psychology ↔ Large Language Models

Relations (1)

related 0.40 — supporting 4 facts

Psychology informs the development and evaluation of Large Language Models by providing frameworks for human cognition and personality representation as described in [1] and [2]. Furthermore, the two fields are linked through comparative studies of memory structures [3] and academic research exploring the application of LLMs within the psychological discipline [4].

Facts (4)

Sources

A Survey of Incorporating Psychological Theories in LLMs - arXiv arxiv.org arXiv 4 facts

referenceDorottya Demszky, Diyi Yang, David S. Yeager, Christopher J. Bryan, Margarett Clapper, Susannah Chandhok, Johannes C. Eichstaedt, Cameron Hecht, Jeremy Jamieson, and Meghann Johnson authored 'Using large language models in psychology', published in Nature Reviews Psychology in 2023.

claimIn psychology, memory entails structured encoding and recall, whereas in LLMs, memory typically refers to context windows or parameters.

claimPsychology can inform the development of Large Language Models (LLMs) in three key areas: evaluating emergent capabilities like reasoning, improving task performance in domains involving human cognition, and designing socially aware, multi-agent systems.

claimDevelopmental models in psychology could enable more coherent and interpretable personality representations in LLMs, offering a deeper alternative to static prompt-based personas.