It is contended that perceptions of groups are affected by particular variables that do not apply to individuals (e.g., intragroup similarity and proximity). Importantly, the perception of outgroup threat has incomplete analogs at the individual level. Results from 3 studies support predictable distinctions between representations of individuals and of groups. Study I showed that priming of the word they produces more extreme negative judgments of the protagonist(s) in a story about 4 individuals acting jointly than in the same story with a single person acting alone. The opposite result holds for priming with the word he. Study 2, with Korean participants, demonstrates that actions by individuals or groups elicit differing preferences for redress. Individual responses (e.g., getting mad) to an individual racial insult (e.g., a snub by a waitress) are preferred to collective responses (e.g., circulating a petition), whereas the reverse preferences hold for a group insult (e.g., taunts from a gang of White youths). In Study 3, cues to the sensitivity of a group are introduced. This concept, introduced by Donald Campbell (1958), distinguishes different degrees of "groupness." Visual depictions of collections of unfamiliar humanoid creatures (greebles) were used to convey that they were either similar or dissimilar and either proximate or scattered. Results confirm the expectation that similarity and proximity-two entitive conditions-elicit more negative judgments of the group. Attention to other cues for entitivity may enrich social psychological views of stereotyping and prejudice by focusing on perceptions of groups as coordinated actors with the potential to bring about negative consequences. Such experiments point to the need for greater research focus on the vastly understudied but fundamental problem of the social cognition of group behavior.