Using Large Language Models to Simulate Multiple Humans and Replicate Human Subject Studies

Gati V Aher, Rosa I. Arriaga, Adam Tauman Kalai
Proceedings of the 40th International Conference on Machine Learning, PMLR 202:337-371, 2023.

Abstract

We introduce a new type of test, called a Turing Experiment (TE), for evaluating to what extent a given language model, such as GPT models, can simulate different aspects of human behavior. A TE can also reveal consistent distortions in a language model’s simulation of a specific human behavior. Unlike the Turing Test, which involves simulating a single arbitrary individual, a TE requires simulating a representative sample of participants in human subject research. We carry out TEs that attempt to replicate well-established findings from prior studies. We design a methodology for simulating TEs and illustrate its use to compare how well different language models are able to reproduce classic economic, psycholinguistic, and social psychology experiments: Ultimatum Game, Garden Path Sentences, Milgram Shock Experiment, and Wisdom of Crowds. In the first three TEs, the existing findings were replicated using recent models, while the last TE reveals a “hyper-accuracy distortion” present in some language models (including ChatGPT and GPT-4), which could affect downstream applications in education and the arts.

Cite this Paper

BibTeX
@InProceedings{pmlr-v202-aher23a, title = {Using Large Language Models to Simulate Multiple Humans and Replicate Human Subject Studies}, author = {Aher, Gati V and Arriaga, Rosa I. and Kalai, Adam Tauman}, booktitle = {Proceedings of the 40th International Conference on Machine Learning}, pages = {337--371}, year = {2023}, editor = {Krause, Andreas and Brunskill, Emma and Cho, Kyunghyun and Engelhardt, Barbara and Sabato, Sivan and Scarlett, Jonathan}, volume = {202}, series = {Proceedings of Machine Learning Research}, month = {23--29 Jul}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v202/aher23a/aher23a.pdf}, url = {https://proceedings.mlr.press/v202/aher23a.html}, abstract = {We introduce a new type of test, called a Turing Experiment (TE), for evaluating to what extent a given language model, such as GPT models, can simulate different aspects of human behavior. A TE can also reveal consistent distortions in a language model’s simulation of a specific human behavior. Unlike the Turing Test, which involves simulating a single arbitrary individual, a TE requires simulating a representative sample of participants in human subject research. We carry out TEs that attempt to replicate well-established findings from prior studies. We design a methodology for simulating TEs and illustrate its use to compare how well different language models are able to reproduce classic economic, psycholinguistic, and social psychology experiments: Ultimatum Game, Garden Path Sentences, Milgram Shock Experiment, and Wisdom of Crowds. In the first three TEs, the existing findings were replicated using recent models, while the last TE reveals a “hyper-accuracy distortion” present in some language models (including ChatGPT and GPT-4), which could affect downstream applications in education and the arts.} }
Endnote
%0 Conference Paper %T Using Large Language Models to Simulate Multiple Humans and Replicate Human Subject Studies %A Gati V Aher %A Rosa I. Arriaga %A Adam Tauman Kalai %B Proceedings of the 40th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2023 %E Andreas Krause %E Emma Brunskill %E Kyunghyun Cho %E Barbara Engelhardt %E Sivan Sabato %E Jonathan Scarlett %F pmlr-v202-aher23a %I PMLR %P 337--371 %U https://proceedings.mlr.press/v202/aher23a.html %V 202 %X We introduce a new type of test, called a Turing Experiment (TE), for evaluating to what extent a given language model, such as GPT models, can simulate different aspects of human behavior. A TE can also reveal consistent distortions in a language model’s simulation of a specific human behavior. Unlike the Turing Test, which involves simulating a single arbitrary individual, a TE requires simulating a representative sample of participants in human subject research. We carry out TEs that attempt to replicate well-established findings from prior studies. We design a methodology for simulating TEs and illustrate its use to compare how well different language models are able to reproduce classic economic, psycholinguistic, and social psychology experiments: Ultimatum Game, Garden Path Sentences, Milgram Shock Experiment, and Wisdom of Crowds. In the first three TEs, the existing findings were replicated using recent models, while the last TE reveals a “hyper-accuracy distortion” present in some language models (including ChatGPT and GPT-4), which could affect downstream applications in education and the arts.
APA
Aher, G.V., Arriaga, R.I. & Kalai, A.T.. (2023). Using Large Language Models to Simulate Multiple Humans and Replicate Human Subject Studies. Proceedings of the 40th International Conference on Machine Learning, in Proceedings of Machine Learning Research 202:337-371 Available from https://proceedings.mlr.press/v202/aher23a.html.

Related Material