Note-taking for Research on Games and Simulations with Jan Plass
In this post I’m summarizing some writing about the foundations of research on games for learning. It’s a dry topic, so to enliven it I’ve included a bunch of screencaps from Mega Man 2. They have nothing to do with anything, but they look cool.
Plass, J.L., Homer, B.D., & Kinzer, C. (2015). Foundations of Game-based Learning. Special Issue on Game-based Learning, Educational Psychologist, 50(4), 258–283.
What is a game exactly? One definition: “a system in which players engage in an artificial conflict, defined by rules, that results in a quantifiable outcome” (Salen & Zimmerman 2004, 80). Gamification is the grafting of points and stars onto existing tasks, like completing your boring homework. By contrast, game-based learning is more like Logical Journey of the Zoombinis – organically placing learning activities into a conflict structure to make them interesting and engaging.
The essential activity in games is play, i.e., the method by which mammals learn. Play activates kids’ schemas in ways that let them transcend their immediate reality. Pretend play enables kids to practice holding multiple representations of the same object in mind, which builds toward symbolic thinking, theory of mind, and literacy and numeracy.
Good games induce a state of flow by keeping the player within the zone of proximal development per Vygotsky. They allow for graceful failure and foster self-regulated learning as players set goals, monitor their own achievement, and assess the effectiveness of their strategies.
All games include a challenge, a response, and feedback, forming a loop. Behaviorist games give players a limited menu of possible actions, and give feedback in the form of binary right and wrong, success/failure (e.g. Super Mario Bros, Tetris). Constructivist games let players set their own challenges and find their own responses, and also create peer feedback (e.g. World of Goo, tower defense games).
There is no single model of game-based learning, because that would require a model of learning generally. Games for learning can have several possible goals: preparation of future learning; directly teaching new knowledge and skills; practicing and reinforcing existing knowledge and skills; or “developing 21st-century skills,” that is, socioemotional skills related to teamwork, collaboration, problem solving, creativity, communication, and so on.
Games provide opportunities for situated learning, providing information at the moment when it will be the most useful to the learner. Also, games can simulate real life, and can thus facilitate transfer of learning. The tutorials in entertainment games use scaffolding, progressively fading out supports as the learner progresses.
A well-tuned challenge can be intrinsically motivating; otherwise why would we want to spend so much time on an activity as flamboyantly pointless as Tetris? But while games are great for motivation, you want to be careful that the motivation to play actually translates into motivation to learn the subject matter. Otherwise kids can “game the system” (amazing term!), entertaining themselves without learning anything. Of course, this can be true of any aspect of school.
What is the difference between playful learning and games? Playful learning is learning by doing a game-like task, as opposed to gamification, which is adding game elements to an non-playful task.
Mayer, R.E., & Johnson, C.I. (2010). Adding Instructional Features that Promote Learning in a Game-Like Environment. Journal of Educational Computing Research, 42, 241-265.
The authors compare value-added research, cognitive consequences research, and media comparison research. Compare to ideation research, e.g. IDEO, and playtesting. The game involves progressively harder challenges which the player performs with a minimum of guidance in a rule-based environment that responds to the player in a cumulative way–the game’s present state depends on the player’s past actions.
The authors evaluate their game using a measure of learning outcome—in this case, performance on an embedded transfer test. Their research is based on the cognitive theory of multimedia learning. By this theory, learners have separate channels for processing verbal and pictorial information, and each channel has a limited capacity. Players learn meaningfully when they engage in cognitive processing, which includes attending to relevant information and ignoring everything else; organizing information into a coherent representation; and relating the information to prior knowledge.
The authors found that players learned faster and were better able to transfer their knowledge to new problems if they had to reflect on their own thinking during the game, by having to give the reasons for their actions (self-explanation) or being shown why a certain action was the correct one (feedback).
Isbister, K., & Schaffer, N. (2008). Game Usability. New York: Morgan Kaufman. Chapters 3, 4
User testing for usability is more of a qualitative methodology, like ethnography, so it can be performed on small groups or individuals. Designers can observe and discuss during testing, but they need to try not to coach the subjects, because of course, the designers don’t get shipped with the game.
People buy and play games for entertainment voluntarily, so it matters how they feel when they’re using them, unlike productivity software. Though don’t feelings matter there too? How productive are you going to be if you’re enraged or demoralized? Anyway, game devs have a strong incentive to pay attention to user experience, since that’s the only thing that the customers are paying for. Games are just as hit-driven as movies or music; eighty percent of them lose money.
Richardson, P., & Kim, Y. (2011). Beyond Fun and Games: A Framework for Quantifying Music Skill Developments from Video Game Play. Journal of New Music Research, 40(4), 277–291.
This study of learning outcomes from music games used pre- and post-assessments of players’ musical skills using both quantitative tests and qualitative surveys. These included the Musical Aptitude Profile (MAP), basically a musical SAT. It’s a comparative music-listening test addressing mental representation of musical forms, taken as a form-fill listening test. It attempts to assess students’ ability to form internal mental representation of musical phrases and features without assuming any music theory terminology or formal training. That’s good. But how rigorously empirical can this assessment really be?
The MAP exercises are based on qualitative matching/fitting of musical prompts, and they are probably debatable. My experience of standardized testing on humanities subjects is that they present subjective questions as if they were unambiguous matters of fact. This leads to absurdities like a poet being unable to answer test questions on her own work. I would take MAP results with a giant grain of salt.