Join us on October 10, 2024, when John Horton of the MIT Sloan School of Management stops by the Lab for his talk, “Automated Social Science: Language Models as Scientist and Subjects.”
This is a hybrid event open to everyone. Click here to register to watch on Zoom.
Members of the Stanford community are invited to join us in person. Register to attend by filling out this form.
We present an approach for automatically generating and testing, in silico, social scientific hypotheses. This automation is made possible by recent advances in large language models (LLM), but the key feature of the approach is the use of structural causal models. Structural causal models provide a language to state hypotheses, a blueprint for constructing LLM-based agents, an experimental design, and a plan for data analysis. The fitted structural causal model becomes an object available for prediction or the planning of follow-on experiments. We demonstrate the approach with several scenarios: a negotiation, a bail hearing, a job interview, and an auction. In each case, causal relationships are both proposed and tested by the system, finding evidence for some and not others. We provide evidence that the insights from these simulations of social interactions are not available to the LLM purely through direct elicitation. When given its proposed structural causal model for each scenario, the LLM is good at predicting the signs of estimated effects, but it cannot reliably predict the magnitudes of those estimates. In the auction experiment, the in silico simulation results closely match the predictions of auction theory, but elicited predictions of the clearing prices from the LLM are inaccurate. However, the LLM’s predictions are dramatically improved if the model can condition on the fitted structural causal model. In short, the LLM knows more than it can (immediately) tell.
John Horton is an Associate Professor of Information Technologies at the MIT Sloan School of Management.
Horton’s research focuses on the intersection of labor economics, market design, and information systems. He is particularly interested in improving the efficiency and equity of matching markets.
After completing his PhD and prior to joining NYU Stern School of Business in 2013, he served for two years as the staff economist for oDesk, an online labor market.
Horton received a BS in mathematics from the United States Military Academy at West Point and a PhD in public policy from Harvard University.