Computational Thinking: Focus on Pattern Identification

Abstract

This article focuses on pattern identification in the context of pupils aged 9 to 15 who are learning programming at school. In this context, programming puzzles that involve moving a robot on a 2D grid using a block-based programming language is common. We consider the ability to identify and formally characterize recurring structures within data or processes, to be a fundamental skill of computational thinking. In this article, we study the case where the motif (i.e. repeating unit) can be identified visually from the grid (obstacles, target..) for tasks involving the use of a loop. We ask what makes motif identification, and thus problem solving, difficult in this context. We provide a quantitative analysis based on the success rates of a hundred tasks from an online programming contest (200,000 participants). We have identified relevant features of the visual motif, which led us to specify five categories according to the degree of correspondence between the visual motif (2D grid) and the algorithmic motif (corresponding loop based program).