Cognitive Foundations for Reasoning and Their Manifestation in LLMs
cs.AI
Submitted: 2025-11-20
Updated: 2026-11-05
Comments: 39 pages, 9 tables, 9 figures
Code: https://github.com/pkargupta/cognitive_foundations
License: http://creativecommons.org/licenses/by/4.0/
The gist: Large language models (LLMs) solve complex problems yet fail on simpler variants, suggesting they achieve correct outputs through mechanisms fundamentally different from human reasoning.
Terminology
Abstract
Large language models (LLMs) solve complex problems yet fail on simpler variants, suggesting they achieve correct outputs through mechanisms fundamentally different from human reasoning. To understand this gap, we synthesize cognitive science research into a taxonomy of 28 cognitive elements spanning reasoning invariants, meta-cognitive controls, representations for organizing reasoning and knowledge, and transformation operations. We introduce a fine-grained evaluation framework and apply it to 192K traces from 18 models across text, vision, and audio, alongside 1,346 human think-aloud traces on a new, domain-agnostic dataset for human-model comparison. We find that models are able to deploy cognitive elements seen in human reasoning but differ in how they use them. Leveraging these patterns, we develop test-time reasoning guidance that automatically scaffolds successful reasoning topologies, improving the accuracy of the strongest models by 9.7 percentage points on ill-structured problems. Meta-analysis of 1.6K LLM reasoning papers reveals the research community concentrates on easily quantifiable elements but rarely targets self-awareness, whose use differs between human and model reasoning. By establishing a shared vocabulary between cognitive science and LLM research, our framework enables diagnosis of reasoning failures and principled development of models that reason through robust cognitive mechanisms rather than spurious shortcuts, while providing tools to test theories of human cognition at scale.
Sources
- BLAB: Brutally Long Audio Bench
- On the Bias of Next-Token Predictors Toward Systematically Inefficient Reasoning: A Shortest-Path Case Study
- The pitfalls of next-token prediction
- The Reversal Curse: LLMs trained on "A is B" fail to learn "B is A"
- Gold-medalist Performance in Solving Olympiad Geometry with AlphaGeometry2
- Can Memory-Augmented Language Models Generalize on Reasoning-in-a-Haystack Tasks?
- DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning
- Length-Controlled AlpacaEval: A Simple Way to Debias Automatic Evaluators
- Cognitive Behaviors that Enable Self-Improving Reasoners, or, Four Habits of Highly Effective STaRs
- OpenThoughts: Data Recipes for Reasoning Models
- Fluid Language Model Benchmarking
- A Survey on Large Language Models for Code Generation
- Large Language Models are Zero-Shot Reasoners
- MMAU-Pro: A Challenging and Comprehensive Benchmark for Holistic Evaluation of Audio General Intelligence
- Tulu 3: Pushing Frontiers in Open Language Model Post-Training
- Measuring Faithfulness in Chain-of-Thought Reasoning
- Reasoning Under 1 Billion: Memory-Augmented Reinforcement Learning for Large Language Models
- Zebra-CoT: A Dataset for Interleaved Vision Language Reasoning
- PrefDisco: Benchmarking Proactive Personalized Reasoning
- PrefPalette: Personalized Preference Modeling with Latent Attributes
Related papers
- Multi-Objective Structured Pruning of LLMs for Latency and Model Size Optimization
- MAVEN-T: Reinforced Heterogeneous Distillation for Real-Time Multi-Agent Trajectory Prediction
- Model Discovery Agent: LLM-assisted Bayesian experiment design for data-efficient discovery of mechanistic world models
- Embeddings for Preferences, Not Semantics
- Can We Trust Item Response Theory for AI Evaluation?
- Diagnosing Task Insensitivity in Language Agents