PURPOSE: Arthroscopic skills training outside the operative room may decrease risks and errors by trainee surgeons. There is a need of simple objective method for evaluating proficiency and skill of arthroscopy trainees using simple bench model of arthroscopic simulator. The aim of this study is to correlate motor task performance to level of prior arthroscopic experience and establish benchmarks for training modules. METHODS: Twenty orthopaedic surgeons performed a set of tasks to assess a) arthroscopic triangulation, b) navigation, c) object handling and d) meniscus trimming using SAWBONES "FAST" arthroscopy skills workstation. Time to completion and the errors were computed. The subjects were divided into four levels; "Novice", "Beginner", "Intermediate" and "Advanced" based on previous arthroscopy experience, for analyses of performance. RESULTS: The task performance under transparent dome was not related to experience of the surgeon unlike opaque dome, highlighting the importance of hand-eye co-ordination required in arthroscopy. Median time to completion for each task improved as the level of experience increased and this was found to be statistically significant (p < .05) e.g. time for maze navigation (Novice - 166 s, Beginner - 135.5 s, Intermediate - 100 s, Advance - 97.5 s) and the similar results for all tasks. Majority (>85%) of subjects across all the levels reported improvement in performance with sequential tasks. CONCLUSION: Use of the arthroscope requires visuo-spatial coordination which is a skill that develops with practice. This simple box model can reliably differentiate the arthroscopic skills based on experience and can be used to monitor progression of skills of trainees in institutions.