Go to ACL ARR 2025 July homepageA Review of Incorporating Psychological Theories in LLMs
Abstract: Psychological insights have long shaped pivotal NLP breakthroughs, including the cognitive underpinnings of attention mechanisms, formative reinforcement learning, and Theory of Mind-inspired social modeling. As Large Language Models (LLMs) continue to grow in scale and complexity, there is a rising consensus that psychology is essential for capturing human-like cognition, behavior, and interaction. This paper reviews how psychological theories can inform and enhance stages of LLM development, including data, pre-training, post-training, and evaluation&application. Our survey integrates insights from cognitive, developmental, behavioral, social, personality psychology, and psycholinguistics. Our analysis highlights current trends and gaps in how psychological theories are applied. By examining both cross-domain connections and points of tension, we aim to bridge disciplinary divides and promote more thoughtful integration of psychology into future NLP research.
Paper Type: Long
Research Area: Resources and Evaluation
Research Area Keywords: Large Language Models, Psychology, Interdisciplinary Collaboration
Contribution Types: Surveys
Languages Studied: English
Reassignment Request Area Chair: This is not a resubmission
Reassignment Request Reviewers: This is not a resubmission
A1 Limitations Section: This paper has a limitations section.
A2 Potential Risks: N/A
A2 Elaboration: This paper is a review article and does not present new frameworks, datasets, or models that could introduce risks.
B Use Or Create Scientific Artifacts: No
B1 Cite Creators Of Artifacts: N/A
B1 Elaboration: This paper is a review article and does not use or create new scientific artifacts.
B2 Discuss The License For Artifacts: N/A
B2 Elaboration: This paper is a review article and does not use or create new scientific artifacts.
B3 Artifact Use Consistent With Intended Use: N/A
B3 Elaboration: This paper is a review article and does not use scientific artifacts.
B4 Data Contains Personally Identifying Info Or Offensive Content: N/A
B4 Elaboration: This paper is a review article and does not use or collect any data
B5 Documentation Of Artifacts: N/A
B5 Elaboration: This paper does not use or create new scientific artifacts.
B6 Statistics For Data: N/A
B6 Elaboration: This paper is a review article and does not use or collect any data
C Computational Experiments: No
C1 Model Size And Budget: N/A
C1 Elaboration: This paper is a review article and does not perform experiments.
C2 Experimental Setup And Hyperparameters: N/A
C2 Elaboration: This paper is a review article and does not perform experiments.
C3 Descriptive Statistics: N/A
C3 Elaboration: This paper is a review article and does not perform experiments.
C4 Parameters For Packages: N/A
C4 Elaboration: This paper is a review article and does not perform experiments.
D Human Subjects Including Annotators: No
D1 Instructions Given To Participants: N/A
D1 Elaboration: This paper is a review article and does not include human subjects.
D2 Recruitment And Payment: N/A
D2 Elaboration: This paper is a review article and does not include human subjects.
D3 Data Consent: N/A
D3 Elaboration: This paper is a review article and does not collect data.
D4 Ethics Review Board Approval: N/A
D4 Elaboration: This paper is a review article and does not collect data.
D5 Characteristics Of Annotators: N/A
D5 Elaboration: This paper is a review article and does not include human subjects.
E Ai Assistants In Research Or Writing: Yes
E1 Information About Use Of Ai Assistants: No
E1 Elaboration: ChatGPT was used only for minor language polishing. We did not explicitly include this in the paper as it did not affect research design, analysis, or conclusions.
Author Submission Checklist: yes
Submission Number: 753
Loading