Description: This presentation is part of a wider project which compares two objects taught through teaching practices in the classroom - the argumentative text and the relative clause - through a multi focal approach (Schneuwly,Cordeiro & Dolz, in press). One of these focuses concerns fundamental didactic acts (Schneuwly, 2000) like use of memory, creating didactic settings, regulation and institutionalisation. In this paper, in order to describe the object of knowledge constructed in classroom, we focus on regulations provoked by students contributions, stating that they influence in a significant way how teachers reorganize contents and regulate the space of learning (Marton & Tsui-Amy,2004) while teaching. The gap between intended and enacted object revealed through classroom interactions gives us an insight what object is actually taught. Two types of contributions allow us to make evident the transformation of the taught object: (1) "critical" contributions: moments when (a) either the environment created by the teacher or a concept is questioned by a student (b) a student spontaneously makes propositions or integrates teacher's talk or action in a different context (c) a student asks for knew knowledge, anticipating a more complex level of the taught concept or the activity (2) obstacles: comprehension problems related to dimensions of the learning object. An obstacle may provoke teachers adjustment either on the object or the didactic setting through different forms: revising a basic concept necessary to understand the knew concept, giving examples, changing the point of view, guiding through a method, (re) naming the object, reformulating a task instruction, giving new explanations and/or making text or exercise function explicit. Methodology: We have conceived a three step analysis of our data as follows: 1) Selecting, classifying and examining context of significant students' contributions based on a first reading of the transcripts. 2) Looking for indices and operationalizing the concept of critical contribution and obstacle. 3) Selecting extracts for a microgenetic analysis in order to describe specific regulation acts. Conclusions: The main results for the argumentative text show a great amount of contributions in two domains: argumentative operations (arguing, anticipating arguments, taking position) and in text planning. While arguing seems difficult for students, taking position and the use of concession stimulates their critical thinking. The use of text models, used by teachers to help students in their own writing, creates difficulties, work on introduction and organisation of arguments inside a paragraph provokes critical contributions. Contrary to teaching the argumentative text, where critical contributions and obstacles occur with the same frequency, teaching the relative clause generates mostly obstacles, which can be detected throughout a lot of sequences, especially in activities where students have to analyse function and forms of relative pronouns or transform two main clauses in one complex clause. Results concerning the microgenetic analysis also show differences in the way teachers regulate obstacles: in the argumentative text they tend to establish a relationship between different concepts, to give new examples or accept changes concerning initially given instructions. On the contrary, in front of students' incomprehension related to the subordinate clause, teachers explain a concept in the same way, repeat the same procedure and use the same materials. These findings inform us, through students' contributions and the way teachers adjust content and didactic setting, what object is actually constructed in class, i.e. on what dimensions teachers are focussing. But they also show how the object is taught, through teachers' way to react to students' questions. Finally, the comparison of the two objects allows to reveal more flexibility in the way argumentative text is taught compared to a grammar object.