My research on social networks is of two varieties. First, I use experimental methods to analyze the cognitive and situational factors that lead people to think about, or "cognitively activate," different subsections of their social networks. In this way I study how heuristics of network perception come to affect people's ability to capture, and squander, the resources available to them in their social networks. Second, I use methods from network analysis and graph theory to build new models of information diffusion that affect the functioning and evolution of markets, including notable the formation of prices. Note: Links to papers may be inactive for papers currently under review or being revised.

Why do equity analysts get it wrong? A (social) structural explanation of forecasting accuracy (with Sun Park)
Working Paper (Additional related slides)

Identities in flux: Cognitive network activation in times of change (with Tanya Menon)
Social Science Research, 45: 117-130

Using a dynamic cognitive model, we experimentally test two competing hypotheses that link identity and cognitive network activation during times of change. On one hand, affirming people’s sense of power might give them confidence to think beyond the densest sub- sections of their social networks. Alternatively, if such power affirmations conflict with people’s more stable status characteristics, this could create tension, deterring people from considering their networks’ diversity. We test these competing hypotheses experimentally by priming people at varying levels of status with power (high/low) and asking them to report their social networks. We show that confirming identity—not affirming power—cognitively prepares people to broaden their social networks when the world is changing around them. The emotional signature of having a confirmed identity is feeling comfortable and in control, which mediates network activation. We suggest that stable, confirmed identities are the foundation from which people can exhibit greater network responsiveness.

Status Differences in the Cognitive Activation of Social Networks (with Tanya Menon & Leigh Thompson)
Organization Science, 23(1), 67-82

We develop a dynamic cognitive model of network activation and show that people at different status levels spontaneously activate, or call to mind, different subsections of their networks when faced with job threat. Using a multimethod approach (General Social Survey data and a laboratory experiment), we find that, under conditions of job threat, people with low status exhibit a winnowing response (i.e., activating smaller and tighter subsections of their networks), whereas people with high status exhibit a widening response (i.e., activating larger and less constrained subsections of their networks). We integrate traditional network theories with cognitive psychology, suggesting that cognitively activating social networks is a precondition to mobilizing them. One implication is that narrowing the network in response to threat might reduce low-status group members’ access to new information, harming their chances of finding subsequent employment and exacerbating social inequality.


Tempering the Glass Ceiling? Differences in how Men and Women Perceive Their Social Networks under Conditions of Threat and Opportunity (with Tanya Menon)
Working Paper.

A prominent explanation for the gender gap in labor market outcomes is that men have more resource-rich social networks than women. We present evidence that gender disparities can also arise from differences in how men and women cognitively activate or call to mind, their social networks in the face of certain social situations. When women were experimentally primed to feel a lack of control, versus feel in control, they cognitively activated fewer contacts, a smaller proportion of weak-tie contacts, more super-ordinates, and fewer subordinates. These patterns of network recall suggest that women experiencing a lack of control may cognitively simulate a "psychological glass ceiling," or a surrounding social network structure that reflects constrained power and control. In addition to addressing the gender gap by countering discrimination and fostering equal access to opportunity, an implication of these findings is that it may also be necessary to design interventions that allow women to envision the richness of their social networks.

Networks as Self-Defense: Identity Threat and Compensatory Cognitive Network Activation (with Cindy Wang and Tanya Menon)
Working Paper.

We argue that social networks function as more than "pipes" and "prisms" that transmit tangible and intangible resources interpersonally: they also serve people’s intrapersonal goals of identity maintenance. Three experiments considered how identity primes affected people’s representations of their networks. People who received feedback that disconfirmed their gender (Study 1) and political/ideological (Study 2) identities, subsequently recalled, or "cognitively activated," networks that were smaller, denser, more emotionally supportive, more likely to be composed of people they have known for longer, and more likely to be composed of people associated with their disconfirmed identity. While identity-disconfirming information unsurprisingly triggered negative affect, people who then cognitively activated the types of networks mentioned above reported elevated moods. These results are indicative of an affirmational, compensatory function of (cognitive) social networks that allows people a psychological respite from situations that disaffirm self-perception. A final study (Study 3) investigated how this psychological process could affect more instrumental networking goals such as information search. By utilizing networks to serve identity-related goals, people may distort and block the network pipes that service effective network mobilization.

The affective antecedents of cognitive social network activation (with Catherine Shea, Tanya Menon, and Kyle Emich)
Social Networks. Forthcoming.

How might people’s moment-to-moment feelings affect the social network contacts they call to mind? Three datasets indicate that experiencing positive affect leads people to cognitively activate larger and more sparsely connected social network structures. A preliminary association emerged between positive affect and activating large, diversified network structures in the General Social Survey. To isolate causality, we then conducted two experiments where participants were randomly assigned to experience either positive or negative affect which replicated the hypothesized relationship between affect and cognitive network activation.


Structure vs. Capture of Social Capital: Explaining the gap between having and using networks (with Tanya Menon)
Working paper.

Social network research typically focuses on the structure of social capital—that is, how network characteristics such as size, density, and topology affect both network processes such as the diffusion of information and emergence of trust as well as various individual-level behaviors and outcomes. By contrast, the present research emphasizes an alternate perspective, focusing on the capture of social capital. We focus on activation and mobilization as two important barriers to extracting value from social networks. First, we describe how activation, involving cognitive principles such as the use of decision-making heuristics and interpretive frames, can distort people's network perceptions and produce gaps between real and imagined network structures. Second, we consider various social orientations that can inhibit network mobilization, creating a gap between knowing a network and utilizing it. Capture matters because it represents an important pathway by which social structures come to affect individual- and system-level outcome.

A Model of Robust Positions in Social Networks (with Matthew Bothner & Harrison White)
American Journal of Sociology, 116(3), 943-92.

This article introduces a network model that pictures occupants of robust positions as recipients of diversified support from durably located others and portrays occupants of fragile positions as dependents on tenuously situated others. The model extends Herfindahl’s index of concentration by bringing in the recursiveness of Bonacich’s method. Using Newcomb’s study of a college fraternity, we find empirical support for the contention that fragility reduces future growth in status. Applications of the model to input-output networks among industries in the U.S. economy and to hiring networks among academic departments are also presented. Implications for future research are discussed.