A survey was conducted among graduates of two Canadian medical schools who have been in practice for more than 9 years. The purpose of the study was to test the hypothesis that graduates of a problem-based curriculum differ from graduates of a traditional curriculum in their attitude to and participation in continuing medical education (CME) activities. Differences were noted in the rate of participation in certain CME activities (attendance at national and international conferences and meetings) between specialists and family doctors in both groups of alumni. However, the data indicate that the differences in learning-teaching methods employed in the course of the undergraduate medical curriculum do not exert a decisive influence upon the learning habits of the graduates.