There is much positive research on computer-supported collaborative learning (CSCL) environments in asynchronous distributed learning groups (DLGs). There is also research that shows that contemporary CSCL environments do not completely fulfil expectations on supporting interactive group learning, shared understanding, social construction of knowledge, and acquisition of competencies. There appear to be two major pitfalls impeding achievement of the desired social interaction in CSCL environments: taking social interaction in groups for granted and the lack of attention paid to the social psychological dimension of social interaction outside of the task context. Current solutions offered to avoid the pitfalls placed responsibility on instructors and teachers to encourage collaborative learning and social interaction. To both free educators from this burden and be more cost effective, we propose an intelligent CSCL environment. The environment is based upon a theoretical framework that suggests embedding certain properties in the environment to act as social contextual facilitators- social affordances- to initiate and sustain learner’s social interactions. Finally, a group awareness widget (GAW)- a software tool providing the learner group awareness about the others in the task and in the non-task context- is introduced as an embodiment of this theoretical framework.

The practical enquiry model describes a process by which experience leads to understanding though a cyclical process of deliberation-action, perception- conception. The four-phases cycle begins with a triggering event, moves through exploration and, as exploration reveals possible insights, integration, and concludes with resolution. Using the practical inquiry model as conceptual grounding, in this study Garrison, Anderson & Archer’s (2001) procedure for analyzing conference transcripts at the message level was compared with a sentence-level method, using the Transcript Analysis Tool (TAT). Three categories of the TAT were developed, aligning it with the critical inquiry model under different assumptions about and interpretations of the model’s four phases. One of the alignments was shown to accord almost perfectly with the critical inquiry model, with both procedures showing that exploration was most common, followed by integration, triggers and resolution. Other alignments showed different proportions, suggesting that further research (preferably at the sentence level) might be useful in establishing the variation of the proportions of the model’s elements in online interactions of different types and purposes, conducted under different conditions of social and moderator presence.