476 C. Brassac and M.-C. Mietkiewicz of the way it feels itself and others to exist(Boutonnier, 1953, p. 25). In the course of interaction, the clinician has access to the drawing child's activity; on one hand, he is ndeed able to follow at leisure the drawing process, and, on the other hand, to have at his disposal the outcome of this process, the drawing in itself that can be saved as a trace of the activity. This drawing being finalised, it represents the condensation, the ongoing process of creation, of the dynamic of tracing-erasing that led to it As a researcher, the psychologist works at analysing the semiotic value of the draw ing. This value is the result of a task of interpretation of the graphic representation being studied. Doubtless to say, this work depends on the production frame, on the conditions under which the drawing-object is being created and, of course, on the drawing-object itself. This drawing-object represents the whole of the graphite marks left on paper support by the child. In other words, and with few exceptions, the interpretation object is the graphic itself, the permanent configuration of marks left on paper by the subject We could add that, whether under therapeutic care or subject of theoretical research the tradition consists in the study of a childs individual production Limiting ourselves to the research side of the drawing of the family, we herewith would like to offer a new approach to children's graphic realisations. The new approach relies on a particular construction mode(vs. collection) of empirical data on which we shall base our analysis. This proposal results from the differentiated research fields of both authors. One of us works on a regular base with situations of joint conception, by small groups of human subjects, of manufactured, digital, textual or graphic objects(Brassac Gregori, 2001; Le Ber Brassac, 2000, for example). The other researcher deals with the fact that siblings(including several pairs of twins), will not come out with the same graph- ical representation when asked to draw their family. Through his or her drawing, each child describes its own perception, its point of view about the ties that make up the family, the elective closeness and the omissions that reveal the child's specific position in the family network(Mietkiewicz Schneider, 2005). This dual origin leads us to the following proposal, expressed in a few words: why not ask siblings to draw their family together and why not film them during their activity? One of the goals of this article is to show the added value the study of a joint graphical produc tion may obtain by taking into account its development. To analyse the process of interac tion in the elaboration of a joint drawing of the family, could this be, for researchers in psychology, something new in the understanding of mental determinants of this mental activity? These are the types of questions for which we would like to bring some answers throughout this text. We shall try to do so by first describing the methodology which we adopted, and the empirical material which we built to this effect. Second, we shall propose an analysis of the interaction of both children, producing jointly this drawing, to conclude finally with the results and discuss the accuracy of such an approach Method and empirical material A clinical approach to a collaborative activity Traditionally, studies conducted in the interpretation of children's drawings are based on vast collections of drawings made by large groups of subjects. Therefore, it becomes possi ble to take measures, both qualitative and quantitative, which lead to strong statistical stud- ies whose significance or non-significance will allow to confirm, or invalidate, hypotheses put forward a priori. Let us be very clear: our approach is not in line with this methodological paradigm. 2 This is easily understandable, considering the fact that the built-up empirical476 C. Brassac and M.-C. Mietkiewicz of the way it feels itself and others to exist’ (Boutonnier, 1953, p. 25). In the course of interaction, the clinician has access to the drawing child’s activity; on one hand, he is indeed able to follow at leisure the drawing process, and, on the other hand, to have at his disposal the outcome of this process, the drawing in itself that can be saved as a trace of the activity. This drawing being finalised, it represents the condensation, the ongoing process of creation, of the dynamic of tracing–erasing that led to it. As a researcher, the psychologist works at analysing the semiotic value of the draw￾ing. This value is the result of a task of interpretation of the graphic representation being studied. Doubtless to say, this work depends on the production frame, on the conditions under which the drawing-object is being created and, of course, on the drawing-object itself. This drawing-object represents the whole of the graphite marks left on paper support by the child. In other words, and with few exceptions, the interpretation object is the graphic itself, the permanent configuration of marks left on paper by the subject. We could add that, whether under therapeutic care or subject of theoretical research, the tradition consists in the study of a child’s individual production. Limiting ourselves to the research side of the drawing of the family, we herewith would like to offer a new approach to children’s graphic realisations. The new approach relies on a particular construction mode (vs. collection) of empirical data on which we shall base our analysis. This proposal results from the differentiated research fields of both authors. One of us works on a regular base with situations of joint conception, by small groups of human subjects, of manufactured, digital, textual or graphic objects (Brassac & Grégori, 2001; Le Ber & Brassac, 2000, for example). The other researcher deals with the fact that siblings (including several pairs of twins), will not come out with the same graph￾ical representation when asked to draw their family. Through his or her drawing, each child describes its own perception, its point of view about the ties that make up the family, the elective closeness and the omissions that reveal the child’s specific position in the family network (Mietkiewicz & Schneider, 2005). This dual origin leads us to the following proposal, expressed in a few words: why not ask siblings to draw their family together and why not film them during their activity? One of the goals of this article is to show the added value the study of a joint graphical produc￾tion may obtain by taking into account its development. To analyse the process of interac￾tion in the elaboration of a joint drawing of the family, could this be, for researchers in psychology, something new in the understanding of mental determinants of this mental activity? These are the types of questions for which we would like to bring some answers throughout this text. We shall try to do so by first describing the methodology which we adopted, and the empirical material which we built to this effect. Second, we shall propose an analysis of the interaction of both children, producing jointly this drawing, to conclude finally with the results and discuss the accuracy of such an approach. Method and empirical material A clinical approach to a collaborative activity Traditionally, studies conducted in the interpretation of children’s drawings are based on vast collections of drawings made by large groups of subjects. Therefore, it becomes possi￾ble to take measures, both qualitative and quantitative, which lead to strong statistical stud￾ies whose significance or non-significance will allow to confirm, or invalidate, hypotheses put forward a priori. Let us be very clear: our approach is not in line with this methodological paradigm.2 This is easily understandable, considering the fact that the built-up empirical