Archive for Games

Reward That!

In third grade, I participated in the MS Readathon. Just like a Walkathon, the Readathon asked you to collect pledges for each book you read during a specific period of time. Depending on how much money you raised, there were prizes and rewards – books, posters, certificates, and more. Plus, our school held an assembly where the highest earners were publicly honored.

Now, as you probably know from following this blog, I love to read. Back in third grade, that was just as true. I spent my recess time with my nose in a book, and my favorite activity was going to the library.

I also, as it happens, love public recognition and praise. So as a third-grader, I set my sights on being the highest earner in the school. I wanted to stand up in front of everyone and have the principal say what a good reader I was, and how I was so committed to community service, and that everyone in the school should try to be more like me. I literally dreamed about it for weeks.

I was a smart kid, so I sat down and did the math. I could spend my time recruiting lots of donors – not really my strength. Or I could get a couple of per-book pledges, and then just read a whole hell of a lot.

I somehow managed to wangle a dollar-a-book pledge out of my parents, who should have known better. After hitting up my grandparents and my parents’ closest friends, I was up to $2.50 a book. After researching how much the previous year’s winner had raised, I figured that I needed to read at least a hundred books in thirty days in order to win.

Let me say that again. A hundred books. Thirty days.  If I wanted to get on that stage, I needed a plan.

On the first day of the Readathon, I went to the library. Instead of heading for my usual haunts, the science fiction and adult sections, I went straight for the kids’ room. I picked the shortest, thinnest chapter books I could find. Ramona Quimby? Two hundred pages. Out. Frog and Toad Are Friends? Sixty-four pages. In. Forty books were no problem for me to carry home, and at that length I could handle at least one a day with ease.

Next I hit up my sister’s collection. She was in kindergarten at the time; even though she’d also been an early reader, her books were nice and short and simple. She wouldn’t let me take them out of her room, so I put a pillow on the floor and dug in. I could easily read five a day, in addition to my library booty.

To make a long story short: in thirty days, I read a hundred and ninety two books. My donors were flabbergasted. I raised a whole lot of money. I got up on that stage, just like I wanted. And after the ceremony, the principal called me into his office to explain that I hadn’t exactly kept to the spirit of the contest.

“You mean raising a lot of money?” I said. “I did that!”

“Well,” he said gently, “how you do it also matters.”

I wasn’t going to smart-talk the principal, but I knew better. If the how mattered, they would have put it in the rules.

I’ve been thinking a lot about my Readathon experience lately, because it’s a classic case of what’s referred to as “undermining intrinsic interest with extrinsic rewards.” I loved to read because I loved disappearing into a story. I loved learning new things and challenging my mind. I loved talking to my dad about books, and I lived for the approval he’d give me when I had something interesting to say about them. But for thirty days, all that was on hold. My approach to reading became an instrumental one, all about quantity rather than quality. Instead of thinking about what I was reading, I was thinking about how fast I could finish and what I would look like up on stage. I was thinking about what reading would get for me, not what reading meant to me.

Now, that doesn’t mean I think that all rewards undermine your interest in things. In fact, most complicated rewards start out as extrinsic to the individual. We’re biologically programmed to find certain things rewarding (food, affection) but over time we develop more complex desires. Even if humans have a common hierarchy of needs, the ways we think we can fulfill them are defined by our shared culture and our individual experiences. I may have an innate need for novelty, achievement, respect, and creativity, but learning to get those things through reading took time and was supported by rewards that weren’t always related to reading per se.

The question, then, is how to design an external reward system that supports intrinsic interest rather than undermining it. I found three lessons for reward system design in the Readathon, so I’ll share them with you.

First, reward the right things. The Readathon rewarded the total amount of money raised, which lent itself to exploitation. I knew kids who saw that as a challenge to read as little as they could. Instead, they spent all their time soliciting pledges. I was terribly jealous of a girl in my class who was getting more than ten dollars a book! Similarly, my “read only easy books” approach was all about gaming the system.

Second, control reward potency. Part of the reason that I dropped my pleasure reading for a month was because the external reward – being publicly praised – was so powerful for me. Now, not everyone feels as strongly about recognition as I do, but not everyone loves reading as much, either. The key is to start with a good ratio between the value of the extrinsic and intrinsic rewards, and to lower the relative value of extrinsic rewards as intrinsic motivation increases.

