Campbell, in 1958, introduced the term “entitativity” as “the degree to which a social aggregate is perceived as “having the nature of an entity, of having real existence” (p.17). In this sense, a social group can vary along a continuum that ranges from low to high entitativity. Hamilton and Sherman (1996) reintroduced the concept and found empirical data that confirmed the hypothesis that perceptions of entitativity, i.e. seeing social targets as possessing unity and coherence, have important implications on how one organizes information about groups and forms impressions. Thus, perceivers are more likely to engage an integrative processing in forming impressions of groups high in entitativity, than of groups low in entitativity (e.g.: MacConnell et al.,1997; Susskind et al., 1999). In this sense the perceivers would make more extreme trait ratings and faster responses to groups with high entitativity than to groups with low entitativity. The main aim of this study is to test these hypotheses in relation to a place, i.e., neighbourhoods of a big city. Recent research explores the concept of “entitativity”, as used in social psychology, to understand the way people organize information about places and people that live in those places (Bernardo, 2008). A group of university students rated a sample of 20 neighbourhoods in the city of Lisbon, Portugal, on 23 social and physical properties of groups and perceived entitativity. The results show that the neighbourhoods vary in terms of entitativity and identify a group of physical properties that are strongly correlated with group entitativity. The main prediction was that participants in high entitativity conditions, both social category and neighbourhood, would make more extreme trait ratings, respond faster and recall more of the stimulus behaviours, than participants in low entitativity condition. An experimental study was conducted with a 162 subjects, randomly divided in four target conditions. Thus the experiment consists of a 2 X (high entitativity vs. low entitativity) X 2 X (neighbourhood vs. social categories). The social categories considered were gipsies and economists, and the neighbourhoods “Bairro Alto” and “Parque das Nações”. The study used the E-Prime software, in order to control the response latencies. The procedure includes the presentation of 16 statements that describe behaviours performed by members of a group. The behaviours were selected to give information about four themes: athleticism, sociability, political activism, and intelligence (based on Susskind et al., 1999). After this task the participants completed a filler-task during 3 minutes, and then they completed three dependent measure tasks: a trait judgment task, a recall task and a perceived entitativity measure. The results shows that when forming an impression of a group with high entitativity, perceivers made more extreme judgments and choose the response faster than in relation to low entitativity groups, both in social categories and social aggregate based on the belonging to a specific place. These results confirm the hypothesis of a more integrative information processing and more dispositional inferences in relation to members of entitative groups. The results are discussed in relation to the intergroup relations in an urban context.