The study of family mediation by the social sciences has rarely been interested in the social and cultural representations of the interveners themselves, who nevertheless arrive in the process of intervention by being charged with their own prejudices about "others". For this very reason, this research aims to study the representations and interventions of family mediators for mixed couples in conflict. Qualitative in nature and carried out in Quebec and France, the study was carried out with 41 respondents, i.e. 25 Social Workers in Family Mediators (SWFM) and 16 resource people with different skills (pioneers of family mediation, specialized researchers, legal experts and institutions) grouped together under the "REFM" category (Researchers and Experts in Family Mediation). The main objective is to identify and analyze the representations which, explicitly or implicitly, influence the intervention of family mediators in an intercultural context. The analysis of the data collected makes it possible to highlight several points. Historically, the issues of immigration, socio-institutional integration and the judicialization of family mediation in France and Quebec. In terms of professional practice, the context of the emergence and the characteristic features of the representations of mediators working with interethnic couples are identified. On the epistemological level, the dominant approach of social psychology of representatives in social work is criticized, which tends to consider them as simple defense mechanisms relating to the interest of the anthropological perspective, the study clarifies the stakes of an analysis of representations as a resource of power and domination.The results obtained during the analysis of the various information and the positions taken by the various respondents show that if opinions can be decided concerning the profession of (intercultural) mediator, its practice and its history in Quebec and in France, a consensus emerges. Made on the representations of interveners in mixed couples conflicts as a bias in the intervention and in the quality of professional efficiency. The absence of marked variation on this last point and on the need to work more to curb, or at least to control this bias, clearly indicates that the subject of the representations of intercultural mediators is worthy of interest: if this study questions and analyzes representations as a problem in social work, so that mediators can curb them in practice, and to clear a thematic avenue to be further explored by future research.