Using Large Language Models to Simulate Multiple Humans and Replicate Human Subject Studies
This work addresses the problem of assessing LLM capabilities for simulating human behavior in research contexts, with potential impacts on applications in education and the arts, though it is incremental in extending existing evaluation methods.
The authors introduced Turing Experiments (TEs) to evaluate how well large language models (LLMs) like GPT can simulate human behavior and replicate findings from classic studies in economics, psycholinguistics, and social psychology, such as the Ultimatum Game and Milgram Shock Experiment, while identifying a 'hyper-accuracy distortion' in models like ChatGPT and GPT-4.
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.