Gamasutra: The Art & Business of Making Gamesspacer
Persuasive Games: Shell Games
View All     RSS
November 1, 2014
arrowPress Releases
November 1, 2014
PR Newswire
View All

If you enjoy reading this site, you might also want to check out these UBM Tech sites:

Persuasive Games: Shell Games

March 3, 2010 Article Start Page 1 of 3 Next

[Just what will the achievementization of the world mean? Author and game designer Ian Bogost ponders Jesse Schell's DICE talk and blends his interpretation with research. Will it work... and, more importantly, is it good?]

In a widely disseminated talk at DICE last month, Carnegie Mellon University Entertainment Technology Center professor Jesse Schell made a provocation: can game-like external rewards make people lead better lives?

To answer the question, Schell explored hypothetical scenarios that might combine awards of Xbox Achievements-like scrip with emerging sensor networks that would track our everyday behaviors. Teeth brushing might earn sponsored awards from Crest, for example, and taking the bus might earn awards from a government mass transit program.

Sounds farfetched? Schell points out early signals that support his vision, including the dashboard of the new Ford Fusion, which features an image of a plant that flourishes or droops based on how efficiently the driver pilots the vehicle.

Reactions to the talk have been mixed, with some heralding him as a brilliant oracle of a desirable future, others wondering how he could have failed to mention related, high-profile work by Frank Lantz (Area/Code) and Jane McGonigal (Institute for the Future), and others dismissing Schell's prophesy as dystopian nightmare.

Here, I want instead to explore a few philosophical problems that arise from Schell's position -- problems which proponents and detractors should both consider.

S(c)hell Games

In addition to his professorship at CMU, Jesse Schell operates a game studio. They make electronic and location-based entertainment, under the shingle "Schell Games." It's a clever name, because it plays its founder's surname off a well-known gambling game, the shell game.

Everyone knows it: a small ball is hidden under one of three walnut shells. The operator shuffles the shells around quickly, and a player is made to guess which shell houses the ball. Wagers are usually made, and indeed a typical run may appear to draw in a great many bets around a makeshift table on a street corner.

But the shell game is not a game of chance at all. It's a confidence trick. A skilled operator can remove and replace the pea at will, insuring that the player only wins if the operator chooses (which he might do in order to encourage additional bets). Most of the audience is in on the swindle, and their bets serve to encourage a target to place (and then ratchet up) the wagers. The shell game is not a game at all. It's a fraud, a swindle, a con.

With apologies to Mr. Schell's good name and studio, we might give the title schell games to video game incentive tricks that cause people to "do the right thing," such as brush their teeth or drive their cars efficiently. (I'll spell it in the lower case to distinguish the type of game from Jesse Schell's company.)

Just as the shell game dupes players into believing that they are making wagers against chance, so the schell game dupes players into believing that they are completing virtuous actions toward righteous positions.

When seen from the perspective of their outcome alone, a wager in a shell game looks like an earnest gamble. But when seen from the perspective of the operator, who is altering the game to suit his desired outcomes, the wager is merely a foil to be used against the player.

Likewise, when outcomes alone are considered, an action taken in a schell game looks like an earnest honor. But when seen from the vantage point of the agents who dole out incentives, that honor again becomes merely a foil to motivate an action.

Article Start Page 1 of 3 Next

Related Jobs

Rensselaer Polytechnic Institute
Rensselaer Polytechnic Institute — Troy, New York, United States

Assistant Professor in Music and Media
The College of New Jersey
The College of New Jersey — Ewing, New Jersey, United States

Assistant Professor - Interactive Multi Media - Tenure Track
Next Games
Next Games — Helsinki, Finland

Senior Level Designer
Magic Leap, Inc.
Magic Leap, Inc. — Wellington, New Zealand

Level Designer


Glenn Storm
profile image
Since the common thread of this discussion is not just human behavior, but cultural behavior, it seems less and less concrete the more derivative the discussion becomes.

If the simple novelty of some of the games proposed in Schell's talk wears off, and we begin to look for 'the next thing', as we often do as a culture; the original point being debated would become moot. I was certainly imagining the backlash from cool teenagers while I listened to the talk. ("My parents are slaves to the game. I'm going to ditch my rewards account.") For me, the only reasonably reliable prediction we can make about cultural behavior is this: if we like it at the time, we'll keep it and if we don't we won't.

