This research addresses the question of how an important computer programming skill -- debugging -- is acquired by elementary and middle- school children, and what the cognitive consequences are of acquiring that skill. One preliminary goal of the proposed research is the development of a well-specified and empirically-supported account of what a child has to know in order to debug a computer program. The proposed work extends preliminary studies of the debugging skills acquired during the course of a normal LOGO curriculum, and it provides for further assessment and elaboration of a computer-simulation model of the precise components of debugging skill. The work will have several interacting components: (1) Empirical evaluation of the model based on a study of experienced programmers' debugging processes; (2) Further development of precise assessment procedures for determining what component skills a student has acquired; (3) Extending the model to a wider range of programming contexts; (4) Using the model to guide specific instructional procedures in teaching debugging skills; (5) Extending and applying a complete one-semester LOGO curriculum with an emphasis on instruction in an assessment of debugging skills; (6) Determining the extent to which debugging skills, once taught, can transfer to near and far tasks; (7) Construction of a model-based prototype debugging aid on a powerful graphics workstation.
