Abstract

Emotions in collaborative learning both originate from and are externalized in students’ socio-emotional interactions, and individual group members evidently contribute to these interactions to varying degrees. Research indicates that socio-emotional interactions within a group are related with the occurrence of co- and socially shared regulation of learning, which poses a need to study individual contributions to these interactions via a person-centered approach. This study implements multimodal data (video and electrodermal activity) and sequence mining methods to explore how secondary school students’ (n = 54, 18 groups) participation in socio-emotional interactions evolved across a series of collaborative tasks. On this basis, it identifies subgroups of students with distinct longitudinal profiles. Furthermore, it investigates how students with different socio-emotional interaction profiles contributed to their groups’ regulation of learning. Three profiles were identified: negative, neutral, and diverse. Each profile represents a particular socio-emotional interaction pattern with unique characteristics regarding the emotional valence of participation and physiological emotional activation. The profiles relate to students’ contributions to group regulation of learning. Students with the diverse profile were more likely to contribute to regulation, whereas the neutral profile students were less likely to contribute. The results highlight the importance of person-centered methods to account for individual differences and participation dynamics in collaborative learning and consequently clarify how they relate to and influence group regulation of learning.