Demotivation by External Rewards

Jesse Schell’s “Future of Games” talk at DICE is getting a good deal of buzz and discussion.

So here are a few thoughts: The second half of Schell’s talk centers on external reward systems such as the Xbox Live achievement system, and he describes a possible future (that he appears somewhat ambivalent about) in which everything – TVs, toothbrushes, pianos, cars, schools are designed as game-like systems that constantly reward you with points for various actions. Such as, well, brushing your teeth.

Schell’s basic argument is that external rewards are an incredibly strong psychologically motivator.

Yes and no. If you think about the car that gives you points for a mundane activity such as driving fuel-efficiently, then certainly external rewards can work as a motivator.

But I think that Schell a.o. overlook that external rewards are also known to be strong demotivators. A famous 1973 experiment (“Undermining children’s intrinsic interest with extrinsic reward“) showed that when nursery school children consistently received external rewards for drawing, they lost interest in drawing and began drawing less.

Here’s the graph from a write-up of the results that also mentions similar studies of adults. (Expected reward = consistently receiving a reward for drawing.)

But how can that be? Aren’t rewards motivating? Not necessarily, because rewards may trick you into forgetting your original motivation. You think you like drawing, but when you are consistently showered with rewards for doing it, you start thinking that you are really in it for the rewards …

Marc LeBlanc once pointed me to the business book Punished by Rewards: The Trouble with Gold Stars, Incentive Plans, A’s, Praise, and Other Bribes. From the book description:

Promising goodies to children for good behavior can never produce anything more than temporary obedience. In fact, the more we use artificial inducements to motivate people, the more they lose interest in what we’re bribing them to do. Rewards turn play into work, and work into drudgery.

As you can see, this is completely at odds with the argument that Schell is making. I can’t claim to be an expert on the psychology here, but it does seem that external rewards may have a kind of reversal effect: If you dislike the activity, external rewards make it more attractive, but if you like the activity, external rewards make it less attractive.

So one theory to use could be Michael Apter’s Reversal Theory, according to which we switch between telic (focused on external goals) and paratelic (focused on immediate enjoyment) states. Hence the introduction of an external reward will switch us from the paratelic state where we are enjoying a task, to the telic state where we are performing a task for external rewards.

On a more personal note: I sometimes hear of universities that want to reward researchers for publishing papers or talking to the media. This always strikes me as unpleasant because it discounts the pride that we (hopefully) have in our research. If we started to do research only for the external rewards, our productivity would surely drop.

And that, my friends, is why external rewards can be demotivating.

19 thoughts on “Demotivation by External Rewards”

  1. Like you mentioned, we might benefit from a reward for mechanical tasks that we’d rather avoid (that ties in with a favorite TED talk of mine: http://www.ted.com/talks/dan_pink_on_motivation.html). But, I think games might actually motivate people whether or not they enjoy doing X.

    Take his example of going on a walk to lower your insurance premium. You can look at that two ways;
    1) Your reward for walking is a lower insurance premium
    or 2) Your goal is to improve your help and pay less for insurance, so you walk.

    If a game can change your perspective over time, a task that was a chore might become part of a larger goal and motivating in it’s own right. The same might be said of another example: practicing piano. At first a reward might be a motivating or demotivating but if the game helps you discover that music enriches, any other reward is just a bonus. I’m not sure if Schell eluded to it, but games might have the power to change your perspective and points/score might just serve to remind you of where you are and where you’ve been.

  2. So, in the end, this technique of giving out points for everything will ultimately fail in the advertising world but will thrive with goverment programs like, for example, the one regarding public transportation?

  3. I also came to link to Dan Pink’s talk at TED, but d+pad beat me to it.

    Dan Ariely describes, in Predictably Irrational, a kind of behavioural shift that happens when points/costs are involved that seems similar to Apter’s switch between paratelic and telic.

    He speaks of a daycare that wanted parents to come pick up their children on time, so instituted a fine for coming late. The fine, however, caused parents to arrive even later. Parents began to think of their lateness as “five bucks” instead of “letting down the daycare workers” and “keeping my kids waiting.” Even when the fine was removed, the shift in thinking was permanent: the parents were later than ever.

    So, yeah, sometimes giving things a point value can result in bad changes in behaviour, and those changes can be permanent.

