A new study suggests that people may be able to make surprisingly fast predictions about strangers simply by identifying the social roles they occupy. Rather than treating every person as an entirely unknown individual, the human mind appears to use role-based information as a compact guide to three questions that shape everyday interaction: how someone is likely to behave, what that person is likely to know, and how difficult it would be to replace them. The findings, reported by A. Baker, Y. Dunham and J. Jara-Ettinger in Communications Psychology, offer a technical explanation for a familiar social shortcut. A uniformed paramedic, a classroom teacher, a restaurant manager or a software engineer may be judged differently before saying more than a few words, not necessarily because of personal reputation, but because the role itself carries expectations about behavior, expertise and importance.
The researchers’ central argument is that social roles function as efficient predictive tools. In computational terms, a role can be understood as a high-level latent variable: an unseen category inferred from visible clues and then used to estimate several hidden properties at once. When someone identifies another person as a doctor, for example, they may immediately raise their estimate of that person’s medical knowledge, expect certain professional behaviors and assume that the person performs tasks that cannot be transferred instantly to anyone else. This kind of inference reduces the mental effort required to evaluate every new social encounter from scratch. Instead of calculating an individual’s abilities, intentions and responsibilities independently, the brain can use the role as a compact model that generates likely answers.
That shortcut may be especially valuable because social life is filled with incomplete information. People rarely know a stranger’s full history, skills, motives or limits, yet they must still decide whether to ask for help, share information, follow instructions or depend on that person during a crisis. Roles provide an organized structure for making those decisions rapidly. The study’s title highlights three linked predictions: what others will do, what they will know and whether someone else could take over their work. These judgments are not identical. A person may be expected to behave reliably without possessing specialized knowledge, or may hold rare expertise while occupying a role that is easy to fill. By connecting all three dimensions, the research frames role perception as a broader system for reasoning about social organization.
The behavioral prediction is perhaps the most visible part of the process. Roles often come with norms, duties and expectations that guide conduct. A judge is expected to follow formal procedures, a lifeguard is expected to monitor danger and a delivery worker is expected to transport goods to a destination. These expectations do not guarantee that every individual will behave accordingly, but they create a prior probability—a starting estimate—before personal evidence arrives. In Bayesian terms, the role supplies a prior belief about likely behavior, while later observations can update that belief. A person who acts contrary to the expectations associated with a role may therefore attract attention precisely because the mind had already generated a prediction.
The same mechanism can be used to estimate knowledge. Social roles divide expertise across a community, allowing people to infer who is likely to know what without testing everyone directly. A mechanic is presumed to understand vehicle systems; a pharmacist is presumed to know about medicines; an information-technology specialist is presumed to understand certain digital problems. These assumptions are probabilistic rather than absolute, and they can be wrong, but they make cooperation faster. The role becomes a signal about access to information, training and experience. In this sense, role recognition resembles an information-compression strategy: many details about education, practice and responsibility are reduced to a single social label that can guide immediate decisions.
The third prediction—replaceability—adds an important layer to the picture. People do not only ask whether someone can perform a task; they also consider how easily another person could step into the same position. Replaceability depends on factors such as specialized knowledge, institutional authority, unique relationships and the time required to train a substitute. A cashier may be replaceable in one setting but not during a critical shift with no backup. A researcher may be difficult to replace if they alone understand a complex project’s history. By including replaceability, the study treats roles not merely as descriptions of individual activities but as components of larger systems. The value of a role can be inferred from how much disruption would follow if its occupant disappeared.
This framework may help explain why role-based judgments can influence cooperation, trust and social status long before people develop personal relationships. If a role is associated with competence, authority or scarce knowledge, its occupant may receive attention or deference. If a role is seen as easily replaceable, the person may be treated as less indispensable, even when their individual contribution is substantial. Such judgments could affect workplace decisions, division of labor and responses to emergencies. They may also produce systematic errors. A role can create expectations that overlook individual differences, exaggerate expertise or assign importance unevenly across occupations. The same mental shortcut that accelerates coordination can therefore reinforce stereotypes and obscure the people whose work is less visible.
The research also speaks to a broader question in cognitive science: how humans manage the enormous complexity of other minds. Predicting another person requires estimating intentions, capabilities, information and future actions, all under uncertainty. Role-based reasoning offers a solution by linking these variables through socially learned structures. It may allow people to make useful predictions with limited data, while remaining flexible enough to revise those predictions when new evidence appears. The study does not imply that roles determine behavior or that strangers can be understood accurately at a glance. Instead, it identifies roles as a powerful source of prior information—one that helps the mind navigate a world in which cooperation depends on knowing who is likely to do what, who can answer which question and whose absence would matter most.
The findings are likely to resonate beyond psychology because modern societies increasingly depend on distributed expertise and specialized work. Hospitals, laboratories, schools, businesses and online communities all require people to infer responsibilities quickly and coordinate with individuals they may barely know. Understanding the mental machinery behind those inferences could improve the design of teams, training systems and artificial-intelligence models that interact with humans. It could also encourage more careful thinking about the difference between a useful prediction and a fixed assumption. A role can be an efficient map, but it is not the territory: individuals may exceed, reject or redefine the expectations attached to their positions. By showing how rapidly people use roles to anticipate behavior, knowledge and replaceability, Baker, Dunham and Jara-Ettinger place an ordinary social habit at the center of a larger scientific story about prediction, cooperation and the hidden architecture of human communities.
Subject of Research: Social roles and how people use them to predict others’ behavior, knowledge, and replaceability.
Article Title: People use roles to quickly predict how others will act, what they know, and when they can be replaced.
Article References: Baker, A., Dunham, Y. & Jara-Ettinger, J. “People use roles to quickly predict how others will act, what they know, and when they can be replaced.” Communications Psychology (2026). https://doi.org/10.1038/s44271-026-00513-9
Image Credits: AI Generated
DOI: 10.1038/s44271-026-00513-9
Keywords: social roles, social cognition, prediction, human behavior, knowledge inference, replaceability, cooperation, cognitive science, psychology, social reasoning