I have enjoyed following this high-minded discussion about the power and future of games, I just wish it was more grounded, even if that means it becomes less provocative.

Ian Bogost
profile image
Glenn, it's true that there's a good measure of speculation in this area, but it seems to me that there's plenty of ground already on both sides; it's just not the kind of straightforward *design* or *business* ground we sometimes find. Instead, it's a mushy, swampy ground, where all sorts of worldly messiness muddles.

For a fairly definitive counterposition on rewards enhancing behavior, take a look at the 1973 experiment Jesper Juul recently discussed (, which shows that external rewards can be demotivating.

Eric Hardman
profile image
Ultimately, most people aren't going to think too deeply about why they do things: if an achievement gets kids to brush their teeth, then fine, use it until it stops working. Research suggests that motivations people really care about, supposedly, like money and grades, actually prove to be demotivating. So maybe trivial rewards like an achievement will do a better job. Getting kids to *want* to brush their teeth, though, good luck with that. That level of persuasive power is in the realm of the gods alone, games don't even come close.

Glenn Storm
profile image
I appreciate the response and the link, Ian. Gracefully overlooking the irony; that my subtle gripe about over speculation flies in the face of my own theoretical blog posts, is also appreciated. :)

I agree this is mushy stuff.

While I have not read the text of the study Juul discusses, I have a feeling the experiment may actually prove the diminishing returns of static (or consistent) rewarding behavior; something I hint at in my previous comment. I am not a psychologist, but I am a parent. If my daughter receives the same reward, praise or attention in the same way, it almost immediately begins to have a different effect. One thing I do know about psychology: it is not math; it is not absolute.

In the games industry, the new flavor of the month enjoys a novelty boost, in sales and in general interest. It's pretty easy for us to see how games can have this reward influence Schell suggests, but I think this is primarily due to the current climate of turnover of the latest hot game. When we talk about a longer-term, pervasive, comprehensive and consistent social rewards game, I can't help but wonder the cultural lifespan of such a mechanism.

At any rate, just ignore me if the subject is sparking discussion. I will listen regardless, who am I kidding?

Ian Bogost
profile image
Indeed, I share your skepticism about purely psychological or cog-sci answers to the world. Or indeed, perhaps, about definitive answers of any kind!

As for the new flavor of the month, you're right that there's always some new trend. But perhaps the one point I really agree with in Schell's talk is that so-called social games took people by surprise. I think that's generally true, so it would do us well to think and look further out than the next quarter.

Carlo Delallana
profile image
My parents played a shell game on me when I was a kid, i got cool, stuff for getting good grades. At the time it made sense because as a child I couldn't see the long term benefits of a good education so short term and less abstract rewards were needed.

Christian Nutt
profile image
@Eric, Well, that's true -- many people won't think hard about it. That's why it's up to the designers of the systems to think about the true implications, rather than just rushing to make money off of people by using them. Which I suspect is not happening...

@Carlo, Even something like that has its flaws because kids are good at finding the exceptions and loopholes in systems -- maybe better than adults in some ways.

Ted Brown
profile image
I'm sorry, I have to disagree with a piece of the argument. Maybe it's a sliver, or maybe it's a linchpin: the examples are flying fast and free in this article! =)

It's this quote: "When people act because incentives compel them toward particular choices, they cannot be said to be making choices at all." Perhaps I misunderstand, but what of choosing between competing incentives? What if you have 2 minutes, and you only move fast enough to Brush or Floss? Do you measure the rewards? Do you analyze the health benefits? Do you find a way to make more time in your morning?

This is not at all an empty, barren realm Mr. Schell is describing. Rather, it could be a way to help visualize, prioritize, and track progress on things humans simply don't manage well: long term stuff like health, finances, habits, etc.

Nick Fortugno
profile image
What I never understand about this funware/achievements-for-everything argument is why people don't think that users will acclimate to the new environment. I mean, right now I am constantly bombarded with free offers and spam and targeted advertising that I largely choose to ignore. Most people have learned these tricks and self-select what they consume. This is not to say that advertising is ineffective, but rather that we are not swayed by all of the advertising we see, even though we are inundated with it.

I don't know why anything would change when you start calling them achievements and points. I'll remind everyone that most hardcore gamers don't FINISH all of the games they start, let alone come anywhere near getting all of the achievements.

