One of the most potentially profound, but usually overlooked, aspects of public health interventions is the possibility that the intervention will change not only the behavior of individuals in a community, but the social structure of the community itself, even affecting people who were not targets of the intervention. This (R01) project will build on an existing interventional study and create a new dataset, capturing how the structures of real-world networks and the position of aging adults within them change over time in response to a randomized exposure to a health intervention. We will use this data set to answer important questions about how the positions and roles of older adults are affected by public health interventions.

Microbiome Biology and Social Networks in the Developing World

We will merge human genomic, microbiome, and social network data to examine important relationships among our own genes, the organisms living in our bodies, and our social connections to one another. The project aims to add significantly to the current understanding of how the microbiome affects and is affected by human social connections by exploring how the microbiota moves across social network connections and how that spread of the symbiotic microbial environment affects humans.

“Hybrid Systems” of Bots and Humans

In this project, we will develop and execute game theory-driven experimental scenarios (using a sophisticated online experiment system that we have developed) to improve our understanding of how groups of interconnected individuals can better solve collective problems of diverse, important sorts. We hope to invent new AI systems to help with collective action and to make groups work better together.

Developing and Validating a Measure of Intellectual Humility in Traditional Villages of Rural Honduras

Intellectual humility (IH) has emerged as an important construct with particular relevance to many aspects of core human virtues. But efforts to study IH in non-Western contexts and to explore the relevance of IH to various daily practical decisions are still scarce. We will use a sample drawn from our ongoing study exploring changes in health knowledge and behavior of 30,862 people in 176 villages in rural Honduras. In measuring openness to new ideas, practices, and norms, we will broadly examine the potential importance of IH in the acquisition of knowledge important to the wellbeing of individuals and communities.

Yale University and the Tata group have launched a far-ranging research collaboration that builds on their shared strengths in discovery, technology, and innovation. The alliance is led at Yale by Nicholas Christakis ’84, co-director of the Yale Institute for Network Science (YINS) and Sterling Professor of Social and Natural Science. The Tata group has committed to fund the alliance over five years. The project will foster an intellectual exchange between the university and three Tata companies: Tata Sons, Tata Consultancy Services, and Tata Chemicals.

Hybrid Human-AI Systems to Change Collective Behavior

Building on prior work in the Human Nature Lab, we will explore how artificial intelligence agents (also known as “bots”), placed within online social groups, can affect individual behavior and group performance. We will thus study “hybrid systems” composed of humans and bots interacting in complex ways. We will advance our understanding of what is likely to become an even more widespread phenomenon in the coming years, and we will develop applied technologies that could be useful in advancing the welfare of individuals, communities, and societies.

Assessing the Hazard of Elections During the COVID-19 Pandemic

In this project, we predicted that a second wave of the COVID pandemic, in keeping with past respiratory pandemics, would arrive in the US in the fall of 2020. Thus it was likely to coincide with the run-up to the presidential election, which would posse particular challenges. The pandemic behaved as predicted, and now we are evaluating how the process of assembling to vote affects subsequent geographically bounded spikes in COVID-19 cases, and, conversely, how the potential fear of assembling, given the presence of the virus, affected the process of voting, focusing on voter turnout.

Exploiting Social Influence in Networks to Magnify Population-Level MNCH Behavior Change

We will use a detailed understanding of social network structure and function to identify novel ways of targeting influential individuals so as to foster behavioral cascades and population-level behavior change. We will achieve this objective by conducting a randomized controlled trial of network targeting algorithms, to be deployed in a sample of 160 villages in Honduras, with maternal and neonatal care interventions.