The construct of cognitive load (CL) is rooted in the dual-process theory of decision-making, which postulates two distinct cognitive processes that operate largely, but not completely, in parallel (Evans, 2003; Evans & Stanovich, 2013). One of these processes, “Type 1,” is extremely fast, makes minimal demands on working memory, and operates by associatively comparing the current situation to one’s corpus of accumulated prior experiences from long-term memory. Type 1 decision skills are consistent with the recognition-primed decision-making (RPD) approach used by domain experts. By comparison, “Type 2” decision processes involve explicit calculations and conscious deliberation, thereby placing heavy demands on working memory. Type 2 decision skills are consistent with the slow and effortful decision making approach used by domain novices (Kahneman & Klein, 2009). The purpose of the current study was to unobtrusively measure the CL of physicians using wireless, commercial off-the-shelf (COTS) neurophysiological monitors. The participants included a mixed sample of Pediatric Emergency Medicine (EM) physicians (6 novices, 6 experts) who performed four different Virtual Reality (VR) training scenarios (2 clinical scenarios x 2 levels of difficulty). After each scenario, the participants completed a self-reported measure of their mental workload using the NASA-TLX (Hart & Staveland, 1988). The unobtrusive CL measures were significantly correlated with the self-reported mental workload scores. A linear mixed model (LMM) revealed a significant main effect of expertise level (experts had lower CL than novices), as well as a significant expertise-by-clinical scenario interaction. Additional results and implications are presented. This paper concludes with a series of practical recommendations for researchers who wish to use Head Mounted Display (HMD)-based VR training systems while simultaneously using with electroencephalogram (EEG)-based measures of CL.