Game-based Learning Design Optimized for Cognitive Load
The enjoyment factor of video game play makes game-based learning an attractive option to motivate and engage learners of all ages. Game play can provide real world situations to the learning experience, as well as create an opportunity to practice and learn by trial and error. Improvements in technology make creating a game-based learning easy, identifying the need to look closer at how design can influence how information is processed. In general, using technology in education can overload the working memory, but applying Sweller’s (2020) principles for optimal design in educational technology these challenges can be minimized. In the case of game-based learning the mechanics can be applied to minimize the load. The current body of research in regard to game-based learning design and cognitive load is small and offers little insight to address optimal design. It is the position of this paper that game-based learning design and mechanics can address Sweller’s (2020) principles for optimal design in educational technology. Therefore, game-based learning is a viable resource to promote knowledge development.
Game-based learning, cognitive load, game-based learning design, game mechanics.
Technology advancements change the way we live, work and experience entertainment. This technology that has infiltrated our daily lives should and does change the way that we learn (Shaffer et al., 2005). Game-based learning exemplifies the influence current technology can have on learning. Personal computer processing power, affordable highspeed internet, and an array of affordable development tools offers the ability to marry the enjoyment of games with pedagogical outcomes. Creating digital games for learning no longer requires expensive outsourcing, now it can be as easy as dragging and dropping audio, visuals, animations, and interactive elements. This brings up the need to take a closer look at how game mechanics and design can influence the learning experience. More specifically how information is processed during game play. Cohesive integration of game components and learning outcomes has the potential to foster learning, however poor design can increase cognitive load and do more harm than good (Huang, 2010; Lee et al., 2020; Qian & Clark, 2016).
In general, learning with technology can strain the working memory (Sweller, 2020). Just like other digital mediums the visual, audio and interactions in game play might succeed in motivation, however it may also overload the working memory, which impedes learning (Huang, 2010). Currently the small body of research lacks in showing how game-based learning affects cognitive load. Further study is needed to understand the impact and identify efficient design techniques (Chang et al., 2017; Huang, 2010). Even with this gap in research it is proposed that game-based learning can minimize cognitive load. The logic behind this is that game design aligns with Sweller’s (2020) principles for optimal design in educational technology. These principles included worked examples, split-attention, modality, transient, redundancy, expertise reversal and element interactivity, and working memory depletion (Sweller, 2020).
This paper will first look at game-based learning, then investigate the cognitive load theory, and finally discuss how game mechanics and design align with the principles outlined by Sweller (2020).
Game-based learning is the combination of game play and learning theory to enhance the overall learning experience (Roodt and Ryklief, 2019). More specifically, educational computer games integrate the interactive components, challenging activities, and clear goals of game play with pedagogical outcomes (Chang et al., 2017). Mundane, challenging or even well received educational content can leverage the engaging storyline and competitive activities that make video games so popular. Digital game-based learning may include a two-or three-dimensional space that will engage learners and motivate a high level of interaction (Chang et al., 2017). Game mechanics can be adapted to teach specific subjects (Chang et al., 2017), as well as provide real world experiences (Schrader & Bastiaens, 2011).
Activities experienced in a game environment can influence knowledge acquisition, skill development, social interaction, collaborative thinking and encourage problem solving (Qian & Clark, 2016; Shaffer et al., 2005; Yang, 2012). Providing a way to apply the demonstrated benefits of video games to learning, like improved working memory (Lee et al., 2020; Mitre-Hernandez et al., 2021). In addition, game-based learning can improve flow, or the ability to concentrate without disruption from outside distractions (Chang et al., 2017).
When information is taken in it is processed in the working memory and transfers to long term memory for storage. Over time this information becomes part of a large repository of domain specific information, which is how expert skills are developed (Sweller, 2020). Although there are capacity limits on novel information, there are not capacity limits on domain specific information, since it is processed differently (Sweller, 2020). When this information is needed it transfers back to the working memory for use (Sweller, 2020). The working memory is sensitive to complex information which influences the ability to transfer and achieve that long-term storage (Sweller, 2016). Cognitive load can happen when the amount of information taken in exceeds an individual’s working memory capacity (Chang et al., 2017; Huang, 2010).
