Collaborative teaching strategies such as peer instruction and conventional group work have previously been shown to enhance meaningful learning, but they have not previously been compared. In this present study, we compared the impact of solving quizzes with peer instruction and conventional group work on immediate learning in a laboratory exercise. A total of 186 second-year medical students were randomized to solve two quizzes by either a peer instruction strategy (n = 93) or conventional group work (n = 93) during a mandatory laboratory exercise on respiratory physiology, after which all students completed an individual test. There was no difference in total test scores between groups, but students randomized to peer instruction obtained the highest test scores in solving simple integrated questions. Conversely, students randomized to conventional group work provided the best evaluations of the overall assessment of the laboratory exercise. In conclusion, different collaborative teaching strategies implemented during a laboratory exercise appear to affect immediate learning and student satisfaction differently.