Despite theory tying wealth inequality to social structure, empirical evidence has been limited to a few studies based on online social media data. This study uses a very different type of data, expands the global coverage to very different types of societies, and investigates new questions. In particular, we collect data from 3500 sharing units (households) in 46 communities across the globe, representing considerable human social and cultural diversity. In each, we analyze the relationship between people's material wealth and the structure of social networks: borrowing money, sharing food, working together, socializing, etc. In almost all communities, a sharing unit's material wealth is positively associated with the number of other sharing units it both helps and is helped by. A sharing unit's wealth is also associated with the relative wealth of the sharing units to which it is linked—a form of economic homophily. Notably, communities with greater wealth inequality are also characterized by a network structure in which poorer sharing units are less well connected to wealthier ones. We augment our unique cross-cultural data with other community-level environmental, institutional, and economic attributes, opening new avenues for future research into the co-determination of wealth and social networks.

We study the consequences of job markets' heavy reliance on referrals. Referrals lead to more opportunities for workers to be hired, which lead to better matches and increased productivity but also disadvantage job seekers with few or no connections to employed workers, increasing inequality. Coupled with homophily, referrals also lead to immobility. We identify conditions under which distributing referrals more evenly reduces inequality and improves future productivity and mobility. We use the model to examine the short- and long-run welfare impacts of policies such as affirmative action and algorithmic fairness.

We study multiplexing-the extent to which people have multiple, distinct types of relationships with the same others. We document empirical multiplexing patterns in Indian village data. We show that relationships such as socializing, advising, helping, and lending are correlated but distinct, while commonly used proxies for networks based on ethnicity and geography are nearly uncorrelated with actual relationships. We also show that these layers and their overlap affect information diffusion in a field experiment. On the theoretical side, we show that increased overlap in people's different types of relationships impedes the spread of simple contagions-such as diseases or basic information that spread through a single interaction. In contrast, such higher levels of multiplexing enhance the spread of a complex contagion (e.g., a behavior or belief that requires multiple interactions for transmission) when infection rates are low, but impede complex contagion if infection rates are high. Finally, we identify empirical differences in multiplexing by gender and connectedness.

People learn about opportunities and actions by observing the experiences of their friends. We model how homophily – the tendency to associate with similar others – affects both the endogenous quality and diversity of the information accessible to decision makers. Homophily provides higher-quality information, since observing the payoffs of another person is more informative the more similar that person is to the decision maker. However, homophily can lead people to take actions that generate less information. We show how network connectivity influences the tradeoff between the endogenous quantity and quality of information. Although homophily hampers learning in sparse networks, it enhances learning in sufficiently dense networks.

We examine the economic impact of increasingly productive AI and policies that spread its benefits across the economy. Improvements in AI productivity trigger labor reallocation and changes in absolute and relative wages for different types of labor. Wages of labor that is essential for building AI increase faster than overall GDP. Wages of labor that is substituted for by AI decrease in both absolute and relative terms. Wages of labor that is used only in final goods production and is not displaced by AI increase in line with overall GDP. We contrast the impact of productivity gains depending on whether AI production is competitive or monopolistic. Monopoly production of AI restricts its deployment, slowing the transition and impact of AI. Optimal tax and regulatory policies that achieve Pareto-improvements differ depending on whether there is competition in AI production.

We provide a first causal analysis of the behavioral consequences of the friendship paradox—the fact that people’s friends in a network have more connections than average. We find that people’s behavior is biased by their network position: they do not best respond to what they should infer the average behavior of the population to be, but instead simply to the average behavior of their friends. Moreover, we find that they fail to learn to overcome such a bias when relocated within the network, varying their observational environment. In these games of complements, the friendship paradox generates a systematic upward distortion in actions, increases behavioral dispersion, and persists despite learning opportunities.

We examine how agents learn when information from original sources only reaches them after noisy relay. A receiver learns if and only if they have access to sufficiently many chains of noisy relay and they perfectly understand the noise process. However, even slight uncertainty over message mutation rates makes learning from long chains impossible, no matter how many independent sources are accessed.

