In this paper we investigated the impact of feedback and guidance on the development of computational thinking skills. To achieve this, we extended a game-based learning environment that aims to foster computational thinking by teaching programming in self-regulated learning scenarios. The learning environment has been enriched with multiple mechanisms to guide learners and provide feedback that is directed towards the development of computational thinking skills, particularly specific abstractions in programming among algorithmic thinking. To assess the impact of guidance and feedback, we conducted an empirical study with 57 participants. The findings indicate that feedback on the logical artifacts can reduce certain code smells and increase the motivation on the part of the learners.

In this paper we presented an extension of the game-based learning environment ctGameStudio. The new version includes feedback and guidance components that analyse the learner's code and provide appropriate guidance that improves learning, particularly the inspection, reflection, and evaluation of own logical (code) artefacts. We hypothesized that this would result in higher performance regarding computational thinking.
The results regarding the differences in the performance parameters between two groups showed no significant improvement in most performance parameters. However, the significant mean difference in number of duplicate blocks shows that the concepts are well explained to the participants and that the acquired knowledge can be transferred. In addition, there is a tendency for the feedback group to take a bit more time and complete slightly fewer levels, which can be explained by the fact that the participants had to deal with the feedback and interact with the component. However, these differences are not significant. Additionally, a significant improvement of the flow could be observed in the feedback group. Flow is repeatedly described in literature as a particularly important factor in the learning process, so that this result is an indication that the changes to the game experience were effective.
Since there is a considerable amount of literature on the benefits of formative and summative feedback in gaming-based learning environments, e.g. [Gr18, LK11], the reason for the small group differences regarding the performance of the learners may be due to the implementation of the feedback system and the design of the study. Even though there were small performance improvements in the feedback group, there were no significant differences in the ctscore. The ratios of the parameters for the calculation of the score may need to be revised. Although, ctGameStudio is created for first-time game-based learning of CT the participants of the study had a high heterogeneity in terms of programming experience, which may have led to mixed results. Since the game was only played for 45 minutes and it takes extra effort to read the feedback dialogs, the benefits might only be realized with longer observation.