  4. External points systems are basically money. You’re paying people to do things you want them to do.

    If there’s an entertainment factor in receiving the rewards, then it’s a little more complicated. People who spend $50 playing blackjack or roulette or poker in a casino don’t regard it as a failed investment; they regard it as money well spent for an evening’s entertainment. Likewise, people who spend hours playing slot machines aren’t doing it for the reward (if they were, every dollar they paid in would result in a payout of 70c no matter what); they’re doing it because they like the experience.

    A variable feedback reward system to give people money/points for undertaking some action may therefore add an extra layer of entertainment, which could keep interest in higher for longer. I wouldn’t expect it to last forever, though: as soon as people regard it as a right, rather than as a reward, any beneficial effects would disappear.

    Richard

  5. Here’s the comment I made on your FB excerpt before realizing the full blog post was here:

    This sounds like “Punished by rewards” book by Alfie Kohn. It’s a whole educationnal theory based on the idea that external rewards are bad as a method. I dislike external rewards so I would like to believe that study but it’s just one study… And my empirical observation strongly contradicts it: just look at the millions playing WOW or Farmville … See Moreor for that matter, most video games. Or witness the unbelievable power of the external motivator called money that will keep people in jobs they hate all their lives just because it pays well. I think every game designer has had an opportunity to test how placing some external motivators in a weak part of his game just pulls players through. It ‘s artificial, it can even be ethically wrong but unfortunately it works when done right. So coming back to your study (which I haven’t read yet as i am on my iPhone), maybe it’s a question of quantity or pacing. Maybe it’s in the way the rewards are being delivered: handing a salesman his paycheck saying “you’re a winner!” or “here you go, slave” is sure to affect his motivation.

    In any case an extraordinary claim like “external rewards demotivate people” is going to require an extraordinary proof – or at least more than one old study – because it flies in the face of all observable evidence…

    Now that I’ve read your full post I will add that I came out of Alfie Kohn’s book feeling it was weak on science and big on wishful thinking. I also suspect external rewards work best on tasks you wouldn’t really enjoy much doing otherwise.

  6. @d+pad, Kimari, Lucas, Richard
    To me, the clearest example of demotivation by rewards is when you are strongly invested in a specific activity. Because you fundamentally enjoy it (playing piano), or because it is a strong part of your self-image and values (say, making sure that you are doing high-quality research, game design or teaching, or raising a child well).

    Hypothetical example: You see a mother walking with her child. The child suddenly runs from the sidewalk in front of a moving car. You instinctively leap and manage to pull the child to safety. A text message arrives on your phone saying “+10 points for saving the child”. Isn’t there a mismatch between the “reward” and the action which you performed out of your core humanity and empathy? Isn’t the reward basically discounting what you just did? Is the “+10 points” going to motivate you to help other people in the future?

    In such cases, it feels inappropriate if someone gives you gold stars or petty cash bonuses. In fact, in may feel like it is discounting your strong personal convictions and that you take pride in what you do.

    It also seems that while game designers clearly are great at dealing with reward systems, many game designers didn’t exactly choose game design because of the money, but because of a strong personal alignment with game design. So game designers themselves aren’t necessarily motivated by external rewards.

    Perhaps the overall takeaway is that people aren’t simply motivated by rewards (cash or otherwise). We also have things like values and beliefs, and inherent joy from specific tasks. Which means that external rewards isn’t a universal motivation system.

    @Olivier
    Kohn’s book is a popular one (I haven’t read it). The scientific paper I linked to does seem to document fairly well that external rewards can demotivate in certain situations.

    The big question is when exactly that is. How much can we generalize about external rewards?

  7. Oh, here’s an interesting (and heavily quoted) study that I hadn’t seen ( http://psycnet.apa.org/index.cfm?fa=buy.optionToBuy&id=1971-22190-001 ):

    “Results indicate that (a) when money was used as an external reward, intrinsic motivation tended to decrease; whereas (b) when verbal reinforcement and positive feedback were used, intrinsic motivation tended to increase.”

    So there’s external rewards and there’s external rewards.

