Comparing the effectiveness of supports for collaborative dialogic sense-making with agent-based models

Collaborative dialogue has been classified by the ICAP framework (Chi & Wylie, 2014) as the highest interactive level of cognitive engagement. In this study we evaluate the balance between type of guidance and learner characteristics while collaborative learning with agent-based models takes place. Participants were students from health-care programs and were randomly assigned to one of two conditions: One group learned with agent-based models using the vicarious approach, where pairs viewed and discussed recordings of others using agent-based models; the other group explored agent-based models while discussing in pairs a set of text-based prompts. The results reveal that the vicarious-learning approach was superior to agent-based models’ exploration. More detailed evaluation shows that nursing students’ performance wasn’t affected by the type of guidance, while less experienced in the content, non-nursing students, benefited more from vicarious guidance. Findings suggest that dialoguing while observing the dialogue of others can maximize immediate learning gains.