Computer-assisted learning is known to be an effective tool for improving learning in both adults and children. Recent years have seen the emergence of the so-called "serious games (SGs)" that are flooding the educational games market. In this paper, the term "serious games" is used to refer to video games (VGs) intended to serve a useful purpose. The objective was to review the results of experimental studies designed to examine the effectiveness of VGs and SGs on players' learning and engagement. After pointing out the varied nature of the obtained results and the impossibility of reaching any reliable conclusion concerning the effectiveness of VGs and SGs in learning, we stress the limitations of the existing literature and make a number of suggestions for future studies. (Contains 2 tables.)