This doctoral thesis is based on the observation that the individualization of society that is taking place today has led to new differentiations within the domains of lifestyle and residential choice. Our main hypothesis was that lifestyle influences residential choice. What kinds of differentiation with regard to lifestyles are we seeing currently? What connections are there between lifestyles, residential preferences and localization? Theoretically, the typical methods for describing residential choice and environment rely on epistemological, methodological and empirical corpuses that are quite varied and, at times, contradictory. On one hand do we find economic and demographical approaches based on the theory of the "rational" actor who attempts to optimize his localization according to "functional and economic" parameters (i.e. proximity to the workplace, the price and size of the residence depending on income and background); on the other we find sociological and anthropological approaches based on theories of social action where choice as a general rule is "socially" determined based on the individual's family ties, tradition, culture, way of living or even social position (Authier & al., 2010). Above and beyond such rational or social perspectives, other factors also influence the choices we make, including the sensitive dimension related to an individual's physical and emotional experience of space. In order to grasp all three logics of action in a single analytical frame we based our analysis on the postulate that residential choice is "plural." Thus do multiple criteria – accessibility, social rooting of families, social status of the residential residence, appreciation of the morphological qualities of the environment, etc. – figure into the decision. We assumed that based on their lifestyles that households blended one or the other of these logics of action, thus resulting in their ranking their criteria for choice in different ways. We proposed a conceptualization of lifestyle that offers a compromise between classical structuralist perspectives (Bourdieu, 1979) and individualist perspectives (Beck, 1983). In our opinion residential choices are influenced by a three-fold dimension of lifestyle: urban practices and daily life (stylization), values and urban representations (evaluation) and resources/constraints (stratification variables). Thus did we couple classic vertical forms (training, income, gender, etc.) with new horizontal forms (attitudes, opinions, values, practices, etc.) that engender social inequalities. Residential choice therefore results in a compromise (resulting from lifestyle) between the different functional, social and sensitive criteria that are or are not present in the structural, socio-historical, environmental context (potential receptiveness). Methodology The cities of Bern and Lausanne in Switzerland – two cities of comparable size but whose socio-historical backgrounds and urban structures vary significantly, particularly with regard to the quality of public transportation service – served as the field for our study. As families often use long-term residential choices to realize plans for the future, we chose to look at them. We used a combination of quantitative and qualitative methods for our empirical work. At the macro level we describe the territory using a series of functional, social and sensitive indicators that allowed us to characterize "potential receptiveness" at different levels, from citywide to neighborhood. At the micro-sociological level we conducted: a quantitative telephone survey of 500 households in each agglomeration, a qualitative survey comprised of 40 semi-directive interviews in eight neighborhoods that distinguished themselves by their geographical location (central, suburban, peri-urban) and architectural diversity. Principal findings In general our findings show that income inequalities have a discriminatory effect on the size and price of the residence as well as on the probability of finding a living space that is entirely adapted. These inequalities however only play a secondary role when it comes to explaining differences in terms of residential preferences. Whether or not the households had a weak or strong income, certain areas of the city were never considered if they did not correspond to the individual's lifestyle. To begin we showed that the immediate environment plays an important role in quality of life for families, especially with regard to the presence or absence of elements favoring children's autonomy. Access to public transportation for instance is a decisive factor in terms of localization for most families. Above and beyond these commonalities we highlighted six axes of differentiation for residential preference that indicate the qualities most sought after by families: do they want to live in an environment that first and foremost is safe, densely-populated, convivial, traditional, calm or elite? To identify how families classify choice criteria and in so doing position themselves on the different axes we did a typological analysis, thus allowing us to categorize households into seven major "residential lifestyles," each of which valued the environment's social, sensitive and functional qualities in varying degrees. Thus we found there was not a quality of life that held true for all groups but rather qualities of life. Two main groups stood out, highlighting as such the contrast between ways of living typical of (modern) industrial society, based on classic models of upward social mobility, work and material success, and those originating from (post-modern) post-industrial society, based on self-enrichment, post-materialism and criticism of so-called "bourgeois" values. The groups identified are products of the historical sedimentation of the different ways of relating to the environment that have appeared over time. By comparing our analysis of residential lifestyles and the structural/cultural differences between the cities of Lausanne and Bern we were able to highlight the impact of context on the spatial distribution of lifestyles at different scales (from nationwide to the neighborhood level). We were also able to prove that geographical distinctions such as center, suburban and peri-urban become less important; each area has its own individual functional (accessibility, services), sensitive (urban morphology, architecture) and social (social-spatial diversity, segregation) qualities. Based on these characteristics, an environment is more or less receptive to different types of residential lifestyles (can I do everything on foot in this neighborhood? Can my children go to by themselves? Is there a strong social/community dynamic?). With such a detailed qualitative and quantitative description of areas it becomes easier to determine who would like to live in a given neighborhood and who, on the contrary, would not ("describe where you live and I'll tell you who you are?"). To guarantee the success of urban development or regional policies, those responsible for planning and development need to consider urban planning not only in terms of "social mixing" but in terms of the "mixing of lifestyles" as well. Through this notion of potential receptiveness, the urbanism of lifestyles is one of the main functional conclusions of this study, for which we propose avenues of change and recommendations (how to plan and think about urban development based on use?).