Nederlands Tijdschrift voor de Psychologie vol:42 issue:7 pages:364-372
An important argument for teaching children computer programming (in Logo) is the contention that learning to program can foster the acquisition of generalizable cognitive skills and problem solving abilities. In this article, we examine if this 'cognitive-effects hypothesis' is supported by the findings of the available empirical research. The results of this examination do not yield convincing evidence in favor of this optimistic hypothesis. However, our analysis shows also that the disappointing results were obtained in instructional conditions that cannot be considered to optimally foster the development and transfer of cognitive skills. In this respect, some findings of very recent research are mentioned, suggesting that if one succeeds in overcoming the shortcomings of the previous studies, moderate transfer effects can indeed be obtained.