Modeling Individual Decision for Collaboration and Implications for Intelligent Assistance

As collaboration has proven to be beneficial in learning environments, there has been an emerging interest in automating the formation of student groups. However, existing work focuses on optimal formations from the instructor’s perspective based on pedagogical criteria (e.g., maximal group productivity, avoiding orphans or unmatched students). In contrast, we propose a formal collaboration model from the student’s perspective using formation criteria that are important to individual students. As a consequence, resulting formations may contain groups of isolated students. We believe this result opens an opportunity to utilize the model in helping designers develop better adaptive systems for single users. To this end, we derive a formal model that explains why some users choose to collaborate while others choose to work independently. We implement this model and demonstrate in simulation the various factors involved in people’s choices in the context of collaborative story writing. Furthermore, we use this model as the basis of designing an adaptive story writing assistant for individual users and discuss the design implications for intelligent assistance in general.