Reconceptualizing Trust: Defining, Modeling, and Measuring Trust

Simone Erchov

Advisor: Patrick E McKnight, PhD, Department of Psychology

Committee Members: Tyler Shaw, Seth Kaplan

David J. King Hall, #2073A
April 21, 2017, 10:00 AM to 07:00 AM


Research across multiple domains deem trust a relevant construct, yet trust remains poorly understood and measured.  The lack of understanding stems from an inability to concisely and accurately define and model trust within and across domains (Watson, 2005; PytlikZilling & Kimbrough, 2016).  An empirically supported and sound definition (and model) of trust allows for exploration of the universal mechanism by which trust forms and, accordingly, prediction of future behaviors.  This dissertation aims to examine a new conceptual model of trust that proposes a universal mechanism through the interaction of three key components - goal importance, reliance, and uncertainty.  The first experiment examines performance of this model against several alternative model comparisons to identify which model best predicts trust across several scenarios.  The second experiment validates the best performing models identified in the first experiment against a new data set to ensure replicability of the initial findings.  Results support that trust is best defined and measured as an emergent, state variable resulting from the interaction of assessments of goal importance, reliance, and uncertainty.  However, several areas of inquiry remain regarding improvement of the measurement model, understanding of the dynamic, cyclical process of trust over time, and how cognitive trust relates to behaviors.