    But apparently it is a “controversial” point. (http://rer.sagepub.com/cgi/content/abstract/71/1/1 )

  8. I’m wondering if the issue isn’t so much about *external* rewards, but *immediate* rewards. If you saved that kid and got a medal for it some months later, the effect it has on you would be quite different from getting 10+ immediately.
    Getting the 10+ on a monthly report might also be better than immediate.

  9. ‘external rewards can be demotivating’
    There is this famous story about children drawing with a specific marker; the children where intrinsically motivated to draw with the marker. The group was split into two. One group was praised and rewarded for using the markers, while the other group was not. The use (and enjoyment) of the marker declined in the ‘praising group’, while the latter group still enjoyed the markers and kept drawing.

    It was concluded that external rewards ‘killed’ intrinsic motivation. But there can be many other reasons for the decline in motivation. The motivational change can also be attributed to what Bianco et.al (2003) a regulatory misfit, in which the implicit theory of students about drawing did not fit the instruction (or in the case of the children intrusion). Drawing was considered something fun (the implicit theory of children) and transformed by the educators as something of importance (praising and rewarding is important to children).

    The same is done to nowadays serious games or e-learning applications. Instead of focusing on the implicit theory of students about a subject, serious game designers try to ‘fun-up’ exercises that are experienced as something of importance.

    Bianco, A. T., Higgins, E. T., & Klem, A. (2003). How “Fun/Importance” Fit Affects Performance: Relating Implicit Theories to Instructions. Personality and Social Psychology Bulletin, 29(9), 1091-1103. doi:10.1177/0146167203253481

  10. Also, the works of Lepper & Cordova (1996) show that there is no dichotomous relation between intrinsic and extrinsic motivations. In other words, intrinsic motivations do not decline at the same rate as extrinsic motivations (or external rewards) increase.

    Cordova, D. I., & Lepper, M. R. (1996). Intrinsic Motivation and the Process of Learning: Beneficial Effects of Contextualization, Personalization, and Choice, Journal of Educational Psychology, 88(4), 715-730. Retrieved from http://www.sciencedirect.com.proxy.library.uu.nl/science/article/B6WYD-46RVPMV-9/2/89b111b9288c506d06b4ce01b3abde6c

  11. @Menno

    Ah, yes, I was already mentioning the study of the children drawing.

    I find the idea of “regulatory misfit” quite interesting.

    But as far as I understand, there is a continued discussion about the relation between intrinsic and extrinsic motivation – it’s not exactly a settled issue?

  12. If the hunter-gatherer tribe can satisfy itself by taking the fruit of one forest/savannah, it will not move on. Several generations may pass before the tribe gets ‘stuck’ in its environment, with accumulated sacred mythology and totemism, and the environment begins to be ‘used up’ (e.g. North American Dust Bowl or indeed any human-created desert region). The smarter humans invented crop rotation, which is the true key to human settlement – actually a way of reducing immediate rewards!

    Vague hypothesis: Our satisfaction falls off if we are rewarded every time for doing The Right Thing because [human] survival depends on finding new strategies, new environments, new frontiers etc. This will also be true of other opportunist species like rats, foxes, pigeons, racoons etc.

    I seem to remember some experiments with rats in skinner boxes. The rats would press the bar more avidly if the reward came only sometimes. If the reward came every time, the rats became blasé and only pressed the bar when they were actually hungry. If the reward was intermittent, the rats went into a frenzy of bar-pressing. It might have been pigeons.

    I believe this mechanism is very powerful and is the basic ’cause’ of ludomania. Would you get hooked on a game of chance if you knew you could win every time? Scarcity is what hooks the punters, which is also why Burberry upped their prices: The teenage girls buying Burberry scarves for cheap were devaluing the brand.

