Abstract: The past few decades, interest in team learning behavior has increased significantly. Most of previous studies paid attention to the effects of learning behavior on group outcomes, but less work was done on how the effect takes place. In other words, the mechanism which transfers the learning behavior into group effectiveness was far from straightforward. The present study seeks to examine the mediating effect of transactive memory system (TMS) on the relation between team learning and performance; and analyzes the dynamic mechanisms of team learning within a time series framework. We employed an input-mediating-output-input (IMOI) approach as proposed by Ilgen et al. (2005), instead of the traditional input-process-output (I-P-O) model in industrial & organizational psychology.
Data was collected from an operational business-planning competition held in a comprehensive university in China. Each group was requested to have 3~8 members, worked together for 6 months, and finally completed a business plan as a team output. Ninety-nine groups were recruited as our initial samples at the first stage, and we obtained the measurement following a field study paradigm. After three months we traced the initial samples and obtained fifty-five groups at the second stage. Finally, based on testing internal consistency of intra-group data, we aggregated the bi-teammate data into group level data at both stages.
Based upon hierarchical regression analysis, the results show that the mediating effect of TMS on the relation between team learning and performance were promising at both stages, which means the TMS can adequately interpret how team learning influences team performance as a mediator. Another interesting result of this study is that team performance; as an output received at the end of stage one, also acted as an important input variable at stage two, which in turn positively influenced the following team learning process. The circular causal model based on path analysis showed that IMOI approach can be used to interpret organizational mechanisms better than the classic I-P-O approach; the result was consistent with the new trends within the team relevant IO psychological understanding.
Several theoretical understandings and practical implications were obtained from the present study. Briefly, the results suggest that developing and maintaining a TMS is critical to achieve team outputs under a team learning setting. In addition, performance evaluation and feedback are also important factors within team learning processes from a time series framework. Third, we argued that organizational behavior researches based on an IMOI approach would have more generalizability and ecological validity than traditional I-P-O model.

Key words: team learning, transactive memory system, team performance, IMOI, longitudinal design