Intrinsic Motivation and Gamification

Quantified Self QS Open Forum
1

Hi there!

I’m super excited about this forum. Thanks for starting! :smiley:

My question is regarding getting users to stick with a health/wellness app (or other health tech solution) long enough to cause a change in habits. There are numerous apps out there, and many offer rewards and use gaming to get people on board. The problem is that many people don’t stick with it long enough to change bad habits or create better ones.

So…can we use rewards and play while building, not sacrificing, intrinsic motivation?
Are there any apps doing this successfully?

1 Like
2

Hi Karolina,

This is a great question to pose to the forum. Stickiness is a big issue right now with the push towards gamification that has taken over the mindset of a lot of health/wellness app developers. I whole-heartedly agree that deploying reward systems is a great method to engage a user and start them along a path towards better health and wellness decisions, but it probably will not be sustainable over a long period of time.

Intrinsic motivation is always a great goal to work towards, but it is very difficult to “build” systems for intrinsic motivation. Funny thing, as soon as you try and make someone intrinsically motivated you automatically move backwards towards extrinsic methods. I think that best thing we can do as designers, developers and behavior changers is to try and design a roadway, a path or sorts that provides individuals the ability to learn more about themselves, their behaviors, their abilities and how those all interact to influence their decisions.

I think this is a great start to a wonderful discussion on motivation and how to build systems for positive change. If others have examples of apps please share them!

3

Hi Karolina - you posed your question in a way that suggests competition between intrinsic and extrinsic motivation. This is the perspective of Alfie Kohn, who wrote Punished By Rewards, and, if correct, it suggests that current ideas about designing apps for behavior change may be deeply misguided. Here’s a thought experiment, though: imagine designing for people who are already motivated, in a general sense, but who nonetheless struggle. I wonder what would happen if you used intrinsic motivation as the conceptual foundation of the design, and worked mainly on addressing factors that make us stumble despite being motivated.

1 Like
4

hi, social participation is a strong power as one can see from facebook etc. however when it comes to wellness issues it needs extra force. I see from our health net in Israel (CAMONI) that personal acquaintances has a major supportive force and we are trying these days to establish a “mentor” role in the community.

5

These are all very interesting points.

Is it, or should it be, technology’s role to build intrinsic motivation? Maybe not. Like Ernesto pointed out, it can be a great tool to guide users down the right road (solid social support networks, confidence, the right knowledge, etc).

I still wonder whether continually offering extrinsic rewards when it comes to health wellness does more harm than good. This goes back to Kohn’s perspective. Rewards should guide users to the point where they want to do more because they feel benefit from the activity/process itself, not because they’re looking to gain another extrinsic reward after achieving a particular goal.

I think there are now 2 questions: 1. could/should we design technology to be a primary driver of intrinsic motivation (is this its best use in getting people healthier)? Or is it better used as a support and guide tool?

  1. Are extrinsic rewards doing more good or harm in the long-term in getting people to exercise, eat healthier and go to the doctor?

On the second question, I think gaming without actual reward (other than winning or helping your team) works better than material reward.

6

It’s important to avoid to be contrarian for it’s own sake. The idea that extrinsic motivators are always bad is very seducing.

Rewards are quite tricky. Sometimes they work, sometimes they don’t.
http://scienceblogs.com/cognitivedaily/2009/10/what_does_it_take_to_get_kids.php is an article where children increased their intrinsic motivation to eat a certain food when the got extrinsic rewards for the food.

7

I also think we should ask about the definition of “work.” The idea that rewards produce short term changes seems well founded. But where is the evidence for lasting change?

8

[quote]The idea that rewards produce short term changes seems well founded.[/quote]The idea that rewards produce positive changes as long as those rewards continue seems to be accepted by everyone.
There a question of whether positives behavior changes stay after you stop the rewards, whether something changes about your intrinsic motivation.

Whether the change of your intrinsic motivation is permanent is a separate question.

9

Gary, Ernesto — I too am more inclined to use technology to help intrinsically motivated people address the issues that cause them to stumble. However, despite what behavior experts like Kohn say, one hears much more about companies (and VCs) with technologies that are meant to motivate people via extrinsic rewards. Why do you think this is so? Are they all misguided? Or do they have valid reasons to discount the theories of Kohn and others?