Cognitive load research is still evolving, progressing from generalizations to more specific categorizations of intrinsic, extraneous, and germane. Intrinsic load refers to demands placed when completing a task and could vary based on the actual task and the learners experience level with that task (Lee et al., 2020). Extraneous load comes from the physical learning environment, or distractions in the environment that split attention between learning materials and an unrelated source (Lee et al., 2020). In addition, unnecessary elements in learning materials are distracting and increase the extraneous cognitive load (Chang et al, 2017). Germane load is the actual mental effort exerted in trying to learn (Lee et al., 2020).
The current body of research does not fully support game-based learning as a desirable option for cognitive load. Typically, environments that promote exploration, interaction and manipulation have a higher cognitive load (Alexiou & Schippers, 2018). Elements like three-dimensional animation influence user engagement, but also increase cognitive load (Chang et al, 2017). The dynamic nature of game mechanics might succeed in accomplishing engagement, it may also be overloading the working memory, which impedes learning (Huang, 2010). For example, the intense focus required to succeed in serious games for learning or even multiplayer online role games expose learners to a higher level of cognitive demand (Huang, 2010). This level of intensity and complexity of game play has shown to increase cognitive load (Alexiou & Schippers, 2018). Since these storylines aid in an emotional engagement between characters and learners (Alexiou & Schippers, 2018). Heightened emotion can negatively influence extraneous load, with dynamic interactions and simulations causing higher intrinsic load (Lee et al. 2020).
Schrader & Bastiaens, (2012) compared a highly immersive educational game to a less immersive environment using narrated and animated on screen text. The level of immersion in delivery was used to create varying levels of virtual presence for the eighth-grade participants. It was found that higher immersion created more strain on the working memory. One point discussed was the additional cognitive demand required to learn how to work the game. In contradiction Chang et al (2017) compared the level of flow between game-based learning and materials accessed on a website. Flow is achieved when an experience is enjoyable enough to hold a learners attention, even with distractions. Game-based learning showed to weakly enhance extraneous cognitive load, but largely enhance germane load, with no significant difference for intrinsic load. One reason for this difference may be related to categorizing findings. Chang et al (2017) reported load based on categorization, however Schrader & Bastiaens, (2012) believe all categories should be addressed to support optimal cognitive load.
Research in game-based learning finds some agreement on the effect design has on cognitive load. Design of educational games must consider limitations in cognitive processing to reduce load and promote transfer (Huang, 2010). Design consideration should be given to environment and complexity of tasks to reduce cognitive demands (Schrader & Bastiaens, 2012). Game mechanics and learning alignment is necessary to reduce load and facilitate learning (Kalmpourtzis & Romero, 2020). Game mechanics refers to what the game can do, and learning objectives are what we want learners to be able to do after completion (Kalmpourtzis & Romero, 2020).
Principles for optimizing cognitive load
Optimal design of educational games can be fostered by addressing principles of the cognitive load theory. Huang (2010) surmised this theory as a construct that includes mental load based on tasks, learner performance and the effort exerted in the working memory to process information. The cognitive load theory is a framework for interactions between instructional design, learning processes and cognitive load (Lee et al., 2020). It is concerned with the presentation of information and can be applied to instruction to improve that presentation (Sweller, 2020). This can be used as a resource to provide instructional design principles to computer-based learning to achieve desired outcomes (Chen at al., 2017).
Based on the constructs of the cognitive load theory Sweller (2020) identified principles that can be applied to educational technology to minimize load. These principles include worked examples, split-attention, modality, transient, redundancy, expertise reversal and element interactivity, and working memory depletion. In some instances, by nature, game-based learning aligns with Sweller’s (2020) principles, however, thoughtful design can address these principles to lower load and promote learning.
Problem solving requires simultaneous processing of multiple elements, which spends a high amount of working memory load, however, prior knowledge requires less processing effort (Sweller, 2020). Worked examples help novice learners’ problem solve as someone with prior knowledge, which is with reduced working memory load (Sweller, 2020).
Classroom worked examples were exemplified by ter Vrugte et al. (2017) who studied the use of fading. The game and content was identical with the only difference of one group having faded worked examples in game challenges and the other group having none. Fading was described as displaying a full worked example, then partial worked example, with less of the example showing until it was gone (ter Vrugte et al., 2017). This method allowed learners to adapt to problems before progressing to more difficult problems. Findings showed faded worked examples integrated into game play improved proportional reasoning skills, as well as accuracy. Indicating students that had worked examples were more able to apply knowledge, as well as identify the appropriate solution for solving problems (ter Vrugte et al., 2017).