Finally, consider your audience. One of the Readathon’s unspoken goals was to get kids more engaged with reading. But I was already a passionate reader who could handle adult books! I don’t think I read a single book at grade level, let alone at my actual reading level, in the whole month – but for another child the books I read might have been challenging. A system that’s going to include a young Jessica, a grade-level reader, and a struggling reader has to have rewards and roles for all of them.

Fortunately for me, my Readathon experience didn’t change my lasting attitude toward books – but not everyone who encounters a broken reward system is so lucky. That’s why I spend my time working on reward system design, and trying to come up with principles that can help other designers do it better.

Burn It Down!

I was terribly excited to read about Azmo the Dragon, an entry from Rice University students for the Imagine Cup. I think it’s a really inspiring project that has the potential to help a lot of people!

Here’s what I like about their game.

  • It defines a specific problem. Kids with asthma are supposed to be tracking their lung capacity every day, but blowing into a spirometer is frustrating and dull. Only 50% of kids do it; the game tries to raise that number.
  • Playing the game is solving the problem. Getting people to transfer knowledge from one context to another is hard. If the only action you need to take is play itself, the designer’s task gets a lot easier.
  • Better performance on the task correlates with better performance in the game. Breathe long and deep enough, and burn your castle! Fail, and watch them snatch victory from your hands! (Er, snout?)

I strongly suspect I’d have some critiques of their game, too. It sounds like the player has to jump through some unrelated hoops before breathing, and I wonder whether there are any in-game challenges to do while doing the measurement. Still, I think this is a smart problem to tackle and a great game-based way to solve it.

Plus, I’m always excited to see more groups working on breath control in games!

RTFM

I just came across a great study out of MIT about how game manuals get used – but not by people! Professor Barzilay and her team created a computer program that learned to play better by reading the manual.

This is potentially a very big deal. Reading – or, as the artificial intelligence folks call it, natural language processing – is one of those problems that theorists thought would be solved fifty years ago, but turned out to be way tougher than expected. Because there are so many ways to express any given concept, it’s extremely difficult to translate text into the underlying logical concepts that a computer can understand.

As part of her research, Barzilay developed a program that could play the game Civilization. The program didn’t know much about the game, but it could move the cursor, see what was on the screen, and tell whether it had won or lost the game. Over time, the program learned to play the game more and more effectively – winning up to 62% of its games. But next, Barzilay decided to let the program examine the text of the game’s manual. It didn’t know what the words meant, but it could figure out what words appeared in the game and compare them to words in the manual, then take relevant actions. Using this strategy, the program’s win rate shot up to 79%.

As the MIT article points out, this research shows a possible new direction for natural language research. But what excites me about this is how strangely similar it is to the way that people use game manuals.

Jim Gee writes about game manuals and “situated meaning.” He argues that for most people*, the words in the manual are just words. We don’t easily translate them into the actions we’re supposed to take while playing. Instead, we play the game, and then use the manual to help us understand what our actions mean. Reading about, say, the V.A.T.S. in Fallout 3 is very different from using  it, then looking it up in the manual for additional insight. In the latter case, we already have a sense of what the V.A.T.S. does** because we have experienced it, and we know how our use of it may or may not serve our larger goals in the game. All of a sudden we have a specific and meaningful context in which to make sense of the words in the manual – which, otherwise, are just words.

Barzilay is essentially taking Gee’s insight and putting it into practice. She lets the computer make sense out of text in the context of play. Did this insight from the manual help me win the game? Which pieces of the manual are related to the in-game actions I’m taking now? These are questions Barzilay’s program asks itself – but they’re questions we human beings ask ourselves, too.

Neat.

* I think we all know someone who insists on reading the whole manual before starting a new game – but even they probably have trouble applying what they’ve learned until they get a chance to play.

** Of course, different people use the system differently. In my case, I often used the V.A.T.S. to see where enemies might be lurking in the environment, because they would be outlined in red against a dark background.

Fear Fails

I just came across a piece on the likely effectiveness of the new, gruesome cigarette warning labels developed by the FDA. The idea is to scare people into quitting by graphically illustrating the consequences of smoking, right there on the package. Think this will work?

Nope, me neither.