    BTW if anyone has a foil Charizard pokemon card, let me know. (JK)

    [GODDAMMIT jesper . the spam filter this thing ALWAYS rejects my comments these days, and I have no clue why I should be more ‘fishy’ than the next man. Then the 15 second threshold invariably blocks my second attempt to post. Can’t you whitelist me? :)]

  13. Hi Jesper and Co, I finally got around to listening to this speech, and I have to say, i find it terribly depressing because it moves so far in a direction I do not want to go. I had the same thought as Richard above concerning this all-purpose numerical exchange item, call them points or dollars or clams, they are all currency. Don’t we already do all this sort of thing? People already get paid to do market research, to advertise, to solve crimes, to do all sorts of things. I think his basic theory that we should all be paid for EVERYTHING we do is just weird, and I’m not sure it would be a step forward at all. Why would anyone sit down and work hard at any one thing in exchange for dollars when he can wander around performing meaningless, unrelated tasks casually and be ‘rewarded’ (which is somehow different to paid)? Rather than focusing and developing a profession or tradeskill, we instead are paid to buy things. His marketing-centric approach to the talk really made me cringe. If the ‘thing’ is worth paying money for, I don’t need any other external points system for doing it. If the novel is good, I will read it; if the novel is bad I don’t want to read it just for points. If the airfare is fair, I will pay it, I don’t need points… why not just make the service better or the price cheaper? He wants to layer a second economy over the top of the first one, and the only reason he suggests is to make millions of dollars for (presumably) people like him with the foresight to put these systems in place.

    On another note, I find it interesting that (some) theorists seem to think that everything is becoming a game. A few decades ago, everything was a narrative…

    Next note: I have a different understanding of the recent development in ‘casual’ games, which probably comes from reading your work, Jesper. Rather than this notion of ‘external’ and therefore more ‘real’ rewards in the form of achievement and points, I see that these systems are a means to an end that is alluded to in the comments above. We don’t want points, we want recognition and socialisation. The feeling of doing well at a task is (for many/most people) magnified when someone else (who understands the circumstances) sees you do it. Thus these artificial gamer scores that says to people who can’t possibly witness the event itself. You can basically export the milestones in your single-player experience and share with like-minded people in the gaming community. There’s no commercial exchange happening, only a cultural one. The reality of that kind of value doesn’t come from its location in relation to a fantasy game, it comes from its ability to generate recognition in a community. Its a codification of something that happens anyway amongst a community of people who socialise together. Putting a meta-score in everything in life would devastate our ability to socialise, because we’d be constantly looking for the most efficient way to get more points.

    Players of different sports don’t need to be assigned generic scores that transcend the points systems of their various games in order to be recognised. What they get is money–which brings us back around to the first point. Everyone gets money as a marker of the significance of their professional contribution. Things we don’t do professionally don’t require the same reward structure for very good reasons (especially in our late-capitalist society). Sometimes we DON’T WANT TO BE WORKING. Talk to an ex-WoW player about the achievement system, and how that only magnified the feeling that Warcraft is like a second job…

    Nowhere in this speech did art, creativity, expression, learning or socialisation come up. Of course, its a 30 minute presentation and he can’t possibly cover everything, but I feel like the ‘market’ doesn’t really need help from academics.

    Ok, pulling the plug on this rant here.

  14. @Jesper et al: Thank you all for this useful discussion. I never thought a talk I wrote on the airplane on the way to DICE would get so much attention! By no means do I have any clear understanding of when extrinsic rewards are and are not effective. But as virtual economies can be more and more easily connected to the real world, I feel certain that hundreds of experiments will commence in this space. Seems like a good time for game designers to figure this out! I’m going to check out the Punished By Rewards book — it seems to have pointers to a great deal of research in this area. I had never thought very seriously about this issue before — thanks so much for raising it!

  15. @Jesse Thanks for raising the question – you’ve really got us talking.
    It would be great with some more game-specific research on the subject too.

  16. interesting post, although i’m missing one thing which may play a big role when talking about external awards in combination with the xbox live achievement system (and “such”): making your achievements visible to others. there’s a difference between getting an award “silently” or getting an award which creates a kind of reputation and builds “identity” (by points, kind of award, …). in secound case, a new area opens up when thinking about comparing your achievements or rewards with others and –in an extreme case– judge people by their achievements. would be interesting to hear what you think about it.

  17. @Fabricio
    I agree, there is a large and interesting social component that seems not to be well understood. I am browsing through some psychology readings on the subject, and the article below seems to be relevant, for example. Will post if I come up with something interesting.

    Buunk, A. P., Cohen-Schotanus, J., & van Nek, R. H. (2007). Why and how people engage in social comparison while learning social skills in groups. Group Dynamics: Theory, Research, and Practice, 11(3), 140-152.

Leave a Reply

Your email address will not be published. Required fields are marked *