Informatik als Denkwerkzeug hat nicht nur im Bildungs- und Wissenschaftsbereich etwas zu bieten. Das Problemlöseargument zielt auf den Alltag: Informatik stellt Werkzeuge und Verfahren zur Verfügung, mit denen sich im Alltag Probleme strukturiert beschreiben, diskutieren und damit besser lösen lassen – auch ohne den Einsatz von Computern. So können zum Beispiel Flussdiagramme helfen, Abläufe zu verstehen und zu optimieren, unterschiedliche Datenstrukturen wie Listen, Tabellen, Bäume und Graphen unterstützen das problemgerechte Erfassen, Verarbeiten und Darstellen von Daten, und Visualisierungstechniken wie Concept-Maps helfen beim Nachdenken über Strukturen und Zusammenhänge. Bei der Beschäftigung mit Informatik lernt man diese Werkzeuge kennen und schult auch das entsprechende Denken. Viele dieser Konzepte und Werkzeuge existieren nicht erst seit der Entstehung der Wissenschaft Informatik. Doch erst die Informatik macht den Umgang mit solchen Werkzeugen explizit zum Thema. Guter Informatikunterricht fördert daher nicht nur die Nutzung solcher Konzepte und Werkzeuge, sondern hilft Schülerinnen und Schülern auch, künftig selbst die geeigneten Denkwerkzeuge zu finden. So eröffnet Informatik die Möglichkeit, über Problemlöseheuristiken oder den Unterschied zwischen Korrektheit und Viabilität nachzudenken.

Clark (1992) kommentiert diese Annahme nach einer Durchsicht entsprechender Studien: »The original hope that the learning of computer programming would sharpen general thinking and problem-solving skills seems unsupported. Thus, there seems to be a lack of compelling evidence of ›domain-general‹ transfer which is attributable to computer programming expertise« (267ff.). Die Transferqualität des Programmierens sei unbewiesen geblieben, Programmieren bleibe domain-spezifisch (wie es im übrigen ja auch Latein war). Wo Transfer nachgewiesen wurde, war der Computer kein notwendiges Medium, und die Transferfähigkeit wurde mit speziellen didaktischen Methoden trainiert. Clark bezeichnet die Verwechselung der menschlichen Informationsverarbeitung mit dem Programmieren als unzulässige Reifikation einer Metapher für Kognition.

The intuitively plausible claims for the cognitive benefits of programming have broadened in scope and in public attention. Although evidence does not support these claims as yet, their presumed validity is nonetheless affecting important decisions in public education, and leading to high expectations for outcomes of programming in the school and home. In the current climate of uncritical optimism about the potential cognitive benefits of learning to program, we run the risk of having naive "technoromantic" ideas become entrenched in the school curriculum by affirmation, rather than by empirical verification through a cyclical process of research and development. Already at the pre-high school level, programming is taught primarily because of its assumed impacts on higher cognitive skills, not because proficiency in programming is itself an educational goal.

The contrasting belief, in part a reaction to the first belief, is that through learning to program. children are learning much more than programming, far more than programming "facts". It is said that children will acquire powerfully general higher cognitive skills such as planning abilities. problem-solving heuristics and reflectiveness on the revisionary character of the problem solving process itself. This belief. although new in its application to this domain, is an old idea in a new costume which has been worn often before. In its common extreme form, it is based on an assumption about learning - that spontaneous experience with a powerful symbolic system will have beneficial cognitive consequences, especially for higher order cognitive skills. Similar arguments have been offered in centuries past for mathematics, logic. writing systems, and Latin (e.g. see Bruner, 1966; Cole & Griffin, 1980; Goody, 1971; Olson, 1976; Ong, 1982; Vygotsky, 1978).

Despite the absence of substantial proof of a positive relationship, there are still those who support the proposed link between programming language instruction and problem-solving skills. These researchers address deficiencies in those research studies that have not found these predicted positive relationships. These deficiencies can be characterized according to four major issues:

• (a) programming language/ problem-solving studies not being firmly grounded in problem-solving theories (Burton & Magliaro, 1987-1988);
• (b) quality, length, and intensity of the treatment presented (Burton & Magliaro; Palumbo & Reed, 1987-1988; Seidman, 1988; Soloway, Spohrer, & Littman, 1988);
• (c) appropriateness of the programming language selected in increasing problem-solving skills and the method of instruction (Burton & Magliaro; Littlefield, Delclos, Lever, Clayton, Bransford, & Franks, 1988; Reed et al., 1987-1988); and
• (d) selection of an appropriate sample of students, whose age range and ability level will provide the necessary background to benefit from programming language instruction (Linn & Dalbey, 1985; Pea, 1984).

One of the main arguments for teaching mathematics is the development of the exact way of thinking that finally results in the ability to use the exact language of mathematics for describing, analyzing and solving problems in all areas of our life. This ability becomes more and more important. Some colleagues tend to call informatics the “new" mathematics or at least a constructive mathematics. Jeannette Wing, head of the computer science department at Carnegie Mellon University, even envisions that “thinking like a computer scientist" should be a fundamental skill such as reading, writing and arithmetics. Cohen and Haberman regard computer science as one of the five “languages" every citizen should acquire.
The reason for that is the way of working in computer science. Similarly as in mathematics, we begin with an abstract description of a problem and continue with its analysis. But additionally, computer scientists do not only discover an efficient way of solving it, but they also implement the discovered method and provide a product (program) for solving problems of this kind. This work is more constructive than the typical work of a mathematician and ties the exact way of thinking in mathematics with the pragmatic way of working in engineering.

Studies of computer programming and problem solving have yielded mixed results. Studies that failed to find a relationship between computer programming and problem solving had various weaknesses in experimental design and instructional approach. Specifically, many studies employed small samples [8, 12], did not use random selection and/or assignment of the subjects [8, 9, 10, 11 ], or lacked control groups [9].
In addition, the main instructional approach in the non-support studies was non-directive (discovery) [8, 9]. Program planning, development, and debugging were not specifically taught [8, 9]. This is a problem, since using a non-directive instructional approach with a limited amount of treatment time does not appear to be effective.
Another difficulty is that mastery of programming was not measured [8, 12, 10, 11]. Without ensuring that programming is mastered, it makes little sense to talk of problem solving transfer [13].
A number of other studies showed some positive relationship between computer programming and problem solving ability. Some of these studies used random selection and/or assignment [14, 15, 16, 12, 17, 18]. Random selection should result in sample groups that are more closely related to the population. Several studies were for longer periods of time (semester or more) [19, 17, 20]. Longer studies should allow for more treatment time. The instructional strategy used in a number of these studies was directed with specific instruction in program planning and development [15, 16, 12, 17, 18). Tbe results of some of the studies are not conclusive since the programming activities and dependent variables appeared to be highly related [16, 18]. Even though the inferences were not clearcut, these studies are a beginning of an experimental process directed at investigating a possible link between computer programming instruction and generat problem solving ability.