10

I think that the there are so many companies and startups circling around implementing extrinsic rewards because is is the “low hanging fruit” in the world of behavior change. If you want to see immediate results then you automatically graviate towards using reward systems. I think this is still a misguided approach as it undermines the fundamental processes that can help a person learn and achieve real constructive behavior change.

11

I think we are evolving a “counter-conventional wisdom” in this thread - Raj and Ernesto and I are importing into a dialog that has been ongoing at the Bay Area QS meetups - and for this reason I really value Christian’s different perspective. It’s through really exposing our preliminary ideas and admitting we don’t have a certain answer that we could discover some new approaches or ideas to test. A few questions that have emerged for me:

  1. What are the differences between types of rewards? One obvious difference is between consequential and non-consequential rewards. If you need food (and your hunger grows with time between eating), a food reward will keep working, right? No behavior, no food: that’s powerful! Money probably has some of this affect, under conditions of money scarcity. But badges are different, right? Your hunger for badges does not necessary grow with time between badges.

  2. Are there known classes of behaviors that are more controllable with rewards? A lever press by a pigeon - or a click on a mouse by a person - is an action that takes minimal effort. The impulse and the execution are close together. Depositing money in a savings account, say; or cooking a healthy meal; these are complex actions that require more than an impulse. What’s needed is a “frame of mind” in which the rewards are sort of incorporated into the action. Have behaviors been classified according to their responsiveness to rewards?

  3. What has been learned from truly successful efforts to change complex behavior? Ernesto and I talked on the way down to the Bay Area QS Show&Tell last night about the anti-smoking campaign in the US, and especially in California, which led the way. (Ernesto, you can provide more details if you like.) We remarked on how much had to align to make this work: law, culture, tax policy, negative messages, positive assistance for people quitting. It was a giant, multi-decade alignment of forces. On the way back from the Show&Tell, Seth Roberts tagged along in our car, and we talked about food. Food is a lot more complex than cigarettes. For instance, you have to eat food! Also, you can sequester basically ALL the packs of cigarettes in the state behind retail counters, and demand identification from people trying to buy them (for a high price). Just stating these factors makes it clear how much more complex behavior change is when it is applied to something like diet. Still, maybe the smoking campaign has something to teach us about the efficacy and limits of rewards.

12

I am really interested in this subject and I would like to know what apps and systems people use that do this well.

I use and really like 750words.com and healthmonth.com and I like the daily tweets from @habitdesign

Over time, long term what has worked for me has been some kind of social contract to do something and having to be accountable to someone specific that I have some kind of relationship with. Is that an intrinsic or extrinsic motivation?

I think the most powerful reward system would be for the app, or system to show me feedback loops that I might not be as aware of on my own.

13

I agree with Gary that there are some problems with current terminology around this. I suspect partly because if we characterized these things accurately we’d have even more awkward questions about the structure of society to answer. A couple of questions that I think are interesting to ponder:

  1. Are intrinsic and extrinsic motivations necessarily mutually exclusive?

  2. What qualifies as a reward? Gary’s food example seems more like coercion to me. Is there a difference between coercion and reward?

  3. Are systems of extrinsic motivation even possible in the absence of intrinsic motivation? (this is essentially a followup to #2).

  4. How much do we have to consider the relationships between the creators and users of a tool when we consider the reward system? Can the same tool have a different reward structure depending on who provides it and who uses it?

  5. What role does trust play in motivation?

I like Ioan’s point about ‘feedback’. It seems as though there’s fruitful thinking to be had in drawing out the distinction between feedback and reward. I don’t think they’re always exclusive, and this is a key design center.

It’s been shown that biofeedback can allow people to develop regulatory behaviors that persist in the absence of the feedback mechanism. It seems as though this is an example of intrinsic motivation being supplemented by technology without it being a ‘game’ or extrinsic reward system. Could biofeedback be the template for responsible behavior change systems?