A broader context of worked examples can be experienced with simulations. Novice learners can interact in a real-world scenario to address training for situations that might otherwise be left to a textbook. Students gain hands-on virtual experience (Shaffer et al. 2005) and practice, which assist with long term retention (Yang, 2012). This is exemplified by Tsai et al. (2015) who developed a training simulation for disaster prevention. It was found that current disaster training incorporated more traditional theoretical educational methods, which lacked actual skill development for disaster preparedness. Learning goals were combined with game mechanics to provide opportunities to practice and explore techniques in a safe environment. Interaction with the game evolved tangible skills for strategy and manipulation of resources (Tsai et al., 2015). The game environment provided opportunities for trial and error, where students can grow from mistakes without fear of loss (Yang, 2012).
Split attention is separating multiple sources of information, that have a relationship. This requires the learner to take on the extra task of integrating information. For example, when presenting a diagram, text or callouts should be incorporated with the diagram, instead of positioned separately (Sweller, 2020). Multiple sources of information should be smoothly integrated and occur simultaneously (Sweller, 2020).
Game mechanics add a context that unifies visual, audio and tactile elements. Organically incorporating communicative elements into the game will create a cohesive flow (Van Eck, 2010). For example, visual elements like status bars serve as a guide through the experience within the context of the theme (Van Eck, 2010). The learner is not forced to lose focus on the learning information; however, extra support is integrated into the environment if needed.
Components like narrative is depicted in characters, settings and game elements, this is combined with audio to tell a story that unifies learning elements and game interactions (Van Eck, 2010). The game narrative adds that real life context to game play, creating a simulation to facilitate knowledge development for use outside of the game (Alexiou & Schippers, 2018). This can reinforce skills through game interaction, and game environments can provide contexts beyond pedagogy to further aid processing (Alexiou & Schippers, 2018). A single player design tends to involve the learner in the narrative, whereas a multi-player design allows game play to dictate the narrative (Van Eck, 2010). Learning activities should be problem-based and applied within the context of the narrative (Van Eck, 2010).
Modality effect, like split attention, addresses the presentation of complimentary elements. In technology it is not always possible to visually create relationships with information. Delivering complimentary information in different modes, like visuals on a screen that are coordinated with audio, can create that relationship (Sweller, 2020). The working memory increases efficiency when dual modularity is applied (Sweller, 2020).
Game-based learning uses a higher modality than traditional learning and even other technologies, by incorporating the multimodal mechanics of video games. Texts are delivered through a combination of graphics, animations, sounds, and written words (Lee & Ke, 2019). Game play promotes information to transform into learning by the way these multiple intake methods are incorporated (Plass et al., 2015). This experience is complimented with interactive challenges to reinforce concepts. The ability to interact with a dynamically changing environment fosters a relationship between game mechanics and learning information.
Challenges directly incorporate multimodal auditory and visual cues with an opportunity to engage in practice, problem-solve and apply information. They trigger curiosity, which leads to exploration, which leads to learning (Mosiane & Brown, 2020). They can be thought of as competitions played individually or against other learners. Design can be as simple as offering a limited selection of choices, coordinated with corrective feedback (Plass et al., 2015). Ranging to a more complex design where decisions made in game play determine the level of difficulty (Plass et al., 2015). In aligning with pedagogy, challenges can be directly mapped to learning outcomes and used as a method to measure achievement.
Redundancy refers to unnecessary information, or clutter which includes repeating information, and busy visuals. This can be overwhelming and take more effort to process (Sweller, 2020). Redundant information will interfere with learning by becoming more important than non-redundant information (Sweller, 2020). Reducing complexity will reduce working memory load and promote acquisition of new knowledge (Sweller, 2020).
Multiple components of the game environment can add complexity and require simultaneous processing with key learning elements (Schrader & Bastiaens, 2012). Game play offers opportunities to reduce the need for repeating information. Learning components and related activities can fluctuate between structured activities when necessary to unstructured allowing opportunities to experiment and make discoveries (Alexiou & Schippers, 2018).
Redundancy can be reduced with timely and relative feedback (ter Vrugte et al, 2017; Mosiane & Brown, 2020). Feedback can be used to reduce uncertainty, assist in adjusting strategies, identifying status and progress made, and maintain alignment with learning objectives (Alexiou & Schippers, 2018), and track progress in reaching goals (Mosiane & Brown, 2020). It helps learners identify mistakes, recognize correct action and guide the journey to foster knowledge transfer (Goldberg & Cannon-Bowers, 2015).