Foundation models have been increasingly applied to behavioral science domains such as psychology, sociology, and economics. While these models show promise in individual tasks such as survey response prediction and human-subject experiment simulation, there remains no systematic understanding of how well they perform across diverse behavioral science tasks, contexts, and populations. We introduce BehaviorBench, a comprehensive benchmark that evaluates foundation models along four core capabilities: (1) behavior prediction and simulation, (2) strategic decision-making, (3) subject-trait inference, and (4) behavioral knowledge application. Crucially, BehaviorBench evaluates model outputs at both the individual and distributional levels, capturing not only per-subject accuracy but also population-level alignment, an essential requirement for behavioral validity. Leveraging the tasks in BehaviorBench, we further develop Be.FM-1.5, extending the Be.FM family of behavioral foundation models fine-tuned on behavioral data. Our results reveal a considerable gap: proprietary general-purpose models excel at individual-level prediction and knowledge-intensive tasks, whereas behavioral foundation models, fine-tuned on behavioral data, achieve substantially stronger distributional alignment. Notably, Be.FM-1.5 leads on distributional metrics and remains competitive on individual-level metrics, suggesting that proper behavioral adaptation can close the gap. Our results highlight the importance of distributional evaluation, establish BehaviorBench as a foundation for developing and assessing behaviorally aligned AI systems, and demonstrate Be.FM-1.5's potential for a broad range of behavioral science studies. Our BehaviorBench and Be.FM-1.5 models can be accessed via https://umich-foreseer.github.io/behaviorbench/.

When people prefer to coordinate their behaviors with their friends---e.g., choosing whether to adopt a new technology, to protest against a government, to attend university---divisions within a social network can sustain different behaviors in different parts of the network. We define a society's `behavioral communities' via its network's `atoms': groups of people who adopt the same behavior in every equilibrium. We analyze how the atoms change with the intensity of the peer effects, and characterize the atoms in a prominent class of network models. We show that using knowledge of atoms to seed the diffusion of a behavior significantly increases diffusion compared to seeding based on standard community detection algorithms. We also show how to use observed behaviors to estimate the intensity of peer effects.

Do social networks and peer influence shape major life decisions in polarized settings? We explore this question by examining how peers influenced the allegiances of West Point cadets during the American Civil War. Leveraging quasi-random variations in the proportion of cadets from Free States, we analyze how cadets' decisions about which army to join depended on the composition of their peers. We have three main findings. First, there was a strong and significant peer effect: a higher proportion of classmates from Free States significantly increased the likelihood that cadets from Slave States joined the Union Army. Second, the peer effect varies with geography, most notably with the slave population share in cadets' home states or counties, and with cadets' own slave ownership in 1860. Third, peer effects were amplified by shared experiences such as having served together in the Mexican-American War, continuous military service, and belonging to the same cohort, suggesting that sustained interaction is important.

We examine optimal regulation of financial networks with debt interdependencies between financial firms. We first characterize when it is firms have an incentive to choose excessively risky portfolios and overly correlate their portfolios with those of their counterparties. We then characterize how optimal regulation depends on a firm's financial centrality and its available investment opportunities. In standard core-periphery networks, optimal regulation depends non-monotonically on the correlation of banks' investments, with maximal restrictions for intermediate levels of correlation. Moreover, it can be uniquely optimal to treat banks asymmetrically: restricting the investments of one core bank while allowing an otherwise identical core bank (in all aspects, including network centrality) to invest freely.

Young adults face a rising tide of mental illness and loneliness. We propose that an overlooked barrier for social connection is how people perceive each other's empathy. Here, our longitudinal study of an undergraduate student community (N = 5,192) reveals that undergraduates who perceive their peers as empathic report better current and future well-being. Yet we document an 'empathy perception gap': people systematically see others as less empathic than others see themselves. Students who perceived their peers as less empathic were less willing to take social risks and grew more isolated over time. To disrupt this cycle, we conducted two field experiments that presented students with data on their peers' self-reported empathy and behavioural nudges to encourage social risk taking. These interventions reduced the empathy perception gap, increased social behaviours and expanded social networks months later. This work offers a promising, scalable strategy to cultivate social well-being, simply by presenting people with data about each other.

We provide an overview of methods for designing and implementing experiments (field, lab, hybrid, and natural) when there are networks of interactions between subjects.

