This paper describes experiments designed to explore the results obtained from user tests of computer games; the tests sought to identify problems in using the games. The experiments tested adventure-type computer games for young children. The subjects were children (a group of 25) between the ages of five and seven who were selected from an elementary school in the Netherlands. The study looked at differences between problems observed when the children first played the game and problems observed after the children had an hour of experience with the game.

Problems were noted in the areas of how efficiently the children played the game, how effective the children were at realizing the goals of the game, and how much satisfaction (or fun) the children had while playing the game. The results point to a radical difference in the types and severity of problems that are detected, even after a limited amount of practice. The differences follow patterns that are predictable based on human problem-solving theories, so the results of the experiments may be applicable for developers looking to establish user testing for specific purposes.

The paper is well written and informative. The results and discussion contained in it should be of interest to others doing related research, as well as to developers interested in user testing issues.