Expanding the Theory of Cognitive Fit: A Longitudinal Quantitative Study

It is shown in literature that adaptive information visualization based on the variability of specific information content can improve the efficiency and effectiveness of managerial decision making. Existing approaches are predominantly based on the theory of cognitive fit, which accounts for the fit between a specific visualization and a particular task, as well as for the moderating role of various task complexity levels. However, this current approach is restricted in its predictive power given a specific user or user group due to its apparent neglect of individual factors that would determine the optimal visual representation given a specific context concerning data at hand and the users’ respective information requirements. We argue that a more comprehensive model including individual factors is needed to best support information retrieval, cognition, and as a consequence decision making. Thus this paper sets out to further our understanding of how situational and individual factors influence these processes. Using a controlled longitudinal experimental setting, we introduce an enhanced structural model demonstrating a significant increase in predictive power over the previous simple task complexity/visualization model. The predictive power of the adapted model is increased by including data complexity and individual complexity (represented by knowledge, pervious experience, biorhythm, and spatial ability) as additional factors. The overall fit of the proposed model in comparison to the original model increases from 0.129 to 0.064 (based on SRMR).