Collaboration with AI: How group composition affects solution quality in problem solving

While collaborative learning remains a cornerstone of higher education, the factors underlying effective group performance are not well understood, particularly in the context of computer-supported collaborative learning with artificial intelligence (AI) as a co-author. This study explored the relationship between group characteristics (prior knowledge, heterogeneity and group size) and the quality of solutions generated through collaboration with AI. The present study looked at a group performance of 196 bachelor students engaged in group problem-solving activities during a macroeconomics course at a large university. Unsurprisingly, multiple regression analysis revealed that groups with higher prior knowledge generated solutions of higher quality when collaborating with AI. In contrast, greater heterogeneity in prior knowledge within groups was associated with lower quality solutions. Interestingly, larger groups demonstrated superior performance in generating solutions with the help of AI compared to smaller groups. These results demonstrate the importance of considering group composition when setting collaborative activities supported with AI.