Presence, the impression of being existent in artificially created places, is an important factor for achieving maximum experience in virtual environments (VEs) and hence in their use for experiments or therapy. To date, most clinical studies have used self-report questionnaires or physiological measures to appraise the degree of presence. Some studies that have tested behavioral scales have used a single VE, required additional equipment, or did not test psychometric properties during immersion in the VR. The current study described and tested an operationalized, easy-to-use scale of presence, the Behavior Presence Test in Threatening VEs (BPTT), through the observation of overt behavior exhibited by a sample of 40 participants during immersion in five different threatening virtual situations. The BPTT was compared with standard subjective and physiological measures of presence and anxiety. The VEs resulted in clear increases in anxiety and arousal. Most importantly, findings demonstrated the consistency, reliability and validity of the BPTT. Video game experience did not seem to have any impact on presence or on participants' behavior. The test is a step forward in creating an objective assessment of presence during the virtual reality experience and without the need for additional equipment. Future work should focus on developing broader tests suitable for use in any threatening or nonthreatening VEs.