Over 70% of American smokers already want to quit. That’s over 31 million people – an enormous number. Most of them will fail. The gap between desire and performance suggests we’ve been largely successful at educating people about the risks of smoking, but much less successful at helping them actually stop doing it.

The point of failure isn’t educational. It’s behavioral.

Telling someone, “Stop smoking or terrible things might happen to you someday!” is about as ineffective an intervention as you can imagine, even when it’s backed up by images. Our brains don’t do well with might or someday, especially when the decisions that get us to might and someday are repeated, daily behaviors. It’s easy for us to talk ourselves out of a conditional and faraway goal; after all, this cigarette won’t kill you. And if the little decisions get repeated often enough, they may take their toll on your willpower* – or hit you at a time when you’re not in the mood to think about your long-term goals in the first place.

Is it any wonder that fewer than 5% of smokers successfully quit on their own? When we say, “Just stop smoking,” we’re asking smokers to do something extraordinarily difficult and complex. However, we often speak and act as though it’s easy. This is a lie, and it’s one that is deeply damaging for smokers – especially those who have tried to quit and failed.

That’s why I’m so proud to be involved with the Lit2Quit project. When my students first came to me with the notion of a smoking-reduction game, I told them they had one restriction: they could not tell people to quit. That’s not the missing link for most quitters. What they came up with – and what I’ve been honored to be involved with, over the past two years – focuses on the behavioral aspect of quitting. When you’re not feeling serious or goal-oriented, you might still be willing to play a game; when your willpower is not at its best, the game helps reduce your cravings so you’re more likely to be able to manage them.

The difference between our project and the FDA’s new approach is fundamental. The FDA seems to believe that more information will help people quit smoking. The assumption is that people only smoke because they don’t really understand how bad it is for them; if the FDA can just communicate the facts more effectively, people will change their actions. Our project assumes the opposite. People who want to quit know what to do, but the hard part is translating it into action. That’s where we can help. As we collect our data, we’re finding out just how much!

* There’s some indication that the resource-depletion model of willpower may actually be a mindset issue; if you believe that your willpower gets depleted over time, it does, but if not, not. However, there’s enough long-standing research into willpower as a depletable resource that I’m not yet ready to toss the notion without further personal investigation.

My Health Month

A friend asked me this morning, “Does your post about playing nurse mean you think gamification is a bad idea?”

I told her that I think gamification will sometimes be done badly or thoughtlessly – which doesn’t make gamification bad in itself. In fact, it makes me want to be involved with good gamification projects*, because there needs to be projects that can push back against the inevitable deluge of exploitationware.

“But you don’t like points, right?”

I had to chew on that one. Points, I think, lend themselves to lazy design. Want someone to do the thing you’re trying to promote? Give them points for it! Why bother understanding the psychological factors behind their current behavior? Why create a meaningful aesthetic experience while playing your game? Why consider how your game activity fits into their lives? Everyone wants points because points are awesomesauce!!!!

But that’s not the same thing as not liking points. When I play Angry Birds, for example, I pay careful attention to the points**. I’m currently trying to get three stars in every level, and I can use the points to assess how I’m doing at achieving my goal. If I don’t get enough points on my first shot, I can restart the level instead of playing it through to the end, which saves me time. It also lets me compare the relative effectiveness of two shots. I don’t just use the points to make these decisions, of course – I have to analyze the larger structural situation, or I’d quickly get stuck at an ineffective local maximum. I do, though, use them as a way to quantify and evaluate my progress; they’re a very useful shortcut. (And getting three stars on a level is deeply satisfying!)

A more “gamified” example might be my experience with Health Month. I’m still in my first month of using the service, but so far it’s quite interesting. You define health-related rules for yourself, like drinking water or avoiding sweets or exercising regularly. (My rules are to walk at least 12 miles a week and to eat five servings of fruits and vegetables at least five days a week, though I’ll probably add another rule next month.) When you log in and record your progress, you get points, with bonuses for things like logging in every day. At the end of the month, you compare your points with other people in your bracket to see how you did.

I’m not terribly motivated by the competition, but I do try to log in every day to get my +10 points, and to use the integrated micro-journal for tracking mood and activity levels. However, the game has a second currency which I find highly motivating – because if you break your rules too often, you start to lose life points. There doesn’t seem to be a consequence for running out, but the implicit narrative makes me very concerned with keeping to my rules. I lost my first life point this week, thanks to a headache that left me unable to eat much of anything (let alone anything crunchy!) for a couple of days. It was surprisingly emotionally unpleasant, and I’ll be actively working to avoid losing any more.

