The political and social context makes it more necessary than ever to develop the critical capacities of citizens to deal with the mass of documents that are now spreading very rapidly through the media and social networks. Most of the available studies postulate the existence of a general "critical mind" and pay little attention to the essentially disciplinary functioning of school education. The objective of the CCEH project is to compensate for this shortcoming in the teaching of history in schools by questioning the conditions of learning and the transfer of critical historical skills to non-disciplinary situations (media news). It crosses four central theoretical frameworks of the field of history didactics (Cren: Problematization, Cread: Theory of Joint Action in Didactics, Hep Vaud: Relations Memory/History, U. de Montréal and U du Québec en Outaouais: Historical Thinking), in three countries (France, Switzerland, Canada) on corpus of experimental situations in schools. A qualitative methodology of continuous comparison (based on the Grounded Theory of Glaser and Strauss) will produce concepts for (1) better thinking about the conditions for the development of critical disciplinary capacities and their transfer, (2) developing tools for teacher training, and (3) documenting possible curricular developments. Three hypotheses will structure the research. 1. Documentary criticism skills developed in school history do not de facto become general skills because they only deal with knowledge of the past through the study of its traces; 2. Historical critical skills are not limited to applying rules (in particular external and internal criticism of documents), they consist of an evaluation of past situations with reference to models of disciplinary investigations; 3. Teachers themselves must therefore go beyond the pedagogical model of "teaching rules + training to apply them". To do so, they need knowledge on how to teach students to imitate investigation models. The methodology provides for the co-design and implementation, with teachers, of three series of three teaching sequences in the four clusters (corresponding to the four theoretical frameworks of the members of the consortium). These four corpuses will be structured by a common approach resulting from the already rich previous cooperation of the members of the consortium, which will allow a comparison around three shared variables (inquiry, contents, analogy). The comparative process is twofold: between the data produced in the four clusters, and between sequences and series of sequences throughout the duration of the project. On a practical level, each cluster will develop locally a first series of sequences through cooperation between researchers and teachers, the former being responsible for the scientific objectives, the latter for the pedagogical objectives. The recorded data (preparation sessions and classroom implementations) will be transcribed. In the following semester, each cluster will carry out an initial analysis of the data in its theoretical framework, in order to prepare the comparative collective work that will bring together all the members of the consortium during a three-day meeting. This first stage will lead to the identification of the first relevant conceptual categories around the three hypotheses and the three common variables, and to modifications to develop the next series of sequences (the theoretical sampling, according to the Constant Comparative Method of Qualitatve Analysis), organized according to the same scheme the following year. The third and final series will also enable the Nantes cluster to produce a video support for teacher training.