Objective: To assess the effectiveness of a computer-aided learning program on dementia. Design: Fourth year medical students were arbitrarily assigned to groups that used a computer-aided learning program (65) or had a tutorial covering similar material (73). These sessions were in addition to a base curriculum in a two-week course in geriatric medicine. Main outcome measures: The effectiveness of the teaching sessions was judged by the performance on a multiple choice questionnaire about dementia, given to the students on two occasions, one at the beginning and one at the end of the two-week course. Results: Both groups of students scored significantly better on the second test (computer group, 66% [95% confidence interval, 64-69] to 81% 179-83] and tutorial group, 66% [63-67] to 74% [73-77]). The difference between the groups at the start of the course was not significant (F1,136 = 0.61, P = 0.61); however, there was a significant difference between the groups at the end of the course (F1,136 = 21.83, P<0.001). Conclusion: Both groups improved their knowledge of dementia during the two-week course. Students who used the computer-aided learning programs showed a greater improvement in score. Computer learning programs are effective learning tools and are a useful addition to traditional teaching methods. Further study is required to assess the effects of computer-aided learning programs in long-term studies of dementia knowledge.