This can be built in by incorporating feedback with clues or tips, when triggered by the learner. Game interaction will gradually adapt the experience to knowledge and eliminate redundancy (ter Vrugte et al, 2017). It is important to note that external feedback can increase cognitive load, so it is recommended to incorporate into the environment (Alexiou & Schippers, 2018).
Holding large chunks of information for long periods of time, while trying to assimilate with permanent information is difficult for the working memory to perform (Sweller, 2020). The goal is to convert transient information into permanent information. Sweller (2020) suggests transient information can be addressed by eliminating lengthy chunks of information, this will make it easier to process current and previous relationships.
Aligning content, activities and overall game mechanics will automatically initiate some level of chunking (Van Eck, 2010). Instead of a specific mechanic, chunking can be simulated with the way game mechanics are used together and integrated with learning information. As previously discussed in worked problems, the level of fade applied to worked problems is based on game progress (ter Vrugte et al. (2017). In this case chunking is applied based on performance or skill level. Leveling is discussed in more detail under expertise reversal and element interactivity. This can address transient information by designating that start and end point within the game as it associates with learning content chunking (Van Eck, 2010). Information can be broken out into logical start and end points, which may include a book chapter, learning module, or lesson.
Information chunking within each level should relate new and prior knowledge, which can be accomplished through the use of challenges. In addition to chunking information further, challenges present a break from processing new information and a way to practice what was presented. This offers a way to reinforce new information by constructing knowledge in a realistic context (Alexiou & Schippers, 2018). This not only benefits the construction of new knowledge but can also be used to enhance existing skills (Alexiou & Schippers, 2018). The challenge is followed by immediate feedback based on performance. This would end the cycle for this chunk of information, and promote next steps based on the type of feedback provided.
Expertise reversal and element interactivity:
Expert learners require less information for processing, presenting information to the granularity a novice might require is less effective and could do more harm than good (Sweller, 2020). This requires a balance based on the skill level of the intended audience. An easy task may bore experienced learners, and a difficult task my frustrate novice learners. The level of difficulty should align with the learner (Chang et al., 2017).
Game mechanics should include some form of leveling, which is triggered by the skill required by a player to progress in the game (Mitre-Hernandez et al., 2021). Activities in game play gradually incorporate more complex tasks which will accommodate various experience levels and promote scaffolding (Alexiou & Schippers, 2018). Scaffolding occurs when a learner is guided to achieve a task that was previously beyond their capability (Plass et al., 2015). This approach will build knowledge as an understanding is formed during interaction with the game environment (Alexiou & Schippers, 2018). Information can be presented, mastered and then fade into more challenging concepts (Plass et al., 2015). For example, components such as new player tutorials are present when needed but fade to allow players to focus on game activities. Game mechanics can be designed to assess one level to determine mastery then identify the next level and challenges that the learner should encounter (Plass et al., 2015).
Mitre-Hernandez et al., (2021) discussed two approaches to leveling. Manual leveling would allow players to determine difficulty. This would allow learners to personalize the difficulty of the game based on their skills and experience (Alexiou & Schippers, 2018). This can be approached similar to video games, where interaction starts at a comfortable level and gradually becomes more challenging (Mitre-Hernandez et al., 2021). Manual leveling presents a solution to addressing expert learners, however, there are also some challenges. For example, learners may not know which level they belong in, which requires playing a level (Mitre-Hernandez et al., 2021). Offering leveling guidance can help learners quickly adjust and mitigate any potential losses.
Another option is automated or adaptive leveling, which allows the game to dynamically adjust based on learner interaction (Mitre-Hernandez et al., 2021). There is indication this approach is more successful in achieving learning outcomes then manual leveling (Mitre-Hernandez et al., 2021). Mitre-Hernandez et al., (2021) used eye tracking to study pupil dilation to determine how difficult or easy a task was during game play. This is based on how the pupils become dilated based on level of mental effort. When a lower effort was experienced, the game dynamically adjusted to more challenging concepts. Significance was found in the diameter of the pupil and level of difficulty. Game performance aligned with findings, meaning leveling slowed when players perceived something as difficult. Highlighting an opportunity to use player performance to determine success or failure at a level and identify when it is necessary to progress in the game (Mitre-Hernandez et al., 2021). An automated design can also include positive and negative feedback loops to adjust leveling accordingly (Alexiou & Schippers, 2018). This progression through levels provides immediate feedback which can guide the learner, maintain engagement, and promote scaffolding (Chang et al., 2017).
