Linking gameplay metrics to computational thinking

Computational thinking (CT) is considered to be a fundamental skill underlying not only programming ability, but also an entire array of computational problem-solving competencies in a data-driven world. The need for accessible and engaging educational tools for CT has recently been acknowledged as significant by educational authorities. Glitchspace is a firstperson puzzle game where the player has to manipulate objects in the game world through a visual programming interface in order to escape the cybermaze and win. The game received critical acclaim for its innovative and engaging gameplay, including a British Academy Scotland Award (BAFTA). However, the educational potential of Glitchspace in relation to CT ability has not been sufficiently explored. As success at playing the game relies on the players’ ability to program the game world to suit their goals, it is hypothesised that common measurements of CT will positively correlate with game ability (measured by game metrics, such as completion times and number of failed attempts of various sub-tasks). We propose to develop methodologies to explore the relationship between gameplay behaviour and computational problem-solving ability in the game Glitchspace. Such methodologies will allow researchers to investigate which game mechanics elicit and promote CT in a reliable and measurable way. The outcome of this research will be of great value to instructional designers seeking to leverage games for CT education, as well as learning game analysts seeking to map gameplay behaviour to learning outcomes.