Purpose: Others have shown the ability of an endoscopic simulator to differentiate between psychomotor skills between novice and experienced (>1000 endoscopies) endoscopists. Our goal was to determine whether the virtual reality endoscopic simulator GI-Mentor (Simbionix, Tel Hashomer, Israel) can distinguish the psychomotor skills of intermediately experienced endoscopists (100-250 endoscopies) from those of endoscopic novices. Methods: A total of 5 intermediately experienced (IE) endoscopists and 6 novices (N) endoscopists were evaluated using an abstract psychomotor task on the GI-Mentor virtual reality endoscopic simulator. Each subject received basic instruction and orientation to the simulator and then performed 3 repetitions of the Endobubble task on level 2, which requires the subject to navigate a virtual pipe and burst balloons using a virtual injection needle. If the procedure does not progress in a timely fashion, the balloons will disappear and will not be able to be burst. The number of balloons burst, time required for completion, and number of contacts of the needle with the wall of the pipe was recorded. Results: Subjects in the IE group had performed an average of 146 total endoscopies, while the N group had performed an average of 2.5 total endoscopies. The mean number of balloons burst was 22.9 for the IE group and 8.6 for the N group (p< .0001). Mean time to completion was 103 seconds for the IE group and 116 seconds for the N group, and mean number of wall strikes was 0.25 for the IE group and 2.6 for the N group. Neither of these results reached statistical significance. Conclusions: The GI-Mentor is capable of identifying differences in psychomotor skills between a group of novices and a group with an intermediate level of experience (100-250 endoscopies). These findings suggest that the GI Mentor in its current configuration is sensitive enough to differentiate IE endoscopists from N. While the effects of VR simulator training on performance of real-life endoscopy remains to be evaluated, these data suggest that training should focus on endoscopists early in their learning curve, and that the GI-Mentor allows monitoring of progress in skill acquisition in this group.