Imagine it's the 1960s. Got it? Okay. You look good in those clothes, actually. Now further imagine that you're a student at the Carnegie-Mellon University participating in a psychology experiment. The researcher presents you with a small pile of plastic shapes that can be fitted together to make bigger shapes. He then shows you drawings of such forms and asks you to recreate them. The first one looks like a duck. (This detail isn't really important or even probably true, but I like ducks, so...) The experimenter says he's interested in how many puzzles you can solve within 13 minute chunks of time.
Around the middle of the session, the experimenter excuses himself form the room, saying he will be back in a few minutes and that you're free to amuse yourself however you like. You notice that there's a stack of magazines on the table next to you that includes, weirdly enough, a copy of Playboy. (Unlike the duck thing, this detail actually IS true. I know, right?) There are also still the puzzle pieces and drawings. What do you do? Specifically, how much time do you continue to spend working on the shape puzzles until the experiment resumes?
It turns out that the answer to that depends on whether or not you were being paid to make the shapes.
The experimenters found that when they did not offer payment, subjects spent on average about 248 seconds (over 4 minutes) of their unsupervised time to voluntarily keep working on the puzzles. We can assume that their motivations for solving the puzzles were intrinsic --they was internal to the person and for the sake of the activity itself. Maybe the subjects found the puzzles somewhat fun or interesting.
Then the experimenter told some of the subjects that they'd be paid $1 for every puzzle they solved correctly. This introduced an extrinsic motivator --the cash was external to the mind of the person doing the task. This time during their unsupervised time subjects spent an average of 313 seconds (well over 5 minutes) solving puzzles. A 26% increase!
Now, the kicker. For final puzzle solving session, subjects were told that there was no more money to be had, so the $1 per puzzle reward was retracted. Subjects' average free time spent on the puzzles plummeted to 198 seconds --not much more than 3 minutes. That's 20% lower than the 248 seconds spent before cash payments were even brought into the conversation.
The TL;DR version:
Offering an external reward and then taking it away is sometimes worse than never offering it in the fist place. It's something psychologists call "the overjustification effect" and it has been found in various other studies as well. What's more, I think it happens sometimes in video games.
Take achievements, trophies, and badges for example. These are generally considered to be external motivators. Unlocking an achievement on Steam or the Xbox is a little reward that's generally independent of the game, even if you had to do something in the game to get it. So are systems like gamerscores or percentage of trophies unlocked.
I asked Michael Hanus, a researcher and Assistant Professor at University of Nebraska if this was a reasonable stance to take, and here's what he said:
It's pretty safe to say that achievements and trophies are extrinsic motivators. They're coming from something outside of you. The game designer is making these, you're playing the game, and the game is giving them to you in terms of goals. And the rewards themselves are extrinsically motivating. If you do this thing, you'll go and achieve this outside reward given to you by someone else.
(That was part of a larger conversation Hanus and I had about the motivating effects of achievements. Hear the whole thing in that episode of the podcast.)
Whether an external reward will trigger the overjustification effect depends on a few things. The person must be intrinsically motivated to start with, then she must start receiving an external reward that gradually takes center stage in her mind. It also helps if the external reward comes at first as a surprise.
Another factor that's important is whether the external motivator is seen as controlling and even manipulative rather than simply informational. It matters if a game tells you "Hey you aren't playing right if you don't aim for this achievement." That's controlling and hurts intrinsic motivation. But if it tells you "Hey, you've been doing your thing and that earned you this achievement" that's informational and should increase intrinsic motivation.
Hanus puts it this way:
The way that [researchers] talk about rewards is that they're broken up into two things that you think about. Are they controlling me? Are they telling me what to do? Or are they supporting me and giving information that says "Hey, what you've been doing is a good job. It's along the lines of what you already wanted to do internally for your own reasons." So how you see the reward in that given moment depends on whether you take that as an internal or external motivator.
The way that a game constructs and presents its achievements could have a big impact on whether they actually help or harm a player's motivation to keep engaging with the game. Because like the subjects in the study above, when the extrinsic motivation goes away (because you got the achievements or you beat the main quest line) a player's motivation to keep experimenting and noodling with the game --things they loved to do in the beginning-- may shrivel up when there are no more external rewards to do so. Trophies and achievements may actually hurt in the long run.
While this has been studied extensively in other contexts it hasn't really been studied enough in the context of games or gamification. But it seems a very reasonable hypothesis to put to the test given how ubiquitous achievements, trophies, and badges are in video games.
Deci, E. L. (1971). Effects of Externally Mediated Rewards on Intrinsic Motivation. Journal of Personality and Social Psychology, 18, 105-115.
Deci, E. L., Koestner, R., & Ryan, R. M. (1999). A meta-analytic review of experiments examining the effects of extrinsic rewards on intrinsic motivation. Psychological Bulletin, 125, 627-688.
Hanus, M. D., & Fox J. (2015). Assessing the effects of gamification in the classroom: A longitudinal study on motivation, satisfaction, effort, and grades. Computers & Education, 80, 152 - 161.
Lepper, M. R., Greene, D., & Nisbett, R. E. (1973). Undermining children's intrinsic interest with extrinsic reward: A test of the "overjustification" hypothesis. Journal of Personality & Social Psychology , 28, 129-137.