Working memory depletion:
The working memory becomes depleted in long interactions with new information and needs to recover with resting periods (Sweller, 2020). Working memory capacity is greater when information is spaced out (Sweller, 2020). Similar to chunking for transient information, game design can incorporate appropriate spacing of information, accompanied by opportunities to practice. Knowledge can transfer through repeated practice of a skill or higher-level application like problem solving (Plass et al., 2015).
Previous research identified segmenting, sequencing and prior training as a counter to the decline experienced from higher load (Lee et al., 2020). Exemplified by Lee et al., (2020) who integrating a pausing effect into the game. This effect simulates chunking to reduce a flood of challenging information. Pauses can be used to relax the working memory, reflect on information or give extra time for processing. Lee et al., (2020) studied this in a medical simulation game, where critical patient care was required, with the experimental group being given opportunities to pause. Findings indicated cognitive load was reduced during game play pauses, with significant improvement found in overall performance. Since pauses were voluntary the success of this technique requires self-regulation from the learner, however this technique can be thoughtfully integrated into the game design for planned breaks or lighter activities dispersed throughout exhaustive concepts.
This paper broke out game mechanics to exemplify how the design can minimize cognitive load, with some topics like split attention and transient information giving some thought into how mechanics can work together. Even more consideration should be given to how components relate to each other. The design should make sense and have purpose. The use of mechanics for the sake of using can contribute to clutter and do more harm than good. A successful learning game does not need to include everything, but the most appropriate elements to achieve desired learning goals.
The type of learning content was not discussed here. There is still some debate as to what content is best suited for a game-based learning intervention. There is some belief that educational games are not effective with traditional instruction, like math (Lee & Ke, 2019). Games create iconic relationships, which aid in adding context and skill development. Lee & Ke (2019) points out that although iconic relationships might not teach the actual math calculation, it will teach how to determine which calculation should be used to problem-solve a situation. The ability to analyze a situation and determine the best course of action is a higher-level thinking task, and the ultimate goal of teaching the calculation. In this context the use of game-based learning depends on desired learning outcomes.
There is still a great deal of research needed on how game-based learning influences cognitive load. This relates to showing significance in a game intervention as well as how specific mechanics can influence processing. Researchers do agree that design matters, yet research lacks how to design game-based learning to influence cognitive load (Chang et al., 2017; Huang, 2010). Educational games are met with skepticism by scholars because of a lack of theoretical understanding in how game design, mechanics and pedagogical integration can influence knowledge development (Alexiou & Schippers, 2018).
This paper attempts to offer some basis by aligning game-based learning design and mechanics with Sweller’s (2020) principles for optimizing load in educational technology. Research would benefit from studying Sweller’s (2020) principles in a game-based learning experience to add to the body work and establish design standards. Sound design decisions, alignment with learning outcomes, and fully integrated mechanics can contribute to an impactful learning experience. While taking advantage of the enjoyment, motivation and engagement that game-based learning provides.
Alexiou, A. & Schippers, M. C. (2018) Digital game elements, user experience and learning: A conceptual framework. Educ Inf Technol, 23, 2545–2567.
Chang, C. C., Liang, C., Chou, P. N. & Lin, G. Y. (2017) Is game-based learning better in flow experience and various types of cognitive load than non-game-based learning? Perspective from multimedia and media richness. Computers in Human Behavior, 71, 218-227. https://doi.org/10.1016/j.chb.2017.01.031.
Chang, C. C., Warden, C. A., Liang, C. & Lin, G. Y. (2018) Effects of digital game-based learning on achievement, flow and overall cognitive load. Australasian Journal of Educational Technology, 34(4).
Chen, O., Woolcott, G. & Sweller, J., (2017) Using cognitive load theory to structure computer-based learning including MOOCs. Journal of Computer Assisted Learning, 33, 293–305, doi: 10.1111/jcal.12188.
Dickey, M. D. (2007) Game design and learning: a conjectural analysis of how massively multiple online role-playing games (MMORPGs) foster intrinsic motivation. Education Tech Research Dev, 55, 253-273.
Erhel, S. & Jamet, E. (2013) Digital game-based learning: Impact of instructions and feedback on motivation and learning effectiveness. Computers & Education, 67, 156-167.