What I think would happen in Schell's future is that people would be become so accustomed to getting points that they would devalue them, except for the small subset of points that actually matter to them. Identity would be in part defined by which points systems you choose to pursue. And I will put money down that the tooth brushing progress meter will not be high on the value chain of most 8-year-olds.

Alexander Jhin
profile image
What Schell is proposing is basically a "gamified" version of feedback. Psychologists have known forever that providing people with measurable statistics for everyday actions can change their behavior in everything from heart rate (yes, you can control your heart rate with a biofeedback monitor and enough practice) to eating habits to home energy use to personal finance. The difference is Schell proposes marking actions as "good" or "bad."

Bogost and commentors have basically suggested two questions: 1) Is simply showing stats morally superior to stats + good/bad judgment? 2) and which version is more effective at changing behavior? To this, I would suggest that the government has no problem saying to hell with "morality." It provides tax breaks and incentives for companies to act "morally" to reduce green house gases. It imprisons and punishes those who transgress. Morality is good, but incentives work better.

I would suggest some other ideas: What if these raw statistics were simply made public, without any external judge? Would people say, "Yikes! I'm using twice as much energy as my neighbor! Maybe I should change my behavior so I don't look bad." Is that morality or external judgment and can the two really be separated? (Or imagine a forced gamertag, visible to all which says, "Little Sister Killer." Would that change people's gaming behavior?)

Finally, Schell fails to acknowledge a simple psychological truth: Providing too abundant a reward schedule leads to fast extinction of desired behavior. To be realistic, Schell's proposal needs a more variable reinforcement schedule. You can't get points every time your brush your teeth -- you can only get them some of the time.

Ian Uniacke
profile image
I agree with Bogost. This is a prime example, IMO, of the classic duck speak idea presented in 1984. That is, the danger lies not so much in the outcomes (I'll avoid the whole argument of whether we have a right to dictate good behaviours), but in the way it would shape the mind. If in the average persons mind all they have been brought up to believe is brush the teeth 5 points, drive the car 2 points, kill the maid -10 points, then they will over time have no capacity for rational thought, since it's the language as we understand it that drives us to think rationally. If we spend our lives thinking in points then our capacity for higher level language will decay and soon we won't even be able to "think" of higher ideals let alone contemplate them.

I can totally see I-Robot style situations where a person who has become accustomed to earning points, for example, abuses the system because they just want the points. At what point does a parent tie down their children and brush their teeth for them to earn some more points and is this parental abuse?

No one said that education is easy. As a parent, and this same argument applies more generally, you have to concede that the best you can do is educate and encourage and if the child (or member of society per se) does not want to do the behaviour you wish, then you have to let it be (or alternatively give the punishment if applicable/let the consequences occur (eg your child is the stinky child)). To do otherwise is removing the humanity from that person.

You can probably tell I'm more on the radical end of the political spectrum then most but these are just my 2 cents on the topic.

Ian Bogost
profile image

That's a good one, isn't it? :) I'm making a distinction between the outcome and the choice.


Yeah, that's part of the argument in the research Jesper posted (linked above). But it's a different argument than the one I'm making here, which is not about efficacy.

Ian Bogost
profile image

About statistics: making data visible doesn't necessarily imply any sort of incentive structure whatsoever. Information can provide evidence that motivates decision. But philosophically, we must still distinguish between a society that behaves a particular way because it believes such behavior to be virtuous, and one that calls itself virtuous because it appears to behave in a particular way.

Austin Walker
profile image
Ian, quick comment/question: The way you phrase things it makes it seem as if you believe that there can only be one type of incentive for a given action. That if a person is given points for brushing their teeth, it will be the only reason they perform that action. I think that it's fairly easy to build that argument when you're talking about turning actions children want to do into these schell games. But I think that it's more dicey when you talk about the activities of adults, or in general any activity that has an ethical/social, qualitative, or really any "rational" component.

Imagine that it's election season. The DNC and the GOP are both incentivizing volunteer work (working at call centers, door to door canvassing, etc. Let's ignore that this is probably against a rule I'm unfamiliar with and if it hangs you up that much, tell me and I'll try to come up with an alternative example, but I think that this is the most persuasive.) It strikes me that what you're saying is that people will volunteer for whichever pays more points. They won't be volunteering due to their political beliefs, but because they want a higher score.

I can't speak for everyone, but the promise of a bigger payout wouldn't make me volunteer for the other side. In a world where everything is scored, I suspect personal preference will still come into play. In fact, that a person has chosen to gain points by volunteering - instead of watching tv, going to the gym, or playing flash games - already says something about their proclivities.