Health Month, by the way, makes an interesting contrast to the MeYouHealth Challenge, which I am also experimenting with. MeYouHealth sends me a daily challenge email, which can range from eating a red fruit (yum, cherries!) to contacting an old friend to doing a quick set of lunges. I really enjoy taking on the challenges, because they bring delight and variety to my day, without turning into a commitment or a chore. Doing the challenge is itself the reward.

The metric I track, with MeYouHealth, is whether I’ve done every single challenge and recorded it. I know they have some kind of points system attached, but I don’t care much about it***. What I care about is being able to look back at all the challenges I’ve done and remember the experiences I had doing them – especially when I’ve added a brief note about how I did it. This motivation bleeds through into my reaction to their other design choices. For example, it’s frustrating that you only have 24 hours to record doing each challenge, because I can’t necessarily sign on within a given 24 hour period; I don’t care about getting the points, but I do want to log what I did so I can reflect on the fun I had while doing it.

Contrasting Health Month and MeYouHealth helps me talk to people about gamification, because it lets me show how two apparently similar systems can produce different kinds of fun – and use points more or less effectively to do so.

* Full disclosure: I actually am involved with both Dr. Lee’s Gamifying Education group and the Think Play project out of RIT. I think they are going to do this thing right. To me that means picking answerable questions backed by appropriate theory, designing a playable (and non-lame!) game layer, and doing meaningful assessment. So far, so good!

** Another reason I like my field: a discussion with my adviser about play styles turned into a detailed comparison of how we use points in Angry Birds.

*** This deserves a more careful analysis, but my current theory is that their points system is both visually and conceptually cluttered compared to Health Month.

On Playing Nurse

I’ve been trying to pull together my thoughts on gamification, but I haven’t figured out how to pick out a chunk that’s blog-post-length. So instead I’m going to tell you about nursing my husband through surgery, which was my main non-dissertation activity through most of May, and why it’s made me think carefully about pleasure, process and rewards.

My husband needed to have some growths removed from his throat, which is not a risky surgery but has quite a long recovery time. It’s also extremely painful. I knew my job as nurse would involve both physical care and keeping an extremely cranky patient entertained.

I was, I admit, dreading the experience. I am neither especially patient nor particularly nurturing. When my alarm went off for yet another middle-of-the-night round of medications, would I just be irritated by the whole experience?

It turned out that the answer was no. Sure, some of the 4am wakeups weren’t so fun. But I found it deeply satisfying to fall into a task-based routine, an almost monastic existence. I had two humidifiers, three cases of Gatorade, a giant bottle of pills, and a comfortable chair from which I could see my husband’s sleeping face. I learned I couldn’t expect to get much writing done, but I could enter bibliographic data into Mendeley, or read those articles I’d always meant to get around to, or finally copy-edit the paper I’d intended to finish months ago. Yes, I’d be periodically interrupted, but the interruptions and the work and the napping and even the sleep deprivation were all of one piece, woven together into a world that made coherent sense from the inside even as it was a step away from my ordinary life.

Here’s what I didn’t do. I didn’t give myself points when I administered medication on time. I didn’t collect the empty Gatorade bottles to symbolize the triumph of calories successfully administered. I didn’t advertise my activity through social networking, though I did use it to update my community on how the recovery was going.

I did, I admit, wear a nurse costume. But only once, and no, you can’t see the pictures.

This is the heart of what I have to say about gamification: that what’s meaningful about it is not any particular trappings or techniques, but whether we can successfully create a magic circle within which certain activities become important, necessary and right. I played nurse, and it was magical. I want to figure out how to get people to play learning in the same way.

The Thousand Year Game Design Challenge

Daniel Solis has issued a challenge: can you design a game that will still be played in a thousand years? If you think you can, go check out the Thousand Year Game Design Challenge.

Personally, I doubt anyone’s going to succeed, but the constraint is a brilliant one. It forces you to prioritize elegance and simplicity, and to create a game that can resist (or embrace!) variation over time. You also have to think about the way your game will be passed through social institutions. Do parents teach their children? Do children teach each other? Are you relying on a formal social institution, like schools? And if so, what do you think that institution will look like in a thousand years?

