Abstract:
Several reviews of trust research suggest that trust is suitably measured by a singular, mechanistic model, yet research on trust has yet to yield one. Research on trust is fragmented both between and within different fields of study because the state-dependent, subjective nature of trust is difficult to examine. As a result, trust is often modeled by behaviors or beliefs that fail to generalize. The lack of generalizability perpetuated the fragmented state of trust research. As previously stated, several reviews suggest that a universal model is not only possible but appropriate considering researchers often identify a common set of core ingredients for trust. I - along with my colleagues - responded to this suggestion by developing and testing a more generalizable model that incorporates these common ingredients. Together these common ingredients provide a more parsimonious, universal mechanistic model of trust that we refined over several empirical studies. The current study addresses several lingering concerns from this previous work and then confirms any changes that may be both theoretically and empirically warranted. From two independent efforts - one exploratory and one confirmatory - I demonstrated that the original model predicted trust best via a fixed effects model specifying trust as the result of a three-way interaction of goal importance, reliance, and uncertainty. Additionally, I discussed the implications of this model with future ideas for refinement and continued testing.