Ian, I understand your trepidation here, and I do think that Juul's argument is compelling. But I'm curious if it holds up inside of a system change. I've always played video games, and I don't /think/ that I play them worse now that I'm awarded trophies or achievement points. But maybe that's because I'm not a child. And even as a child, I think that I responded very well to some incentive programs (the Book It! reading program specifically sticks out in my mind.)

I think that's my driving force here - why are we saying that the future will be just incentivized activity or must be the opposite of that. The factors at play already suggest that we'll see some of real life "Lockerz"ized. Perhaps the lesson here isn't that we have to throw away the schell game all together, only do our best to limit its pervasiveness.

(I also have some bizarre determinist notions that make me think that even without explicit point schemes, every action we do is incentivized by innumerable factors, and that's why we do them. But that's neither here nor there.)

Austin Walker
profile image
First para, I say "children want to do." I meant "children don't want to do." This is what I get for not sleeping.

Bart Stewart
profile image
Thereís a crucial political question to ask here. Itís been asked in this very form for a long time, but itís worth asking again in the context of Schellís presentation: if real life is to be gamified, who gets to set the rules of the game?

Letís take brushing oneís teeth as an example. Who gets to say that brushing your teeth in the morning is worth 5 points in Game A? Presumably itís not you, or you could pay yourself anything you wanted. So you must be playing someone elseís game... but whose?

Is it your familyís game? Does Mom get to decide whether to incent toothbrushing? What happens to parents or to people who donít have an immediate family to whom toothbrushing data can be sent -- are they not permitted to play the game?

Is it businessís game? Should the toothbrush or toothpaste manufacturer get to set the rules of the game? What if different companies create different games -- whatís to stop players (i.e., people who brush their teeth) from shopping around to see whose game gives the most points?

Is it your governmentís game? Do you want to pay for the bureaucratic organization that will need to hire civil servants whose only function is to monitor your toothbrushing data, and to establish panels that arbitrarily set how many points you receive for brushing your teeth? When they can collect toothbrushing data, how do you argue against them also collecting other data? Can you choose not to play their game?

In other words -- and as has always been the case -- the real question is what amount of choice we will have when our real life actions are rewarded with XP and achievements.

Will we have a choice in whose games we play?

And will we be able to choose not to play at all?

Nick Braccia
profile image
I haven't watched Schell's video yet, only read the highlights, but I did read your piece in full, Ian. I think you're right though. Initially, the presentation of a World of Chorecraft sounds dazzling and seductive. This approach has existed in many forms and systems to address issues in self-control/will-power and discipline. Usually, the context is "life or behavior training of some sort" . I'm thinking about Weight Watchers or the "technology" behind EST and Landmark Education. In these cases, the systems work when they're accompanied by open, honest dialog, personal expression, understanding and development. They don't succeed when the euphoria of groupthink is the driving principle. That's something the carnival barkers will be looking to exploit. There's a reason why the old Medicine Shows had music and comedy acts. They had to seduce people to enter the tents hear the pitch and buy their cure-all snake oil. There's also a reason why people often stop attending 12 Step meetings. The work is not easy (even if it's built against a clear and seemingly simple system/platform). Take away: if it looks like there's an easy, cruise-control way to incite a change in behavior, it's probably a ruse. At the very least, subscribers should be candid enough (with themselves) to admit powerlessness. I don't believe automation is evil, but mendacity...that's tried and true.

Scott Brodie
profile image
"The heart of games is not points, but process. Games have the capacity to persuade us because they can depict perspectives on how things work, and they can give us insights into the complex and often ambiguous connections between them."

Beautiful final section. Great article Ian.

Ian Uniacke
profile image
Another issue I have thought of in regards to this system is it is completely ignorant of free market capitalism. The amount of money each person has to spend will not change. Therefore the cost of toothpaste will increase, since people with the achievement points get a credit on toothpaste. Therefore, the system will move from being a "reward" for good behaviour, to being mandatory if you wish to survive in a competitive world. You can see the exact same thing happening in Australia with government grants for housing, pushing housing prices higher.

Daniel Cook
profile image
Having built effective games of the type discussed here, there are a couple points to be made.

- Yes, games are a process. In the case of games involving people, we can create reproducible processes that have experimentally validated outcomes. This is not new. Governments, religions and social organizations have been doing very similar things for many thousands of years.