  1. Actual computer games require a suspension of disbelief to accept the ‘rewards’. Are there any examples of successful use of gamification in context that aren’t actually games?
14

Great post, Robin!

I think in the idea of “gamification”, sometimes we forget the starting point of the buzzword: Games!!! In a game you:
a) perform a task
b) get feedback on it
You get feedback in very rich ways that make it compelling, and keep you doing it. And you not only perform the task that the game is asking you to do *without *extrinsic rewards- you pay the game industry for the privilege of doing it!

It may be argued that tasks in games are inherently more “fun” than real life. But PLAY angry birds for a while. Or become a master bartender in Fable. What are you actually doing? Clicking a button over and over again when a moving dot gets to the right part of the screen. Or dragging your finger on your phone. Absent the feedback, honestly, I’d rather be flossing!

Because of this, I think the key to creating healthy rewards is feedback- which requires
a) Very good ways to gather data around the habit you want to change (cue cool new biometric sensor devices! Basis, GreenGoose, Proteus)
and
b) A good feedback system for the user.

Take, for example, the “leaf” displays in hybrids- just by showing driving efficiency with clever, meaningful graphics, people become more efficient drivers.

The thing is, we have decades of industry experience behind learning what feedback to give users to change their habits, and now we’re getting to the point where we can also easily gather user data around interesting habits without requiring that they enter it themselves- instead of re-inventing “gamification”, why not just go back to the source and learn better lessons from it?

Yes, I didn’t mention anything about social networks here :slight_smile: Too much for one post!

15

I think the counterpoint to ‘feedback’ might be 'context.

When you talk about flossing vs angry birds, I think most of the time I actually would rather be flossing.

This isn’t because I inherently dislike little computer games. It’s because there’s usually something in my life that I’d rather be doing.

If I was waiting for my flight to be called at an airport, I’d probably find that playing Angry Birds made the time pass much more pleasantly than anxiously watching people mill around. It would be because the context was less pleasant than the game.

Similarly, at certain points in my professional life, spending time on Facebook seemed like fun. Now I can’t stand it. I don’t think it’s because I or Facebook have changed all that much. I think it’s because there are now things in my context that I recognize as much more appealing.

So, feedback is necessary for an engaging system, but the context determines the quality of the engagement.

To me this suggests that perhaps behavior change attempts might be more effective if the systems aren’t focussed on strengthening a behavior pattern, but are designed to help the user change their context.

In my mind, the context is a combination of your identity and your environment.

I think 750 words is a great example of this. It uses a constraint and progressive feedback to help you write. But I think what’s critical about it is that the achievement is real and in the world. After you’ve written the 750 words, you’ve actually written something. The system draws your attention to this, so that you recognize the achievement, but there is another factor - you have the writing itself to prove to you that you are now more of a writer than you were the day before. The system uses feedback to provide a smooth path of behavior that results in the user changing their own context.

16

Gamification has a dark side. Our current knowledge of reward pathways in the brain suggests that all rewarding activities and substances activate the same dopamine-mediated nucleus accumbens/locus ceruleus pathways over different time-courses.

Want to make an otherwise self-respecting person do degrading things? Get them hooked on crack and provide it after each desired activity. Even better, provide it only 60% of the time. Activity -> feedback (reward)… Activity -> feedback (reward). Even better if it’s a really efficient process and easy to get the reward. Works for crack cocaine, works for Angry Birds/Farmville.

If the activity is “desirable” it’s easy to justify this gamification strategy. But ultimately you’re just getting people hooked on the reward, and other research suggests that once you train the addiction pathways, they will replace one rewarding stimulus for another. i.e. people stop drinking, and increase their smoking (just visit an AA meeting); stop smoking, and increase their carb binging.

Drugs, work, food, whatever.

So I would agree that the more interesting and perhaps ethical approach is to take intrinsically-motivated people and help them achieve their goals better.

If not intrinsically motivated, then I suggest a skilled practitioner of Motivational Interviewing to cultivate their ambivalence and develop intrinsic motivation (one way or another) over time. And for this I think people may be more effective than a machine solution.

Just some initial ramblings off the top of my head, but this is a great topic.

