Harnessing Artificial Intelligence to enhance the power of experimental social science research to improve the human condition
About our project:
Talking to Machines
INTRODUCTION TO TTM
“Talking to Machines” is a project that aims to leverage artificial intelligence in the design and implementation of experiments. This project is a unique combination of both academic and industry researchers working in very diverse institutional contexts throughout the world. We are an international collaboration with a talented global team of faculty, staff, postdocs, and graduate and undergraduate students. Research results from these studies have been published widely in leading social science peer-reviewed publications.
Through our commitment to evidence-based policies, we engage with national and international governmental agencies to evaluate the effectiveness of current policies and design new programs and tools with the potential to ultimately affect the lives of millions of individuals.
The project is led by Sonja Vogt at HEC University of Lausanne and Ray Duch at Nuffield College University of Oxford. Our primary funding is from the Swiss National Science Foundation (#100018M-215519).
An important challenge in designing public policies is the effective communication of policy-related information to the average citizen.
Our Focus
Our focus is on the information content of video communications. People differ in many ways that are relevant when it comes to behavioral responses to the information content of videos. Any intervention – nudges, edutainment, information campaigns, or legislative changes – has heterogeneous effects among the targeted population.
The Talking to Machines project uses recent methodological advances to create a platform for designing experiments with information treatment arms that consist of video narratives.
How do we know what information should be incorporated into a particular treatment and how should the information be framed and delivered?
Moreover, how can we address heterogeneous attitudes and needs when developing communication campaigns for diverse societies?
Our Goal
We tackle both challenges, namely optimal design of information treatment arms that are sensitive to heterogeneous treatment effects.
The project aims to incorporate intelligent technologies in the design of experiments informing policies related to human decision making in areas such as health, personal finances, and education. We combine narrative information with intelligent technologies to develop effective video interventions for heterogeneous populations.
We leverage the significant ongoing advances in computational social science that provide the tools for optimizing the design of digital content and the development of information interventions.
Important elements of the project include:
- Automating the conversations with subjects who participate in these experiments
- Programming ChatBots to facilitate these conversations
- Generating large corpuses of text and voice data
- Large language models (LLM)
- Machine learning data analytics
- Machine-enhanced design of video treatments
