TraIN: Transfer of cognitive skills from learning computer science at school

Researchers Michal Berkowitz Urs Hauser Giovanni Serafini

Computer science is the scientific discipline concerned with the development of algorithms and programs. It is at the core of the contemporary digital transformation of society, economy, and politics. Formal problem solving by means of concepts and methods of computer science is usually termed as «algorithmic thinking» or as «computational thinking». According to educators and researchers, algorithmic thinking is not directly promoted in the subjects that are traditionally taught in school. Therefore, not only in Switzerland, but everywhere in the world, computer science is rapidly becoming a mandatory school subject.

Computational thinking and mathematical problem-solving draw both on spatial abilities. School curricula in computer programming and educational robotics as well as teaching computer science by means of unplugged activities are vehicles that expose pupils to computational thinking.

In the TraIN (Transfer Informatik, in German) project, the Chair for Research on Learning and Instruction and the Chair of Information Technology and Education research far transfer effects of teaching computer science on spatial abilities, on deductive reasoning, on mathematical problem-solving, and on complex problem solving.

The TraIN project currently comprises two randomized intervention studies:

• In a study with 5th-graders, we plan to investigate the differential transfer effects of teaching Logo programming and of teaching computer science through unplugged activities. The primarily goal of the project is to compare curricula for the two teaching approaches and to assess whether they help acquiring computational thinking skills. Moreover, the project is driven by the question if those curricula promote transfer beyond the boundaries of computer science itself. To this end, new instruments for measuring computational thinking skills, and for assessing mathematical word problem solving skills will be developed.
  
  
• In another study, we investigate and compare the transfer effects of educational robotics and of turtle-graphics programming on problem-solving skills in secondary school learners (9th and 10th grade). In particular, we are interested in the extent to which computational thinking, spatial reasoning, and complex problem solving can be promoted using the two different approaches. Complex problem solving involves tasks that change dynamically over time or through user interaction, where regularities of the environment can only be uncovered through successful exploration and integration of the information gained in the problem solving process. Within the study, both curricula and programming libraries as well as new tests (for evaluating the power of computational thinking) are developed and validated.