THE EFFECT OF COLLABORATIVE KNOWLEDGE MODELING AT A DISTANCE ON PERFORMANCE AND ON LEARNING

Abstract. This study examines the effect of co-elaborating a knowledge model in dyads at a distance on performance and on learning. Participants (N = 48) were trained to represent knowledge taken from a text, using an object-typed knowledge modeling editor software tool. Knowledge modeling is similar to concept mapping, except that the former is based on a typology of knowledge objects and a typology of links, and that the structure of the knowledge representation is not necessarily hierarchical. After a 75-minute training session to knowledge modeling, each participant constructed a knowledge model individually. The experimental session consisted of elaborating a knowledge model in dyads. In the first condition, participants constructed and shared the knowledge model at a distance, using a whiteboard and a chat tool (synchronous group). In the second condition, participants elaborated one knowledge model with a turn-taking approach; they used e-mail to share their work-in-progress (asynchronous group). In the third condition, participants worked face-to-face at the same computer (control group). Pre- and post-tests were administered to measure learning in the domain. Results show that the quality of the knowledge models was better for dyads in the face-to-face condition than for the ones in the asynchronous condition, but only for the score related to knowledge objects (and not for the score related to propositions). We did not find a significant between-group difference on learning, but results indicate a tendency that working at a distance in a synchronous mode was more beneficial than working face-to-face and synchronously at a distance. These results should be interpreted with caution considering the short duration o