User acceptance of knowledge-based system recommendations: Explanations, arguments, and fit

Knowledge-based systems (KBS) can potentially enhance individual decision-making. Yet, recommendations from KBS continue to be met with resistance. This is particularly troubling in the context of deception detection (e.g., border control), in which humans are accurate only about half the time. In this study, we examine how the fit between KBS explanations and users' internal explanations influences acceptance of KBS recommendations. We leverage cognitive fit theory (CFT) to explain why fit is important for user acceptance of KBS evaluations. We also compare the predictions of CFT to those of the person-environment fit (PEF) paradigm. The two theories make conflicting predictions about the outcomes of fit when it comes to KBS explanations. CFT predicts that explanations with a higher cognitive fit will have more influence and be evaluated faster whereas PEF predicts that individuals will take more time in evaluating explanations with greater fit. In our deception detection scenario, we find support for CFT in the sense that people are influenced more by cognitively fitting explanations, however PEF is supported in the sense that people take more time to evaluate the explanation.