Teacher collaboration is a key question amongst educational research. The individual aspects of teachers' activities are well known. Nevertheless, teacher work is not limited to the mere action of an individual practitioner with a group of pupils. Thus the collective aspects of teacher work are studied, often in order to describe the implementation process of new teaching strategies or to determine the school effectiveness (Grangeat & Gray, 2008). Nevertheless, a few studies aim to understand the interactive processes which underpin teacher collective work and which contribute to change practitioners' conceptualisations and practices. But these questions are well known since the 90s, with respect to vocational fields, through the ergonomics or activity theory frameworks. These studies produced some findings which could be helpful for understanding the factors which influence teacher collective work (Schmidt, 1991). A first result leads to identify three collective work features. The first associates an increasing form (several teachers who teach the same subject) with a collaborative function (these persons act with similar methods): within this feature, the difficulty consists to ensure the coherency, the similarity, amongst the actions of the different actors (Tiberghien, 1994). The second associates a completive form (teachers from different subjects act together) with a co-operative function (teachers deal different tasks to accomplish a same goal). Here, the difficulty is to maintain both the specificity of each actor and their mutual support (Hansen, 2008). The third associates a co-active form (actors works within the same space and time) with a debative function (they confront their ideas). The challenge is the respect of each point of view and the helpfulness of the common resources which result from the debate (Tillema & Orland-Barak, 2006). A second result highlights the collective work effects. Research shows how informal interactions transform the professionals' way to think about their work. They show also that professionals extend the range of elements which are caught in the situation in order to steer the activities at work. This extent occurs when actors could interact either through specific training sessions -aiming to confront methods about a specific professional problem- either through projects accomplished with partners (Engeström & Young, 2001). From these results, some research questions occur. They concern the modalities and the effects of: - teacher collaboration within a same subject fields (e.g. science) in order to improve teaching methods. - co-operation amongst teachers and partners within a specific project (e.g. in order to improve learning outcomes for disadvantaged pupils). - debates amongst teachers about specific teaching problems (e.g. new science teachers exchange about the gain and loss of inquiry based methods). The effects concern either teachers' conceptualisations and practices either pupils' learning outcomes. The paper aims to understand effects of expert teachers' collaboration on their conceptualisations and practices towards inquiry based science teaching (IBST). Teacher collaboration consists of debates amongst expert teacher aiming to conceive and conduct training sessions about IBST toward other science teachers.

According to the activity theory framework, four elements could organise conceptualisations about work and practices: teacher's goals, clues picked out of the teaching situation and which trigger an action, repertoire of actions available for the teacher with respect to his or her goals, reference knowledge which contribute to justify and orientate his or her action (Engeström, 2000). These elements which contribute to accomplish a same goal compose a teaching process-knowledge (TPK); the organisation of these TPK according to the teacher's different goals compose the teacher operative model (TOM) (Grangeat, 2009). For this first study, three science teachers are observed during a sequence about IBST; the sequence is videotaped. Then each teacher comments his or her video. These comments are audio taped. The records are transcribed. Eventually a content analysis of these transcription allows to elaborate TPK and TOM.

It is expected than the TOM are organized according to three critical dimensions: the action devolution toward the pupils, including self-regulated learning; the sequence steering, including the pupils' diversity management; the explicit making of the taught content, including assessment preparation with pupils. Teachers could make explicit that most of their TPK results from collaboration amongst them in order to conceive common training session.