Despite their success in numerous fields, the potential of foundation models for modeling and understanding human behavior remains largely unexplored. We introduce Be.FM, one of the first open foundation models designed for human behavior modeling. Built upon open-source large language models and fine-tuned on a diverse range of behavioral data, Be.FM can be used to understand and predict human decision-making. We construct a comprehensive set of benchmark tasks for testing the capabilities of behavioral foundation models. Our results demonstrate that Be.FM can predict behaviors, infer characteristics of individuals and populations, generate insights about contexts, and apply behavioral science knowledge.

We discuss the three main areas comprising the new and emerging field of "AI Behavioral Science". This includes not only how AI can enhance research in the behavioral sciences, but also how the behavioral sciences can be used to study and better design AI and to understand how the world will change as AI and humans interact in increasingly layered and complex ways.

We evaluate a large-scale set of interventions to increase demand for immunization in Haryana, India. The policies under consideration include the two most frequently discussed tools--reminders and incentives--as well as an intervention inspired by the networks literature. We cross-randomize whether (a) individuals receive SMS reminders about upcoming vaccination drives; (b) individuals receive incentives for vaccinating their children; (c) influential individuals (information hubs, trusted individuals, or both) are asked to act as "ambassadors" receiving regular reminders to spread the word about immunization in their community. By taking into account different versions (or "dosages") of each intervention, we obtain 75 unique policy combinations. We develop a new statistical technique--a smart pooling and pruning procedure--for finding a best policy from a large set, which also determines which policies are effective and the effect of the best policy. We proceed in two steps. First, we use a LASSO technique to collapse the data: we pool dosages of the same treatment if the data cannot reject that they had the same impact, and prune policies deemed ineffective. Second, using the remaining (pooled) policies, we estimate the effect of the best policy, accounting for the winner's curse. The key outcomes are (i) the number of measles immunizations and (ii) the number of immunizations per dollar spent. The policy that has the largest impact (information hubs, SMS reminders, incentives that increase with each immunization) increases the number of immunizations by 44% relative to the status quo. The most cost-effective policy (information hubs, SMS reminders, no incentives) increases the number of immunizations per dollar by 9.1%.

I discuss economic and social sources of inequality and elaborate on the role of social networks in inequality, economic immobility, and economic inefficiencies. The lens of social networks clarifies how the entanglement of people's information, opportunities, and behaviors with those of their friends and family leads to persistent differences across communities, resulting in inequality in education, employment, income, health, and wealth. The key role of homophily in separating groups within the network is highlighted. A network perspective's policy implications differ substantially from a narrower economic perspective that ignores social structure. I discuss the importance of “policy cocktails” that include aspects that are aimed at both the economic and social forces driving inequality.

Low socioeconomic status (SES) undergraduates are often worse off in well-being than high-SES peers. These “well-being gaps” lessen when low-SES students self-report being socially connected; however, one’s perception of their own connectedness in a network differs from external proxies. Within a network of 785 undergraduates, we examine two social network measures of connection—self-reported number of friends (outdegree) and number of undergraduate peers who reported them as a friend (indegree). Low- (vs. high-) SES students have a lower outdegree yet have a similar indegree. Critically, low-SES students who report a lower outdegree are also poorer in well-being, even when controlling for indegree, though the effect is small. This work underlines the perception of connection.

We develop a new class of random graph models for the statistical estimation of network formation-subgraph generated models (SUGMs). Various subgraphs-e.g. links, triangles, cliques, stars-are generated and their union results in a network. We show that SUGMs are identified and establish the consistency and asymptotic distribution of parameter estimators in empirically relevant cases. We show that a simple four-parameter SUGM matches basic patterns in empirical networks more closely than four standard models (with many more dimensions): (1) stochastic block models; (2) models with node-level unobserved heterogeneity; (3) latent space models; and (4) exponential random graphs. We illustrate the framework's value via several applications using networks from rural India. We study whether network structure helps enforce risk-sharing and whether cross-caste interactions are more likely to be private. We also develop a new central limit theorem for correlated random variables, which is required to prove our results and is of independent interest.

By varying prompts to a large language model, we can elicit the full range of human behaviors in a variety of different scenarios in classic economic games. By analyzing which prompts elicit which behaviors, we can categorize and compare different strategic situations, which can also help provide insight into what different economic scenarios might induce people to think about. We discuss how this provides a step toward a nonstandard method of inferring (deciphering) the motivations behind the human behaviors. We also show how this deciphering process can be used to categorize differences in the behavioral tendencies of different populations.