Objective: The purpose of the study was to investigate the relationship between authenticity (how close to actual practice) of preclerkship instructional formats and preclerkship and clerkship outcome measures. A secondary purpose was to investigate the effect of student's small-group assignment during preclerkship period on instructional formats and preclerkship and clerkship outcome measurements. Methods: A prospective, randomized, crossover study was carried out with preclerkship students taking a clinical reasoning course. Students were randomized to small groups and exposed to three formats of differing instructional authenticity across three subject areas. Three student cohorts were taught using one instructional format per subject area with each cohort receiving a different instructional format for each of the three areas. Outcome measures at the end of the preclerkship year and the clerkship year were selected to determine the effect of each instructional format. Hierarchical linear modeling was performed to assess impact of format on outcomes and to assess potential group effect on outcomes. Results: Increasingly authentic instructional formats did not significantly improve clinical reasoning performance. Small-group assignment did not significantly contribute to the outcomes providing evidence that teacher training was successful. Conclusions: Increasing authenticity of instructional formats does not appear to significantly improve clinical reasoning performance. Faculty can be successfully trained to teach consistently in a clinical reasoning course. Medical educators should balance increasing authenticity with factors such as cognitive load and learner experience as well as engaging in frame-of-reference training to minimize group effects with designing new instructional formats.