Paul Abramson MD
@paulabramsonMD
@quantdoctor

17

This a topic I’m fascinated with. I am also constantly pushing everyone to read Alfie’ Kohn’s Punished by Rewards for a good rundown of most of the twentieth century existing research on rewards and punishment. It helped me gather my thoughts and begin to understand why rewards and gameification (which has become shorthand for rewards) makes me so uncomfortable. He begins to get at the basically transactional view of the world and of human nature that underpins faith in rewards and punishment.

The idea that all our motivations can be externally observed and quantified seems both cynical and simplistic to me. I think designers (and psychologists) who hold such a sour opinion of their users will see in response a skeptical backlash. Rewards are inherently coercive and we sense it even if we can’t articulate it. We don’t like to be condescended to or managed, and trying to externalize and manhandle all our idiosyncratic and weird motivations—especially the most fragile and complex like our curiosity or drive to create—risks breaking some ephemeral contract between both ourselves and others. It propagates the dreary notion that everybody’s got a price.

I don’t think there’s an easy answer, but I do think that a reasonable place to start looking is feedback as distinct from rewards. A big part of succeeding there may have to do with resisting the temptation to have the technology ascribe meaning to what’s being seen and letting the individual determine the meaning and value of the feedback. That’s where you start to see epiphanies, the ultimate intrinsic motivation. I also think it’s worth thinking about the state of play; that is, the intense and absorbing state of play that children slip into that’s something very different from a cinematic version of “fun.”

Whew, good point Paul. So in terms of the desirable physiological response in my users (ha!), my Healthy Behavior Change app could look a lot like a ProAna trigger site.

18

I completely agree! What we usually look to as “rewards” are needed because we can’t tell that the good habit we’re trying to pursue is actually good for us until months or years down the road… tightening up the feedback loop and presenting data in a rewarding manner (again, looking to games for great examples on how to parse and display data) can reduce or eliminate the need for extrinsic rewards. If you’re interested, I did a presentation on this at work about a month ago:


(you can see my script/presenter notes if you click “view on slideshare” and then click on “Speaker Notes” next to the comment tag)

19

There has a been a lot of talk surrounding the differences between feedback and rewards and intrinsic vs extrinsic motivation. I am going to take this opportunity to put my academic hat on and try and provide some information regarding motivational theory. Most of the motivational work comes from a group of four major players in the field of behavioral psychology:

[list]
[]BF Skinner - Operant Conditioning
[]Albert Bandura - Social Cognitive Theory & the power of Self Efficacy
[*]Edward Deci & Robert Ryan - Self Determination Theory
[/list]
Because we’re talking mostly about motivation - why people do the things they do - I figured I would start with my favorite theory. Developed by Deci & Ryan, Self Determination Theory (SDT) “represents a broad framework for the study of human motivation and personality. SDT articulates a meta-theory for framing motivational studies, a formal theory that defines intrinsic and varied extrinsic sources of motivation, and a description of the respective roles of intrinsic and types of extrinsic motivation in cognitive and social development and in individual differences.” It is a great place to start framing what we mean when we talk about motivation. I tend to gravitate towards SDT when discussing motivation because of their methods of elucidating the difference between extrinsic and intrinsic, but also because it teases out differences within those two constructs. So let’s start with the different types of motivation

