Can Large Language Models Simulate Human Cognition Beyond Behavioral Imitation?
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866914430057447424 |
|---|---|
| author | Gu, Yuxuan Liu, Lunjun Feng, Xiaocheng Zhu, Kun Zhong, Weihong Huang, Lei Qin, Bing |
| author_facet | Gu, Yuxuan Liu, Lunjun Feng, Xiaocheng Zhu, Kun Zhong, Weihong Huang, Lei Qin, Bing |
| contents | An essential problem in artificial intelligence is whether LLMs can simulate human cognition or merely imitate surface-level behaviors, while existing datasets suffer from either synthetic reasoning traces or population-level aggregation, failing to capture authentic individual cognitive patterns. We introduce a benchmark grounded in the longitudinal research trajectories of 217 researchers across diverse domains of artificial intelligence, where each author's scientific publications serve as an externalized representation of their cognitive processes. To distinguish whether LLMs transfer cognitive patterns or merely imitate behaviors, our benchmark deliberately employs a cross-domain, temporal-shift generalization setting. A multidimensional cognitive alignment metric is further proposed to assess individual-level cognitive consistency. Through systematic evaluation of state-of-the-art LLMs and various enhancement techniques, we provide a first-stage empirical study on the questions: (1) How well do current LLMs simulate human cognition? and (2) How far can existing techniques enhance these capabilities? |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2603_27694 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Can Large Language Models Simulate Human Cognition Beyond Behavioral Imitation? Gu, Yuxuan Liu, Lunjun Feng, Xiaocheng Zhu, Kun Zhong, Weihong Huang, Lei Qin, Bing Computation and Language An essential problem in artificial intelligence is whether LLMs can simulate human cognition or merely imitate surface-level behaviors, while existing datasets suffer from either synthetic reasoning traces or population-level aggregation, failing to capture authentic individual cognitive patterns. We introduce a benchmark grounded in the longitudinal research trajectories of 217 researchers across diverse domains of artificial intelligence, where each author's scientific publications serve as an externalized representation of their cognitive processes. To distinguish whether LLMs transfer cognitive patterns or merely imitate behaviors, our benchmark deliberately employs a cross-domain, temporal-shift generalization setting. A multidimensional cognitive alignment metric is further proposed to assess individual-level cognitive consistency. Through systematic evaluation of state-of-the-art LLMs and various enhancement techniques, we provide a first-stage empirical study on the questions: (1) How well do current LLMs simulate human cognition? and (2) How far can existing techniques enhance these capabilities? |
| title | Can Large Language Models Simulate Human Cognition Beyond Behavioral Imitation? |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2603.27694 |