I used a variant of this challenge in one of my game design classes a few years ago. I asked my students to propose a game that would be played over the course of a thousand years. I got some great proposals for games in which game position was hereditary; in which a locked vault opened every ten years to allow a new move; in which two foundations were established to competitively play across the generations. I think Daniel’s likely to get submissions that look more like Mancala and less like, well, the awesomely crazy things my students came up with, but I guess we’ll all find out next January!

Strange Systems

There’s a fantastic article by Simon Strange in this month’s issue of Game Developer Magazine. He talks about how to diagram game systems in terms of dynamic equilibria. What game systems impact what other systems? Which have a negative effect and which a positive effect? Which game systems are most tightly coupled to the larger design, and which can be tweaked until fairly late in development? It’s eminently practical, with a running example from Doom II that makes his concepts very clear.

What made me go “Ooh!” is that he seems to be essentially proposing an explicit form of systems thinking for game designers. As soon as I saw his diagrams, I thought of Stella and other tools for understanding dynamic systems. It makes me wonder whether this kind of exercise could be useful both for teaching game design and for teaching systems thinking. I’ll definitely be basing future game design assignments on this technique.

I also wish I’d had this tool earlier in the development process for my current projects. I’ll be drawing some system diagrams of my own game next week for sure!

Thanks, Simon!

Match Three, Feel Better

I’ve been saying for a while that “games for mental health” are the next big frontier in serious games, ever since I came across a study on using Tetris to treat PTSD. Well, there’s been some press (here, for example) about a new study that claims that casual games can treat symptoms of depression. I checked out the slides, but it’s hard to tell from slides alone just how rigorous the research process was. I can already see some flaws in this study – for example, I’m concerned that they allowed players to choose among three different casual games*, and I also wonder whether the researchers could accidentally have invoked the Hawthorne effect. It’s also not clear how they handled recruitment, which might lead to generalizability issues.

That said, I’m really looking forward to seeing the full study when it’s published. Even if this study has some possible problems, it points the way to the next study, which lets us ask better questions and get better answers. Jesper Juul hits it on the nose when he suggests that we need to investigate how casual games compare to other semi-structured activities, and what factors of game-play were actually effective. For example, game-play might have helped by giving subjects a manageable task, a quick hit of success and competence, or they might have helped subjects practice emotional self-regulation.  I’d like to see what theories the researchers drew on, and why they chose the games they did. That might help narrow down possible mechanisms.

I’d also like to see this work tied to research on non-digital play therapies, including art therapy. There’s a bit of a tendency to treat digital games as creations ex nihilo. Let’s not forget that we know at least something about the efficacy of existing therapies, and make sure that’s a part of the conversation. For example, maybe we want to target game-based depression relief at people who might not otherwise seek help, or at people who show poor adherence to other treatments, or at people who can’t afford extended therapy. Maybe we want to do all three, but these groups react differently to different casual games. This issue just gives us more good questions to ask.

Hey, anyone want a research agenda?

* I mean, the researchers have the right instinct: freedom is terribly important in play. But now they’ve got three different experimental treatments without any rationale that they should be treated identically. Additionally, they didn’t offer the control group any choice about what to do. Maybe the whole thing ties back to that initial decision to take an active role in one’s own treatment, especially considering that the control group wasn’t asked to do anything between research sessions.

Collegiality and Consequences

The thoughtful and courageous Courtney Stanton has recently gone from “very excellent” to “extraordinarily excellent” in my personal estimation. (Want to know how to do this? Be awesome. Then keep being awesome, but add data.)

In the course of blowing my mind, she said something that I think is remarkably important and I’d like to single out. At the end of her post, she writes an open letter to one of her harassers, concluding with this:

But kid, let me promise you right now: if you ever try and use the computer programming skills you got in community college to work in the video game industry, I will be expecting an apology from you about the harassing tweets and disgusting images. I can have a long memory when I choose to, and since you made such a compelling case for me to do so, I’m choosing to now. I hope you eventually grow up, but in the meantime your behavior shows you to be exactly what this industry doesn’t need. I’m not afraid of making that fact known should you attempt to enter it. You wanted my attention; you got it. Enjoy the consequences.

