Modeling social influence with hidden attention reveals stronger effects
Learning Social Influence with Unobserved Attention
Social and Information Networks
Summary
People don't pay attention to everything they see on social media, but past studies often assume they do. This paper introduces a new model that guesses who people actually notice and when they choose to ignore opinions. The authors show that including these hidden attention steps improves predictions about how social opinions spread. Their model even works well on many different real-world data sets.
What this means in practice
- •For social media platform engineers: Improve algorithms that estimate user influence by modeling which posts users actually pay attention to.
- •For marketing analytics teams: Enhance prediction of how opinions or trends spread within social networks by accounting for selective attention among users.
Authors
William Thompson, Moyi Tian, Linh Huynh, Alexander Wiedemann, Vicki Modisette, Philip S. Chodrow, Heather Z. Brooks
Abstract
Social influence, the tendency of our social environment to sway our beliefs and behaviors, is often cited as a driver of population-level trends in opinions or beliefs. Many quantitative theories of social influence assume that an agent interacts with its entire social environment. This assumption is often unrealistic. For example, users of online platforms typically see posts from only a small subset of the accounts they follow at any given time. We introduce a simple probabilistic agent-based model of social influence that explicitly models unobserved attention mechanisms. The first of these latent mechanisms is selection: each agent interacts with only one other agent at a time from its social environment. The second is retention: the agent chooses whether or not to ignore the opinion of the other. Our model admits an efficient maximum-likelihood algorithm which allows us to fit it to opinion traces even when these mechanisms are unobserved. On synthetic data, omitting either latent mechanism reduces the overall strength of social influence governed by the inferred interaction kernel, which hides the evidence for the underlying nonlinearity of the social influence. We study a selection of previously analyzed empirical datasets and curate several new ones. We show that models that include the latent mechanisms are often more predictive on these datasets; indeed, they typically improve prediction far more than the choice of interaction kernel does. Modeling latent attention increases held-out predictive likelihood on 14 of these 15 datasets. Each mechanism changes the size of the inferred kernel in a predictable direction, while its effect on the kernel's shape depends on the data.