- Electronic games automate traditionally human processes in ways that are scalable, maintainable and constantly validated. This is new, at least in the degree to which they add efficiency.

- Electronic systems also provide feedback on a much finer granularity than is typically cost effective in other rule-based social systems. This allows games to reach in portions of our life that have not historically seen enforceable governance.

- Philosophical and ethical arguments will at some point need to face the pragmatic reality that these systems do work. A measurable percentage of players will behave as desired by the designers of the systems. This has clear economic value. It has the potential for generating social value as well. And like a shell game, it can be used for ill. Games that reach into our everyday lives are a tool, not something inherently blessed or damned.

As for the article, the following thoughts came to mind:

- For many games, the player never judges if the final outcome is useful. Instead they judge if the immediate next choices are worth spending energy on. Players wear blinders. Only the operator sees system as a whole. In this sense, the operator with their access to the behavioral records of thousands of players, is always behaving in a highly manipulative manner. It is what we do...all good game designers play a rigged shell game with their customers. Otherwise, we cannot effectively train players to slowly gain the skills they need to comply appropriately with our end goals.

- It is a common mistake that points are seen as meaningless baubles that reward players to earn more points. Our players are not that gullible, at least not on a subconscious level. Points, in a well designed game, are merely a currency that players exchange for something that matters, be it status, the love of your mother, etc. To say that game that use points to motivate players in the real world will be shallow is merely claiming they are badly designed games. However, being able to point out one badly designed game does not eliminate the possibility that you can make well designed real-world games that are effective, well balanced and appropriately use both intrinsic and extrinsic motivation.

take care,


Altug Isigan
profile image
I agree that games *are* perspectives and that they present a frame that selects certain aspects for interpretation and leaves others out. The Ford Fusion example here can be seen as some sort of Agenda Setting for example. It does not dictate to you what you should think, but then it keeps you busy with finding solutions *within* the constraints of the particular system that generates the problem. You keep wondering and pondering on how to make the tree thrive (bascially you confirm and conform to fuel-based technologies), but maybe you should prefer to think whether cars should use electricity or whether mass transportation systems should be improved and the number of privately-owned cars reduced etc.

In the moral luck example I think it is important in whose favor you interprete probability. From the point-of-view of the old woman, was it just bad luck to cross the street at that time and place? Should she not expect that drivers will watch the street rather than searching for the radio button or receiving a phone call? Resultant moral luck sounds a bit like "We can be ignorant of others, and if we are lucky noone dies." In my opinion it looks like it cancels out the notion of responsibility while it attempts to fight a narrow-minded conception of morality.

To continue with the Farmville example: Is it OK when I exploit friendship for positive gain in a game? Well, it looks OK, because if I choose to play a game and if this kind of relationship is part of the game contract, why not? But what when it is the game designer who deliberately uses invitation and cooperation mechanisms to exploit "my" network of friends to broaden "his" customer base and sell us alltogether to marketing and advertising companies? Can or should we draw a distinction between the ethics of the game itself and the ethics of the business model that sets certain design goals?

And what if other game designers think that herein lies an opportunity to make a profit and wonder whether it might be a good idea that we get points for brushing teeth? And for washing our hands? Or getting points deduced if we have sex before marriage? Will it be bad moral luck when a truck full of moral games hits me because its drivers are distracted since they're receiving calls from venture capital providers? ;)

David Boudreau
profile image
"It is entirely a matter of luck. In one case, an old woman happened to be crossing the street, in the other she didn't. "

Luck had absolutely nothing to do with either case. Don't think it does and start mucking with radio dials when you should be FOCUSED ON THE ROAD, buddy. Of all laws, coming to a stop at a red light is the easiest to recognize as a good one.

Otherwise, interesting article!

Bart Hufen
profile image
I think the biggest reward for people playing games or doing stuff in life is the positive and (even) negative feedback on behavior. Playing a game is like living a life. You have ups and downs and overcoming bad things in life enhances the experience (what doesn't kill you makes you stronger). We have always been rewarded with points, discount etc. but true strength and opportunity for bonding with brands lies in having a relationship that is worth fighting for (in a game) or investing in (time) and finally - has a dynamic feedback / reward system. Things that made me tickle when I was 20 don't make me tickle anymore (37 years now). It requires a continuous dialogue with you target audience to understand how to maintain the flow!