Daniel M. Romero

Aparna Ananthasubramaniam, David Jurgens, Daniel M. Romero

Adoption of cultural innovation (e.g., music, beliefs, language) is often geographically correlated, with adopters largely residing within the boundaries of relatively few well-studied, socially significant areas. These cultural regions are often hypothesized to be the result of either (i) identity performance driving the adoption of cultural innovation, or (ii) homophily in the networks underlying diffusion. In this study, we show that demographic identity and network topology are both required to model the diffusion of innovation, as they play complementary roles in producing its spatial properties. We develop an agent-based model of cultural adoption, and validate geographic patterns of transmission in our model against a novel dataset of innovative words that we identify from a 10% sample of Twitter. Using our model, we are able to directly compare a combined network + identity model of diffusion to simulated network-only and identity-only counterfactuals -- allowing us to test the separate and combined roles of network and identity. While social scientists often treat either network or identity as the core social structure in modeling culture change, we show that key geographic properties of diffusion actually depend on both factors as each one influences different mechanisms of diffusion. Specifically, the network principally drives spread among urban counties via weak-tie diffusion, while identity plays a disproportionate role in transmission among rural counties via strong-tie diffusion. Diffusion between urban and rural areas, a key component in innovation diffusing nationally, requires both network and identity. Our work suggests that models must integrate both factors in order to understand and reproduce the adoption of innovation.

Edward L. Platt, Daniel M. Romero

Targeted attacks against network infrastructure are notoriously difficult to guard against. In the case of communication networks, such attacks can leave users vulnerable to censorship and surveillance, even when cryptography is used. Much of the existing work on network fault-tolerance focuses on random faults and does not apply to adversarial faults (attacks). Centralized networks have single points of failure by definition, leading to a growing popularity in decentralized architectures and protocols for greater fault-tolerance. However, centralized network structure can arise even when protocols are decentralized. Despite their decentralized protocols, the Internet and World-Wide Web have been shown both theoretically and historically to be highly susceptible to attack, in part due to emergent structural centralization. When single points of failure exist, they are potentially vulnerable to non-technological (i.e., coercive) attacks, suggesting the importance of a structural approach to attack-tolerance. We show how the assumption of partial trust transitivity, while more realistic than the assumption underlying webs of trust, can be used to quantify the effective redundancy of a network as a function of trust transitivity. We also prove that the effective redundancy of the wrap-around butterfly topology increases exponentially with trust transitivity and describe a novel concurrent multipath routing algorithm for constructing paths to utilize that redundancy. When portions of network structure can be dictated our results can be used to create scalable, attack-tolerant infrastructures. More generally, our results provide a theoretical formalism for evaluating the effects of network structure on adversarial fault-tolerance.

Daniel M. Romero, Roderick I. Swaab, Brian Uzzi et al.

The current research used the contexts of U.S. presidential debates and negotiations to examine whether matching the linguistic style of an opponent in a two-party exchange affects the reactions of third-party observers. Building off communication accommodation theory (CAT), interaction alignment theory (IAT), and processing fluency, we propose that language style matching (LSM) will improve subsequent third-party evaluations because matching an opponent's linguistic style reflects greater perspective taking and will make one's arguments easier to process. In contrast, research on status inferences predicts that LSM will negatively impact third-party evaluations because LSM implies followership. We conduct two studies to test these competing hypotheses. Study 1 analyzed transcripts of U.S. presidential debates between 1976 and 2012 and found that candidates who matched their opponent's linguistic style increased their standing in the polls. Study 2 demonstrated a causal relationship between LSM and third-party observer evaluations using negotiation transcripts.