I recently wrote about being honest about people’s flaws, but I was putting it in terms of who you might or might not want to work with as an individual. However, Courtney’s hit on something even bigger: that these choices have an impact on the entire community. Let enough rape apologists, or trolls, or just plain jerks into a group, and eventually they start setting the social norms. That in turn has an effect on who wants to be in the group, and how they are treated when they are there, and I think we all know how this one goes.

My post, too, assumed a basic level of human decency. I’m happy to empathize with a colleague who tends to give harsh feedback on projects; if you’re prepared for it, his feedback is actually very helpful. But am I willing to empathize with someone who deliberately causes pain to others? A harasser? A abuser? A rapist? I’ve got a line, and once you’ve crossed it, crossing back is going to be pretty hard.

This is feeling particularly relevant right now, because recently I’ve done a lot of community, leadership and mentoring work. In the last ten days, I’ve:

  • Made over a dozen professional introductions
  • Arranged a job interview for a former student of mine
  • Critiqued a research design for a colleague, and suggested fixes
  • Volunteered my time to guest lecture in a different colleague’s course
  • Helped a peer refine her dissertation proposal
  • Consulted with a startup about their new product
  • Helped two would-be entrepreneurs understand my sector

And that doesn’t even count the meetings I’ve scheduled for the coming weeks and months!

I’m writing about this not to pat myself on the back*, but because it’s made me think about my responsibilities and obligations. Yes, I try to judge everyone fairly, but I have a limited amount of time and lots of people I could be working with. In whom do I choose to invest my time? When I help someone become part of the game design community, or for that matter the academic community, I should remember to consider how their presence is going to influence that community in the long run.

I’m lucky that, as far as I know, the people I choose to work with are Good People. For example, the only reason I’ve ever had to cringe at a joke has been because of a truly groan-worthy pun. I don’t worry that they’re going to behave unethically or exploit others or engage in random acts of senseless cruelty. But I’d like to formalize that commitment, so here goes.

If you ever would like help from me; if you would like introductions to my network; if you would like direction in your job search; if you would like to collaborate with me on a research project; if you would like me to guest lecture in your class; if you would like my advice about graduate school admissions, or your dissertation, or how to be a productive scholar; if you would like my insight into game mechanics or psychology or learning; if you would like any of these things, don’t be evil, because I will hold you to account.

You are probably wondering what I mean by evil. I happen to like Jonathan Blow’s summary: “selfishness to the detriment of others or to the detriment of the world.” In the end, though, I’ll be using my personal moral judgment. I’m by no means perfect, but I spend a lot of time thinking about how to behave ethically in the world and how to be a decent human being. I already use those ideas to guide my work, my relationships**, and even how I spend my money. It’s time to apply the same ideas to my life as a mentor, leader and teacher***.

Yes, there are a lot of corner cases and ambiguous scenarios. (“What if I did something evil a long time ago, but I’m reeeeeeally sorry now?”) If you’ve changed, or you didn’t mean it, or it was totally justified****, fine! The world is a complicated place, and I get that. But the burden of proof is on you, not me – and at the end of the day, I may still decide you aren’t a good investment of my limited resources. For that matter, my resources aren’t infinite. I may have to turn you down no matter how awesome you are.

By making this commitment public, I’m hoping to inspire others to do the same. I know so many people who mentor and advise and teach and lead, and I know that many of them likely already do think carefully about who they invest their time in. But what if we let people know that? What if people who care about our industry had an explicit model for decency? What if virtue – yes, I know, an unpopular word – were a way to access informal networks and formal power? I don’t know about you, but I think the game industry would start to look rather different from the way it looks today.

(I give it five minutes before I get trolled for this, but I’m posting anyways.)

* Okay, maybe a teeny bit of back-patting.

** In my personal relationships the bar is actually quite a bit higher than “don’t be evil,” but that’s another post for another time.

*** In some ways, I already think about this as a teacher. I have to teach the students I’m sent, as long as they conform to my institution’s standards, but I can infuse every lesson with a clear moral stance. I’ve been reading The Sociological Imagination and it’s becoming increasingly clear to me that, in fact, it’s impossible to teach without one. I just try to be conscious and explicit about mine. This is probably also a topic for a longer post, huh?

**** On second thought, you’re probably not gonna get very far with this one.