This post is a response to the Chris Hecker talk given Day One of the main GDC conference this year.
I'll write a synopsis of this talk, "Achievements Considered Harmful?", in my own words, but I'm not the first person to do so. See also Game Set Watch's writeup and this writeup from David Sirlin (scroll to the very bottom of the entry for the Hecker talk). I'll also try to, as I just did, cross-reference relevant information about the talk as necessary to explain it. Hecker's talk about Achievements was full of cross-references to books and videos that describe the phenomina that he's trying to explain.
Hecker started out by saying he wasn't here to give a talk about how Achievements were considered harmful. He was going to give a talk that asked the question, are Achievements considered harmful, with a great big question mark. He jokingly showed a slide that also mentioned that Considered Harmful Essays are Considered Harmful and told us to take anything that anyone considers harmful with a grain of salt.
That being said, he got started on this track himself while reading writing not about game design, but about parenting. Meet Alfie Khon, a man who wrote the book Punished By Rewards. Alfie has some beliefs and some biases which are evident in his writing. This book in particular is about the gold stars and other rewards that parents and teachers give their children, and whether or not those rewards actually help, or hinder, childrens' learning and development.
Here's a talk on TED on a similiar vein: Dan Pink on Motivation. Pink also later wrote a book called Drive on the topic of what does, and does not motivate people.
Hecker read both books, but has a couple complaints: the books are very repetetive, for one, and for another, they tend to use a lot of anecdotal evidence to back the claims made in research studies. Anecodes are easier for a mainstream audience to understand than dense psychological studies, but they aren't science. So Hecker went to the studies mentioned in both books and the general literature about the issue of rewards.
He stops at some point to warn the audience that even the hardest psychological studies are not really a hundred percent conclusive. These studies do the best that they can, but in the end, experts disagree on what motivates people, and there's a lot of back and forth fighting in academia that he personally lampoons as being a little immature. People think of fMRI as hard data, but all science is a little bit "dirty," with conflicting evidence and different opinions that color studies.
There is a subset of psychology called Behaviorism, pioneered by B.F. Skinner. Skinner mostly experimented with rats, putting them in a thing called the Skinner Box which they pushed on to get the pellet. If you want to learn more about the connection between behaviorism and modern video games, this article on Cracked.com covers it pretty well, and in a way that a lot of people describe as scary. If that's not enough for you, here's another video: Jesse Schell at DICE 2010. (Have I given you enough reference material yet? It's the internet - it's all connected!) This particular talk got a lot of wide circulation because it discussed the "gamification" of not only games themselves but of the real world. Soon everything, Schell argues, may become a Skinner Box. Push button, get pellet.
A lot of people consider this a little bit frightening.
But Hecker also thinks, based on the studies, it probably won't even work. Schell says in his talk you might tend to read better books, if you discovered that your descendants were actually able to see a list of all the books that you had read. Hecker says the studies show, when people believe they're being monitored, they actually do worse.
On the topic of external incentives to do a variety of different tasks, many studies seem to be inconslusive and there are a lot of different variables. There are also a lot of different types of rewards. There are exogenous rewards, which are rewards given outside of "the realm" of the thing you are doing. One such example is the "Book-It" program from Pizza Hut, a program that gives kids pizzas for reading books. Books and pizzas aren't related, so that's an exogenous reward. However, if you got a new book for reading a book, the reward is related to the act, making it an endogenous reward. There's also the matter of whether rewards are expected or unexpected. It matters if they are dependent on performance. It also matters if they prescribe a certain action or merely state that you performed an action (the difference between "You killed five orcs!" and "You killed five orcs just as you were supposed to do!"). And it also matters if the reward is verbal, tangible, or symbolic. With all these variables, the science of rewards gets pretty muddied.
However, the psychologists can all agree on two points:
For interesting tasks:
Tangible, expeted, contingent rewards decrease motivation.
Verbal, unexpected, informational rewards increase motivation.
Hecker only wants to talk about what motivates people to do interesting tasks. There's a lot of research on how you motivate people to do "dull tasks," as well, but he didn't want to cover them in the talk. Our games should be interesting, after all. If you're making an intentionally dull game, but need to motivate people to play it anyway to bilk them out of their money, Hecker says, "I pity you." (This of course got some serious applause.)
Games should be inherently interesting. But if you're rewarding people with some kind of concrete reward in order to play a game, then you're actually discouraging them from doing so. In effect, what should be play becomes "work" because you're getting "paid." I know I've felt this phenomenon acutely in my own artwork... I seem to do better when it's just for fun.
Here's a problem in game design right now: we have more metrics than we ever had before. As a result, game designers may be getting "blinded" by these. The effect of an external motivator on gameplay is easier to measure than the effect of a person's internal motivation. So when we see, in the metrics, that the external rewards added are getting people to play the game more, we're tricked in to thinking this is the right way to get people to play games, which may lead to duller games over all.
That was basically the gist of the talk, which then moved on to audience questions. Some people seemed to agree with the findings, a lot, and others didn't.
I found myself asking... what is an Achievement anyway? Is it a verbal reward, a tangible one, or a symbolic one? Does what it is vary from gamer to gamer? If an Achievement is unexpected, as in, you didn't look up how to get it beforehand, it counts as unexpected. I think the happiest I ever was to get an Achievement was the "Renegade" Achievement in Mass Effect, because it dropped on me spontaneously only after making the Renegade decision in the game's finale and I hadn't been expecting it at all. But if an Achievement is expected, suddenly, you're "grinding" for it... if I had been trying to just fill the bar and got Renegade at a less dramatic moment it wouldn't have been nearly as interesting.
And I also found this whole topic relevant to my work as an educator. Grades, A plusses and B minuses, are certainly extrinsic and contingent rewards. They might be thought of as symbolic, or concrete. Maybe they don't work very well. Grades are, however, a reality of what I do, and even though in a perfect world I probably wouldn't have to give them, I do. Some teachers are now experimenting with things like "leveling up" and "Achievements" in the classroom to replace or supplement grades, and find that this actually works for their classes. Maybe students of the current school generation would prefer filling a bar to being stamped with a letter. But maybe this is just the same thing over again: replacing one external, exogenous motivator with another one with a different label. It seems that though some kind of evaluation is required of educators, true motivation to learn, to grow, and especially to create, can only really come from the inside.
I would say overall that Hecker is right to say the science isn't conclusive, and more of it is needed. It's all definitely food for thought both in game design and in teaching.