Creating learning that uses the secrets of games
Or, what learning designers can learn from Geoff Engelstein’s ‘Gametek’
Geoff Engelstein is a legend in the board game design world. He’s become most famous for turning over rocks to throw light on questions of maths, behavioural science and other underlying factors. What influences how games work. How they tick.
Some of his key insights are collected in the book, GameTek. It’s a great read, whose lessons go both ways: help with playing and designing games and help taking lessons from those things into life. I thought about writing a review of it but what I’m most interested in is how the insights relate to learning design.
So this is a not-quite review, where I pick out some of the insights that I think would be most beneficial to learning designers, especially those interested in games-based learning and gamified learning. I’ll occasionally build on them with my own additions.
Game theory suggests ideas for learning games
Engelstein talks about how game theory explains player behaviour. Most people know about the famous Prisoner’s Dilemma, where two players independently choose: to cooperate with or betray the other. It’s set up so that both will be worse off if they both betray and better off if they both cooperate but each player is tempted to betray.
Engelstein looks at how that works when the game is repeated over many rounds: players tend to cooperate more because they’re concerned about retribution. Also whether players will take account of the others’ reputation from game to game. The result is a complex web of reputation and personal advantage which could inspire many learning moments around relationships and collaboration.
The Red/Blue Game is a training classic that leverages this principle for conflict management and cooperation but you could leverage some of the ideas in the article (and the two links here) for decision-making, teamwork, communication and plenty besides.
Behavioural science also gives us ideas we could use
Many studied behavioural tendencies show up in games or game-like things. Engelstein calls out how the endowed progress effect was used to great effect by his local coffee store — a loyalty card with ten spaces (two already stamped) made him more likely to use the loyalty scheme than one with only eight unstamped spaces. Learning designers could take note in motivating learners to collect more or less anything.
He also tackles the famous Monty Hall Problem, in which most people’s intuition is wrong about whether they should switch a choice if given more information. He links it to the endowment effect and loss aversion to explore how people are motivated to stick rather than twist, especially in conditions of uncertainty. Learning designers could design around this to ‘trick’ players into revealing their own fallibility and biases in decision-making.
Framing and prospect theory also get great treatment. In many games, what is in effect the same risk-taking decision becomes more appealing if framed as ‘you’d avoid this much loss’ compared to ‘you could win/gain this much’. Take note if you want players or learners to take the risk, or not. Or frame it both ways and then show them how the framing made a difference.
Probability and maths work in strange ways, so take note
Chance is part of many games but most of us have skewed perceptions about probability and careful design around how chances work and are seen can reward designers. Players may feel that a ‘freak’ result comes up too often — for example, the ‘4’ or ‘10’ they want to avoid on two dice — but probability distribution means that in some games, what seems an unlikely result comes up almost one time in three.
Designers could do something to avoid that kind of perceived unfairness (also made famous by complaints about the videogame XCOM 2) by explaining chances better, or by reducing randomness with more dice or via other methods.
One of my favourite sections in the book explains the difference between different kinds of randomness: white noise, brown noise and pink noise. With white noise randomness, the next result is not influenced by the last. With brown noise randomness, the next result is very strongly influenced by the last: If you rolled a five, your next roll must either be a four or a six. Pink noise randomness is where your next result is influenced by the last, but sometimes more so, sometimes less so, using a distribution curve.
It turns out that pink noise randomness is almost always more interesting to players and leads to more interesting choices and more interesting games. Probably the best way to get a handle on how you might use this is to listen to Engelstein himself explain but think about how you generate randomness in your experiences, and how it might feel to players.
Specific puzzles in maths can lead to great learning game ideas
A lot of game ideas come from maths. But you don’t need to understand all the maths to use them. ‘NP-Complete’ problems, like the Travelling Salesman problem or the Three-colour Map problem, are logic problems that are easy to explain and find answers for but hard to find the best solution for. They’re also easy to ‘re-theme’: take the abstract idea and place it in whatever scenario or framing suits what you’re doing.
With the Travelling Salesman problem, players need to find the shortest route that visits each marked city on a map only once. Good solutions aren’t tough to find. But the perfect solution is much, much harder. You could re-frame this as any journey or metaphorical journey relevant to your theme, and use it as a way to get players/learners to get motivated to work together to find better solutions.
Efron Dice are a set of four different coloured six-sided dice, where (for example) the orange die beats the yellow 66% of the time, the yellow beats the green 66% of the time, and the green beats the blue 66% of the time, but the blue beats the orange dice (the first one) 66% of the time! Using these could lead to some interesting decisions, and these ideas could be used to make sure that — as with scissors-paper-rock — each piece or unit in a game is better than some and worse than others.
Work with how people want to feel
There are many more insights in the book. Many boil down to, if we know how people want to feel, then games (and learning experiences) can lean into that. If your game involves trading, people want to get the best value from a trade. This means if you want the trades to be interesting, you have to create asymmetry in value: one item needs to be worth more to some than others. Set collection, such as in Ticket to Ride, is a way to do this. So are differences in the information each team or player has access to.
Most players want to feel powerful, which is in itself a great insight. But it also turns out that one of the best ways to make them feel powerful is to let them break the rules, in certain specific ways. Ideally, ones that are individual to them. So, having basic game rules, but then exceptions based on the ability of each faction or player can make players feel great, especially when they get to use their ‘powers’. Even better if they have to work out for themselves how to do it.
And most people understand by analogy. If I call a game piece an ‘aeroplane’, it had better be able to fly over other pieces. Think about how the way you label things in the game and how you explain the rules, could be understood by analogy — relate it to things the players already understand.
