Semi-structured interface in collaborative problem-solving
We then computed an indicator of a pair's solution path. When events are put correctly at first try into the schedule, monotony is high. On the contrary, when tuning actions like moving the event up and down or deleting it are necessary, the monotony is low. Monotony is related to the proportion of strategic utterances produced during the third of interaction time. We propose the following interpretation: when splitting up the work and distributing the roles at the beginning, people accomplish the positioning as an individual would do it, by facing less coordination problem. This seems to be an optimal behaviour in our experiment as the simultaneous manipulation of events belonging to the same room was difficult for technical reasons (See "Monotonic reasoning" section).
As an AI&Ed application of the 'knowledge flow' approach presented in the introduction, we could detect phases in problem solving by observing a switch in the interface usage. Advice could then be given on the best way to use artifacts for a particular sub-task. Next step in this direction are Moobservers, non-intentional agents which consist of a recognising module which information is more or less displayed in a raw graphical manner to the collaborating subjects. A major question is whether the socio-cognitive behaviour of the peer changes when it gets information about its interaction.
Generated with Harlequin WebMaker