<p>Based on the OPTIMAL (Optimizing Performance Through Intrinsic Motivation and Attention for Learning) theory, we hypothesized that enhancing expectancies by decreasing the perceived difficulty of a surgical task through an optical illusion that decreased perceived task difficulty would improve surgical skill acquisition. We also hypothesized that men would report higher confidence than women, while demonstrating similar task performance. We conducted a randomized controlled trial comparing the Ebbinghaus illusion (which makes targets appear larger than reality) (enhanced expectancies [Enhanced]) to no illusion (normal expectancies [Normal]) while learning an arthroscopic targeting task. Medical student participants of all years were randomized to Enhanced or Normal groups. Both groups underwent pre-test, training (10 task repetitions), immediate post-test (training model without illusion and transfer model), and retention test (same as immediate post-test, 24-hours later). Sixty-seven medical students (52% male) participated. We found no significant difference between study groups (Enhanced versus Normal) in immediate non-transfer (mean difference=-5.6, 95% CI=[-13.1,2.0]), immediate transfer (mean difference=-2.5, 95% CI=[-0.97,4.7]), retention non-transfer (mean difference = 7.2, 95% CI=[-1.8,16.2]), and retention transfer (mean difference = 0.1, 95% CI=[-4.8,5]) performance. In analyses of covariance, men had significantly higher confidence levels compared to women (adjusted mean difference = 0.9, 95% CI=[0.3, 1.5], <i>p</i> = 0.006). We were unable to demonstrate that using an optical illusion to enhance expectancies while learning an arthroscopic targeting task improves immediate, transfer, or retention performance. Male trainees reported lower confidence than women, despite similar task performance.</p>