A study was done of the relation of time for individual study versus instruction time in a non-traditional, problem-based medical curriculum at the University of Limburg (Netherlands). The study collected data on 86 courses conducted in 5 consecutive academic years. In this problem-based approach, each curriculum year in the first 4 years comprises six 6-week courses which are organized around interdisciplinary themes central to medicine. During each course students met twice a week for a 2-hour small group tutorial in which problems were analyzed and learning goals formulated. Most of the learning took place individually. Scheduled activities included 4 hours a week of small-group tutorials, 2-to-4 hours of lectures and 2-to-4 hours of training in medical skills. The independent variable was the amount of instruction time and the dependent variable was the average amount of time spent on individual study. The individual study time was measured by asking students after each course to estimate the number of hours per week he or she spent on self-directed learning activities. Results of data analysis indicated that increasing instruction time generally led to a diminishing increase in individual study time. Included are 4 figures, 1 table and 19 references. (JB)