Push and Pull: A Framework for Measuring Attentional Agency on Digital Platforms
Zachary Wojtowicz, Shrey Jain, Nicholas Vincent
TL;DR
The paper defines attentional agency as a user’s ability to allocate attention to information on digital platforms and formalizes a three-party model (agent, advocate, platform) to quantify pull and push through a unified framework. It introduces a principled optimization, where the platform chooses allocations $\phi_\lambda$ to maximize $\mathbb{E}[\lambda U(\phi,\theta) + (1-\lambda) V(\phi,\theta) \mid m]$, and defines Pull = $U_\lambda/U_1$ and Push = $V_\lambda/V_0$, enabling per-user and population-level analyses of attentional agency. The framework accommodates advances in AI, such as generative foundation models, which enable finer-grained token-level allocations and broader information recombination, potentially expanding overall value but redistributing agency between agents and advocates. The authors discuss applications across search, content, social media, matching, and marketplaces, and outline policy strategies—transparency, push-free options, and neutral algorithms—to balance welfare with platform incentives. Overall, the work provides a quantitative, operational lens to study and shape how AI-powered platforms influence what people see, with significant implications for alignment, fairness, and governance in the attention economy.
Abstract
We propose a framework for measuring attentional agency, which we define as a user's ability to allocate attention according to their own desires, goals, and intentions on digital platforms that use statistical learning to prioritize informational content. Such platforms extend people's limited powers of attention by extrapolating their preferences to large collections of previously unconsidered informational objects. However, platforms typically also allow users to influence the attention of other users in various ways. We introduce a formal framework for measuring how much a given platform empowers each user to both pull information into their own attention and push information into the attention of others. We also use these definitions to clarify the implications of generative foundation models and other recent advances in AI for the structure and efficiency of digital platforms. We conclude with a set of possible strategies for better understanding and reshaping attentional agency online.