While there has been a wealth of performance criteria proposed and validated for objective assessment of surgical skills, only less than one third of the metrics are adopted in clinical practice, and even fewer are used for medical examination. This is due in part to the fact that the evaluation criteria proposed are focused on the task-centered outcomes which are not generalizable to broader sets of emerging surgical procedures. With increasing pressure from the media, regulatory bodies, and patients, to scrutinize physician performance, generalized standard criteria for assessing surgical performance are needed. This can be realized only if Multivariate Performance Criteria (MPC) are defined with respect to inherent human capabilities, rather than to transient task conditions, and if these criteria become related and prioritized consistently. In light of this, the current study proposes a method for relating and prioritizing the multivariate criteria through the use of Learning Curves. To this end, ten desired capabilities were identified from a Task Analysis of the Central Venous Catheter (CVC) placement procedure. In the preliminary study, motion tracking of video footage for six medical residents, taken at the beginning, middle, and end of a three-months-long training session, was used to confirm learning of the desired capabilities. Specifically, the economy of motion, eye-hand coordination, motion stability, and visual detection were tracked and quantified according to the defined metrics. Among them, both economy of motion and eye-hand coordination had a statistically significant learning effect (p<0.05) with the different learning rates-eye-hand coordination needed more training to reach a plateau. Our results imply that eye-hand coordination is relatively more comprehensive capability and requires to set higher priority in training. The current method can contribute to establishing hierarchical structures in MPC based on learning curves, ultimately building a standard system of performance. The results of this proceeding have implications for using learning curves to assess medical skill performance in a systematic manner.