Dissection and vivisection are traditional approaches to biology laboratory education. In the case of human anatomy teaching laboratories, there is a long tradition of using human and animal cadaver specimens in the classroom. In a review of the literature comparing traditional dissection and vivisection lessons to alternative lessons designed to reduce the time spent dissecting or the numbers of animals used, we conclude that it is difficult to come to any conclusion regarding the efficacy of different approaches. An analysis of the literature is confounded because many studies have very low statistical power or other methodological weaknesses, and investigators rely on a wide variety of testing instruments to measure an equally varied number of course objectives. Additional well designed studies are necessary before educators can reach any informed conclusions about the efficacy of traditional versus alternative approaches to laboratory education. In our experiments, we compared a traditional cat dissection based undergraduate human anatomy lesson to an alternative where students sculpted human muscles onto plastic human skeletons. Students in the alternative treatment performed significantly better than their peers in the traditional treatment when answering both lower and higher order human anatomy questions. In a subsequent experiment with a similar design, we concluded that the superior performance of the students in the alternative treatment on anatomy exams was likely due to the similarity between the human anatomy representation studied in lab, and the human anatomy questions asked on the exams. When the anatomy questions were presented in the context of a cat specimen, students in the traditional cat dissection treatment outperformed their peers in the alternative treatment. In a final experiment where student performance on a human anatomy exam was compared between a traditional prosected human cadaver treatment and the alternative clay sculpting treatment, no significant difference were detected, suggesting that the complexity or simplicity of the anatomy representation is less important than the similarity between the learning experience and the testing experience.