Goldberg, B. & Cannon-Bowers, J. (2015) Feedback source modality effects on training outcomes in a serious game: Pedagogical agents make a difference. Computers in Human Behavior, 52, 1–11
Greipl, S., Klein, E., Lindstedt, A., Kiili, K., Moeller, K., Karnath, H. O., Bahnmueller, J. Bloechle, J. & Ninaus, M. (2021) When the brain comes into play: Neurofunctional correlates of emotions and reward in game-based learning. Computers in Human Behavior, 125. https://doi.org/10.1016/j.chb.2021.106946
Huang, W. H. (2011) Evaluating learners’ motivational and cognitive processing in an online game-based learning environment. Computers in Human Behavior, 27(2), 694-704. https://doi.org/10.1016/j.chb.2010.07.021
Kalmpourtzis, G. & Romero, M. (2020) Constructive alignment of learning mechanics and game mechanics in Serious Game design in higher education. International Journal of Serious Games, 7(4), 75-88. http://dx.doi.org/10.17083/ijsg.v7i4.361
Lee, J. Y., Donkers, J., Jarodzka, H., Sellenraad, G. & van Merriënboer, J. J. (2020) Different effects of pausing on cognitive load in a medical simulation game. Computers in Human Behavior, 110. https://doi.org/10.1016/j.chb.2020.106385.
Lee, S. & Ke, F.(2019) The format of problem representation for in‐game learning supports. Journal of Computer Assisted Learning, 35, 390-406.
Liao, C. W., Chen, C. H. & Shih, S. J. (2019) The interactivity of video and collaboration for learning achievement, intrinsic motivation, cognitive load, and behavior patterns in a digital game-based learning environment. Computers & Education, 133, 43-55. https://doi.org/10.1016/j.compedu.2019.01.013
Mitre-Hernandez, H., Carrillo, R. C. & Lara-Alvarez, C. (2021) Pupillary responses for cognitive load measurement to classify difficulty levels in an educational video game: Empirical study. JMIR Serious Games, 9(1).
Mosiane, S., and Brown, I., 2020. Factors influencing online game-based learning effectiveness. The Electronic Journal of Information Systems Evaluation, 23(1), pp. 79-95.
Plass, J. L., Homer, B. D. & Kinzer, C. K. (2015) Foundations of game-based learning. Educational Psychologist, 50(4), 258-283
Qian, M. &, Clark K. R. (2016) Game-based learning and 21st century skills: A review of recent research. Computers in Human Behavior, 63, 50-58. https://doi.org/10.1016/j.chb.2016.05.023.
Roodt, S., & Ryklief, Y. (2019). Using digital game-based learning to improve the academic efficiency of vocational education students. International Journal of Game-Based Learning, 9(4), 45-69.
Schrader, C., & Bastiaens, T. J. (2012). The influence of virtual presence: Effects on experienced cognitive load and learning outcomes in educational computer games. Computers in Human Behavior, 28(2), 648e658.
Shaffer, D. W., Squire, K. R., Halverson, R., & Gee, J. P. (2005). Video games and the future of learning. Phi Delta Kappan, 87(2), 105–111. https://doi-org.libproxy.library.unt.edu/10.1177/003172170508700205.
Sweller, J. (2020) Cognitive load theory and educational technology. Education Tech Research Dev, 68, 1-16. https://doi.org/10.1007/s11423-019-09701-3.
Sweller, J. (2016) Cognitive load theory: What we learn and how we learn. Learning, Design, and Technology. DOI 10.1007/978-3-319-17727-4_50-1
ter Vrugte, J., de Jong, T., Vandercruysse, S., Wouters, P., van Oostendorp, H. & Elen, J. (2017) Computer game-based mathematics education: Embedded faded worked examples facilitate knowledge acquisition. Learning and Instruction 50, 44-53.
Tsai, M., Wen, M., Chang, Y. •& Kang, S. (2015) Game-based education for disaster prevention. AI & Soc, 30, 463–475.
Van Eck, R., (2010) Gaming and cognition: Theories and practice from the Learning Sciences. Information Science Reference.
Yang, Y. C. (2012) Building virtual cities, inspiring intelligent citizens: Digital games for developing students’ problem solving and learning motivation. Computers & Education, 59(2), 365-377. ISSN 0360-1315, https://doi.org/10.1016/j.compedu.2012.01.012.