Most games use a sequence of little rewards to nudge players around. This is very different from most salary and hiring and review structures, which are big rewards in big intervals. Much has been made about the power of the games approach and how it should be used everywhere. To date (I haven't been looking very hard), I'd not seen a hard analysis of how a sequence of little payments might affect the quality of decisions. Do they help people solve new problems? Do they help people remember what they are doing, and transfer the learning to other situations?
Here's a little experiment by economists Neilson, Price, and Shor that comes to these conclusions:
- If you give micro rewards when people are doing the right thing, they remember the right solution for the current problem and transfer it to similar problems. But they don't learn how to solve new problems.
- If you give micro rewards when people are doing the wrong thing, they get better at figuring out new problems, because they learn more about the dimensions of the problem (including bad strategies). But they don't remember much about the current problem and how to solve ones like it.
Comments on Micro rewards have weird effects on decisions:
It'd be nice to think this could be transferrable to game design, but there are lots more reasons to give players lots of little rewards in games, and breadcrumb around the content. I think we're well past the point when most of the people playing all kinds of games want to feel bad about not being able to solve a problem. In the workplace, you're not solving problems for fun and escapism; it is *your job*. Reward systems are important there for the sake of defining the work environment, where "self-direction" is a workplace culture decision, built out of relationship between management and employees.
But besides all that, what are the alternatives to small rewards in game design? What sort of rewards are we talking about? Rewards can be anything from "stuff" in the game world to pleasant audiovisual cues to more exposition from an especially interesting character.
Posted Nov 30, 2012 11:07:11 AM | link
RTA. The researchers exclusively address extrinsic, tangible rewards in their experiment.
One possible takeaway is to rediscover the potential value of intrinsic rewards like "satisfaction", "amusement", "empathy", as complements to mere Pavlovian reinforcement scheduling - which, this single, non-definitive but provocative study suggests, may not be as effective as we tend to assume.
Another possible takeaway is to rediscover the value of research and evidence-driven design, vs being guided purely by one's own instincts, assumptions and personal preferences as a designer - a particularly useful reminder the longer one has designed.
Finally, this kind of research is particularly important, as Ed mentions, in the context of "gamification", which is often reduced to mindlessly and mechanically adopting Pavlovian game mechanics to "realworld" work, dehumanizing employees and missing all the engaging, stake-building, community-building, flow-inducing wonder proper playful design can spark.
Posted Nov 30, 2012 12:25:59 PM | link
This is an interesting concept. I work in recruiting/HR, so I'm definitely aware of how compensation is set up and evaluated, and I agree that the intervals are often quite long between any type of monetary reward, unless you're working for commission. And I also play games, and notice what you're talking about with micro-rewards coming at much shorter intervals.
This really could be used in the corporate world. I had never thought of it. Of course, the drawback is that pay raises and additional bonuses take time to calculate and execute, for payroll and other departments. So paying out a tiny raise every month is a lot more difficult than bumping up somebody's base salary at the end of the year.
Posted Dec 1, 2012 11:58:02 AM | link
Sorry for the shameless plug guys - I love your blog and scoured it for sometime for a contact us link - maybe I'm just blind? But I wondered if you would do me a favour... I have a project on Kickstarter which I'd really like your support for - the project is just the tip of a far bigger iceberg which I'm hoping to strap a motor to a sail all the way to Gamesville. I can't believe I just wrote that. Too many sleepless nights. Anyway, anything you can do to help spread the word about my Kickstarter project would be amazing. Just think of the games that could be made off the back of this tech! thx! (although the clone wars might have to be strategically dealt with).... :-)
Posted Dec 2, 2012 9:31:47 AM | link
Doesn't it depend on the nature of the reward? Intrinsic rewards (ie. the reward is that it was fun) are not the same as extrinsic rewards (ie. the receipt of the reward, but not the reward itself, was fun). In games, the traditional approach is to aim for intrinsic rewards and use extrinsic rewards as a booster to multiply the intrinsic reward that had already been received. You feel good because you killed the dragon, but you feel better because you killed the dragon and the game bowed down before you through its use of anthemic music.
If there is an intrinsic reward, the research you cite doesn't seem to apply: you're not "giving" rewards, because people are getting them all on their own - they enjoy doing what they're doing. It's all about extrinsic rewards, which ultimately come down to bribery: do this and I'll give you this. This can have well-known problems in addition to the ones the article cites.
The first point in the article (bribing people to do the right thing will make them learn the right thing but not necessarily in a transferable way) seems obvious. I wonder, though, if the second point raised (that bribery for doing the wrong things increases the ability to learn new things) is a long-term effect? It seems to me that once you learn you get rewarded for making mistakes, "making mistakes" will eventually become the "right" behaviour. People will settle down into making mistakes that don't lead anywhere except to a reward, which brings it into the same category as the first point addresses.
It reminds me of one day at school when the playing field was so covered in rubbish that we all had to go out and come back with at least 10 pieces of rubbish to tidy it up: the first piece of rubbish I picked up was a sheet of paper, that I tore into 10 pieces. I got the reward, but not in a meaningful way.
PS: I'm not the bot who posted the post above this one...
Posted Dec 5, 2012 3:03:01 AM | link