Examining the convergent validity of shared mental model measures

Despite widespread interest in shared mental models (SMMs) within teams and groups and an extensive empirical literature examining SMM-performance links, very little is known about the convergent validity of commonly used measures of SMMs. In this study, two-person teams (n = 96) engaged in a complex flight task and completed three SMM measures: concept mapping, paired ratings, and causal mapping. Task-based sharedness scores were compared across the measures. Analyses were conducted in two ways: using SMMs of actual team members (n = 96 pairs) and using the SMMs of pairs of participants who worked separately but for whom similarity indices were calculated after the study (n = 18,240). The purpose of the latter pairs, coined pseudo-partners, was to create a sample with considerable power to test the convergent validity of the SMM measures. The results call into some question the convergent validity of these task-based SMM measures.