Interdisciplinary and convergence science are widely recognized as essential for addressing complex societal challenges, yet the field still lacks robust process theories explaining how diverse expertise is transformed into integrative scientific insight. We argue that progress in the Science of Team Science (SciTS) requires moving beyond structural accounts of collaboration (e.g., team composition, disciplinary diversity) toward a cognitive-process understanding of how teams actually build shared knowledge. To address this gap, we advance a SciTS-integrated reframing of the Macrocognition in Teams (MITM) model as a theory of convergence in action.

MITM conceptualizes team cognition as a dynamic process in which distributed, internalized knowledge held by individuals is transformed into externalized, shared team knowledge through iterative knowledge-building processes. Originally developed to explain performance in complex, novel, and high-stakes environments, the model is uniquely suited to convergence science, where teams must integrate heterogeneous expertise under conditions of uncertainty, shifting goals, and ill-defined problem spaces. From this perspective, convergence is not simply the co-location of disciplines, but a macrocognitive achievement involving the construction of shared problem representations that enable coordinated reasoning and solution development across epistemic boundaries.

We articulate how MITM provides a process-level account of three core SciTS challenges. First, for interdisciplinarity, the model explains how teams develop shared understanding through cycles of knowledge elicitation, externalization, and integration, highlighting the critical role of artifacts and representations in making disciplinary perspectives visible and negotiable. Second, for convergence, MITM describes how teams iteratively construct and refine problem models, moving from fragmented disciplinary inputs to coherent, collective representations that support joint action. Third, for team learning, the model positions learning as an emergent, collective phenomenon arising from repeated cycles of interpretation, evaluation, and adaptation, rather than as the aggregation of individual learning alone.

Building on this, we present a SciTS-adapted phase model of macrocognitive problem solving: (1) distributed disciplinary framing, (2) collaborative problem model construction, (3) integrative negotiation and consensus building, and (4) evaluation, adaptation, and revision. Importantly, these phases are not linear but reciprocal. This reflects the iterative nature of knowledge construction in complex scientific work. We further extend this framework by incorporating theorizing on external cognition and sociotechnical systems, describing how cognition in convergence teams is distributed not only across individuals but also across artifacts, tools, and technologies that scaffold both taskwork and teamwork.

This reframing contributes to SciTS in three ways. Theoretically, it provides a unifying cognitive-process model linking convergence, interdisciplinarity, and team learning. Methodologically, it suggests new measurement approaches focused on externalized knowledge (e.g., artifacts, discourse, representational change) as indicators of team cognition. Practically, it offers guidance for designing interventions, training, and collaborative infrastructures that support knowledge integration in science teams.

In sum, the MITM offers SciTS a needed shift from describing who is on the team to explaining how teams think, learn, and solve problems together, advancing our ability to understand and improve convergence science in practice.