Threats to professional skepticism are embedded in the social relationships and interactions between auditors and management. These can affect auditor skepticism and the extent of audit procedures performed. In this study, we conduct an experiment using live simulation to create a realistic audit setting to investigate the effect of these interactions on professional skepticism. Each participant (n = 49) completed a measure of trait skepticism and conducted an audit interview with a professional actor trained to play the role of a client controller. Findings indicate that, in general, participants who interview a friendly controller (rather than an intimidating controller) are less likely to determine questionable cash disbursements to be control exceptions and less likely to recommend more intensive follow-up. However, consistent with social psychology research on construct accessibility, planned contrasts indicate that participants who score low on trait skepticism are least likely to identify control exceptions and recommend more intensive follow-up when interviewing a friendly controller. This study advances research on professional skepticism by examining the impact that type of social interaction (friendly, intimidating) has on professional skepticism using a methodology (live simulation) that allows us to simulate a realistic audit environment. Use of this methodology increases external validity and generalizability of our findings. As a result, this study corroborates concerns that the social relationships/interactions between management and the auditor can be a threat to professional skepticism, and allows us to understand better how, when, and where these threats occur.