Inria / Raweb 2004
Project-Team: Eiffel

Search in Activity Report, year 2004:


Project-Team : eiffel

Section: New Results

Axis 2: Reflective aspects in design: knowledge management and capitalization

This research topic involves the identification and development of new methods and tools for the capitalization of design knowledge. This capitalization is implemented in organizational training and learning and case-based reasoning, and is recommended in software engineering reuse activities. The topic of knowledge capitalization is often considered from a purely technical angle (technical databases) and from a static perspective (preserving existing knowledge). However, the techniques based on such approaches have many limitations: loss of knowledge, high implementation costs, reluctance by users to implement the capitalization since its benefits are not instantaneous. The planned solution is to move from a static viewpoint (accumulate knowledge and information) to an approach that is both constructive (assist the individual or group process of knowledge accumulation) and integrated (integrate knowledge accumulation into the main activity itself, in our case design). This is the approach that is pursued in our research activities. In this context, it is also important to analyze the role of meta-cognition in this developmental process and the way to support it.

The role of local adaptations in the evolution of rules: therapeutic decision making in cancerology

Participants: Pierre Falzon, Vanina Mollo.

In medical practice, practitioners use pre-established protocols, which consist in a set of rules resulting from analyzing the scientific literature [51]. When they are faced to unusual situations (that make the application of rules impossible), practitioners can refer to a pluridisciplinary concertation committee (CCP). The objective of this study is to understand how practitioners adapt these rules to respond to unusual situations, and the role of existent resources (protocol and CCP) in knowledge construction and evolution [14][48]. In an initial analysis, 19 practitioners were asked to think aloud while resolving 15 cases involving one or more factors that make impossible strict application of the protocol. Individual allo-confrontation has also been conducted: practitioners were confronted with colleagues' decisions, and had to comment them [35]. This method leads practitioners to make their knowledge more explicit. Moreover, it constitutes a concrete helpful tool for reflexive activity, allowing knowledge construction and development.

Meta-cognition in collaborative learning

Participants: Françoise Détienne, Laurence Gagnière.

This research aims to study, in computer-supported collaborative-learning situations, the impact of reflection tools that are assumed to help students to develop and improve meta-cognitive skills. The process by which students develop problem solving, decision making and investigation activities in these situations, is determined by the relationship between collaboration, computer tools and meta-cognitive skills. This relationship stems from research in two areas, Collaborative Learning Theory and Meta-cognitive Theory. It is studied in learning situations using a project-based-learning model [45]. This research is performed in collaboration with the Université de Savoie and the Université de Genève.

The originality of our project lies in the integration of two theoretical perspectives in a complementary framework: the impact of meta-cognitive regulation on learning, and the potential role of traces of the learning activity in supporting the regulatory processes. We propose a method of confrontation aiming at supporting conscious awareness of regulation. It consists in confronting learners with the traces of another learner's activity (individually or with a peer-tutoring approach) in order to both make explicit their procedures used to realize the learning task and to improve these procedures. Another important point of this project is to consider the potential of computer environments for encouraging reflection by way of devices that could keep a record of learners' experience.

Mutual learning between users and designers

Participant: Pascal Béguin.

A current field of research considers the continuation of design in usage. Based on activity theory, human instruments are seen as containing components from both artefacts and users' utilization schemes. Users, through their use of artefacts, turn them into instruments. Extending this approach, design can be viewed as a mutual learning process between users and designers [21][17][22][36][18][24][39].

This research question refers to the more general question of how to articulate users' and designers' creativity. Two main approaches have initially been explored. The first one, which consists in ensuring that the creativity of the user's activity will become a source for the designer's activity, leads to the idea that a system crystallizes representations (of functioning, of activities, of work). The second approach, which consists in ensuring that the result of the designer's activity will become a source for the user's activity, calls for a theorization of a system's plasticity and boundaries. A current research plan is centered on articulating the two previous approaches. The main idea is that design is a developmental, dialogical and distributed process of mutual learning between users and designers.