignment Tuning. Experimental results demonstrate that HyCoLLM substantially enhances multidimensional cognitive understanding, enabling an 8B-parameter model to outperform strong baselines, including GPT-4o.

Technology Category

• Machine Learning: Large Multimodal Models (LMMs)
• Cognitive Modeling & Cognitive Systems: Analogy
• Natural Language Processing: (Large) Language Models

Application Category

• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Economics, Online Markets and Human Computation: Humans versus LLMs for data annotation and labeling
• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation

Abstract

Modeling human cognitive states is essential for advanced artificial intelligence. Existing Large Language Models (LLMs) mainly address isolated tasks such as emotion analysis or stance detection, and fail to capture interactions among cognitive dimensions defined in psychology, including emotion, thinking style, stance, and intention. To bridge this gap, we construct CognitiveBench, the first benchmark with unified annotations across the above four dimensions. Experiments on CognitiveBench show that although LLMs perform well on single dimension tasks, their performance drops sharply in joint multi-dimensional modeling. Using Gromov $δ$-hyperbolicity analysis, we find that CognitiveBench exhibits a strong hierarchical structure. We attribute the performance bottleneck to ``Cognitive Crowding'', where hierarchical cognitive states require exponential representational space, while the Euclidean space of LLMs grows only polynomially, causing representation overlap and degraded performance. To address this mismatch, we propose HyCoLLM, which models cognitive states in hyperbolic space and aligns LLM representations via Hyperbolic Guided Alignment Tuning. Results show that HyCoLLM substantially improves multi-dimensional cognitive understanding, allowing 8B parameter model to outperform strong baselines, including GPT-4o.

Problem

• cognitive states
• large language models
• multi-dimensional modeling
• cognitive crowding
• hierarchical structure

Innovation

• Cognitive Crowding
• Hyperbolic Space
• Multi-Dimensional Cognitive Modeling
• CognitiveBench
• HyCoLLM