This study analyses the representations of urban space from the social representations paradigm (Jodelet, 1982). This approach hypothesizes that a same set of elements of representation can be structured differently from one social group to the next; social meanings as such are less important to our understanding of what differentiates a representation from another than the place of its meanings in the structure of representation (Abric, 1976). The psychosociological approaches of representations provide us with the conceptual and methodological tools to analyze spatial representations through the prism of social relationships by investigating their social (individuals’ social histories) and cognitive structure. As the Chicago school has long showed, despite the effects of social structures, individuals can move on to occupy social positions other than those previously held by their parents (original social positions). To highlight the social dimensions of spatial cognition processes, we examine the cognitive configurations of the city from the social mobility of respondents that live in a same neighborhood. We have hypothesized that social position of individuals is not only a social filter for cognitive process: a same social group affiliation can produce different representations of urban space because it depends more on social structure than their actual social positions.

In order to verify our hypothesis, we have questioned 92 residents, living in single-family houses of a neighborhood located in the first ring of the Strasbourg suburbs. This population is mainly composed of managers and officers. The survey participants responded to a face-to-face questionnaire including a section on the representation of space and one on socio-demographic data. In the first section we have used a tool developed to collect social representation: technique of characterization (Flament, 1994), and the sociographic section was collecting the individuals’ actual social position and individuals’ social trajectory.

A hierarchical clustering analysis of the responses to the characterization technique yielded four cognitive groups with different representations of the city of Strasbourg. Then a multiple correspondence analysis allows us to give a clear picture of the relations between the four socio-cognitive groups and the individuals’ sociographic characteristics. The results show that the representation of Strasbourg depends on the social trajectory of individuals (which can be upward, downward or stable compared to social position of the parents). Thus, our results show that the representation of urban space does not follow a principle of self-construction within a social group, that social life within the group (i.e. the cultural status of social factors - Heft, 2013) is not the only factor in the socio-cognitive construction of geographical space. Representation of geographical space ultimately relates to social relationships between groups, to social structures rather than personal experience. We conclude that reference or anchor points, hierarchies between urban places and clusters in cognitive mapping must be analyzed in a social and relational/transactional perspective, to understand more accurately the cognitive mapping processes and its relations on social socio-spatial segregation.