**Amotivation **
Amotivation is referred to as a state of motivation during which the there is no intention to act. There are four types of motivation proposed by another great researcher, Robert Vallerand
[list=1]
[]Individuals believes they do not have the capicity to perform the beahvior successfully.
[]Individual believes that the methods to complete the behavior will not lead to the desired outcome
[]Individual believes they do not have the capacity to put forth the effort required for success.
[]Individual feel helpless and inconsequential compared to the action.
[/list]
Extrinsic Motivation
Extrinsic motivation is defined as participating in an activity in order to gain some separate outcome. According to SDT extrinsic motivation is split into four distinct categories: external, introjected, identified, and integrated regulation.
[list=1]
[]External regulation is the most extrinsic of the four and is also the most nonself-determining. It is commonly referred to as the regulation of behavior through external rewards and contingencies.
[]Introjected regulation begins the movement towards internalization of the regulation of behavior. It is more stable than external regulation but still shares many of the same attributes. In external regulation, rewards and contingencies are brought about by a source separate than the individual, while introjected regulations relates to how the individual controls the rewards for him or herself. Deci & Ryan believe that the introjected regulation lies within extrinsic motivation because it creates tension within the individual.
[]Identified regulation, one begins to act out of choice. There is a choice to act because the action is considered to be important by the individual.
[]Integrated regulation an individual that displays integrated regulation often makes free choices to engage in behavior with knowledge of both the consequences and rewards that are attributed to that behavior and a willingness to accept responsibility for their actions.
[/list]
Pause here before reading on about intrinsic motivation. One of the most important things to understand about SDT is that the above mentioned types of extrinsic motivation operate along a continuum from motivations that result from completely non-self directed behaviors (“I do this because you gave me that reward”) to almost fully autonomous behaviors (“I go to the gym because it’s a good way to meet people”). One could even make the case that as you move towards identified and integrated behaviors you are in a gray area between extrinsic and intrinsic motivation.

Intrinsic Motivation
The development of self begins with intrinsic activities. The characteristics of intrinsic motivation includes but are not limited to, the absence of an external reward, optimally challenging activities, and the undertaking of interesting activities. Intrinsic motivation has also been split into three categories: to know, to accomplish, and stimulation
[list=1]
[]Intrinsic motivation to know is defined as engaging in an activity for the satisfaction experienced through learning, exploring, or understanding something new.
[]Intrinsic motivation to accomplish is concerned with engaging in an activity for the pleasure associated with the attempt to accomplish, create, or to surpass previous standards. This is closely related to the challenge of mastering a certain behavior, not in the end result of an activity.
[*]Intrinsic motivation for stimulation is operating when an individual engages in a behavior in order to feel enjoyable sensations like excitement.
[/list]
When I listen to people talk about games, feedback, badges, points, behavior change, etc I alway looks back on my nights in a tiny room reading paper after paper about SDT and the continuum of motivation. I think it is really useful as a framework that you can use to understand motivation and the methods we use to try and get people to engage in any behavior.

For those of you who want to read more about SDT I highly suggest the following articles:
[list=1]
[]Deci, E.L., & Ryan, R.M. (2000). The “what” and “why” of goal pursuit: Human needs and the self-determination of behavior. Psychological Inquiry, 11, 227-268.
[]Ryan, R.M., & Deci, E.L. (2000). Self-determination theory and the facilitation of intrinsic motivation, social development, and well-being. American Psychologist, 55, 68-78.
[*]Vallerand, R.J., & Losier, G.F. (1999). An integrative analysis of intrinsic and extrinsic motivation in sport. Journal of Applied Sport Psychology, 11, 142-169.
[/list]

20

Ernesto — thanks for that great summary of the continuum of motivation, including amotivation.

I found Daniel Pink’s book “Drive” to be great introduction to these ideas. As a non-expert, I found it much easier to appreciate Deci, et al’s work, after getting a lay-language introduction from Drive. What did you think of Drive? Is it a good summary, or is it perhaps too simplified?[hr]

I think there may be much value to be gained from all these “gamification” efforts, even if one feels that “motivating” people extrinsically is ineffective or worse. Games can be fun, joyful. When we play family board-games, the biggest pleasure comes simply from doing the activity together. But if that was all, any game would do, and we wouldn’t be so particular about which games we choose. In fact, the game quality — the overall game play, the artistic beauty of the game materials, the quality of the cards/tokens/etc. — make a huge difference.

One could argue that these are all efforts by the game designer (and their marketing folks) to motivate us to buy, play and tout their game. So, these are all efforts to bribe us into doing something they want us to do — imposing extrinsic motivations.

Or one could look at this as the same game designers & marketers doing their utmost to satisfy their customers’ desire for a great game-playing experience. Viewed this way, they are supporting our intrinsic desire for fun, family activities.

Similarly, if these gamification efforts lead to products that make what we already want to do for our health much more pleasurable, that would be great. Framing and intent are important. Coercion is bad. But adding joy to the world has to be good.