“So, we kind of started from scratch, no pun intended”: What can students learn from designing games?
Much research attention has been focused on learning through game playing. However, very little has been focused on student learning through game making, especially in science. Moreover, none of the ...
In a constructionist learning environment, game designers engage in generating an artifact, iteratively test and refine it through playtesting by peers, and keep the goal of public communication in the forefront (Laurillard, 2020; Wilson, 2020). Wilson (2020) notes that “the active work of the learner is evident and activated in the building of ‘public’ artifacts” (Wilson, 2020, p. 17) as a key characteristic of constructionism.
In a related study, we report that spontaneous critiques allowed students to be knowledgeable authorities and helped to facilitate iteration as students worked to improve their games (Tucker-Raymond et al., 2019) both with respect to climate content and to the player experience. This study extends these findings by documenting other instances where peer-to-peer interactions supported learning and additional skill development. For example, Jack's comment on Danny and Stavros' disagreement about Boolean operators in Scratch (Screenflow 20161114) resulted in further research and learning while problem-solving discussions within Sharon, Allie, and Nate's group resulted in debating what components of the carbon cycle to include (Screenflow 20161104).
Kafai and Burke (2016) point out that constructionism equally situates “cognition not just in the head but also in space” (p. 86). Furthermore, there is evidence that, as Kafai and Burke (2016) note, systems thinking as instantiated in this study aligned with “growing interest in using complex system thinking as a framework to approach science learning and the notion of CT as designing a system” (p. 33).
As the evidence shows, in this constructionist learning environment students took responsibility for their own learning. Peer programming and collaboration on production of a shared artifact are key features of these learning environments (Dishon & Kafai, 2020; Papert & Harel, 1991). Constructionism afforded students autonomy and agency; students in this study freely chose the science topic and how they would model it in the design of their games. In addition, we saw many instances of their drawing on multiple resources, consulting peers who had more Scratch expertise, and sharing their developing games with others. We observed all three groups conducting Google searches to obtain additional science information on an as-needed-basis.
The students' high level of engagement, persistence and agency was noteworthy for us. The novice group in particular persisted in trial-and-error and thence to more purposeful troubleshooting, only asking for help from peers and the researcher when completely stuck. All groups worked on their games outside of class time; progress on their games was often evident to us on several of the mornings as we entered the classroom.
As Mambrey et al. (2020) and Samon and Levy (2019) point out, a focus on interactions among system components is critical to recognizing and understanding emergent systems behavior.
This study has shown how a thoughtfully designed learning environment shaped by constructionist theory can support meaningful engagement and learning by students with a range of programming experience. The Building systems from Scratch curriculum allowed for multiple entry points for students; all students were expected to grow from the place at which they started; all did.
understanding systems is a critical component of science literacy
We define systems thinking as “a set of analytic skills used to improve the capability of identifying and understanding systems, [and] predicting their behaviors” (Arnold & Wade, 2015, p. 675). Identifying systems components and the interactions among them are very common elements in various systems frameworks and are essential for student understanding of systems in science (e.g., Nguyen & Santagata, 2020; Rachmatullah & Wiebe, 2022; Samon & Levy, 2019), particularly in studying climate change (Bhattacharya et al., 2020). Furthermore, focusing student attention specifically on interactions among system components is needed to support them in understanding causal mechanisms (Mambrey et al., 2020; Penner, 2000; Samon & Levy, 2019; Yoon et al., 2018) and in recognizing emergent behavior, particularly in climate systems (Pallant & Lee, 2015). In addition, understanding which aspects of climate change are anthropogenic is essential and still a challenge for both teachers and students (Lundholm, 2019).
In Building systems from Scratch, students are tasked with applying systems thinking as they identify the components of the system (e.g., greenhouse gases, solar radiation, carbon sources [fossil fuels], carbon sinks [sequestration by trees], and anthropogenic actions that exacerbate or mitigate global warming). They explicitly explore the nature of connections among components and focus on the dynamics or behaviors of the system, including feedbacks (Yoon et al., 2018). They construct systems diagrams to depict systems dynamics, and thus enhance their understanding of causal mechanisms (Khajeloo & Siegel, 2022).
Key and defining features of a constructionist learning environment for game making are that game designers engage in generating an artifact, usually collaboratively (Papert, 1980; Papert & Harel, 1991). Designers iteratively test and refine their games through playtesting by peers, thus considering the player's perspective during design (Dishon & Kafai, 2020). Building a public artifact intended for used by others is an end goal, thus communication is a key design consideration (Laurillard, 2020; Wilson, 2020).
They found that designing educational games allowed students to represent their understanding of science in a personal and meaningful way, while democratizing participation in the classroom through supporting peers.
A unique feature of the Building systems from Scratch program is the inclusion of an established systems-based constructionist framework, triadic game design (TGD), as a heuristic to support the design of serious games (Susi et al., 2007), that is, games designed for a primary purpose other than entertainment. In this program, the purpose of game design is to teach others about some aspect of climate change. We leveraged the TGD framework (Harteveld, 2011) to support student game creation since this framework sets up a system consisting of three interacting game design dimensions: Reality, Meaning, and Play. Reality represents the connection between game and real worlds, suggesting that any game contains an underlying model of reality (Troiano, Schouten, et al., 2020b). This is often deployed through the representation of real objects in-game (e.g., CO2 molecules, bicycles), or through the implementation of real-life physics and mechanics (e.g., ice melts at 4°C; cars go around buildings not through them). Meaning represents the underlying learning goal, for example, to change a behavior or to educate players (Puttick et al., 2018; Troiano et al., 2019; Troiano, Schouten, et al., 2020c). Play represents the genre (e.g., narrative, puzzle, simulation) or gameplay mechanics of a game, which define the experience of the player.
Kafai and Burke (2016) point out that systems-thinking skills are an integral part of programming; making a successful game also involves system-based thinking. For example, in a comparative study of Dutch middle school students in a language class learning from designing versus playing games, Vos et al. (2011) found that game designers showed significantly deeper engagement in systems analysis than students in the play only group.
How do the conceptual interconnections among systems in science, systems representations in a game, and systems thinking in the computational sense act synergistically to support student learning?