Dopamine and gamification

It's been a bit crazy in the gamification world with GSummit and CHI coming one after the other, but I'm back and have been meaning to post on a topic that came up again and again at the GSummit.  So often, in fact, that I started to think of the old drinking game Hi Bob because people were saying dopamine in practically every talk.  It's funny to me, because my degree is in Cognitive Neuropsychology and as part of my qualifying exams, I had to map out the entire dopaminergic system.  In fact, I might still have the notecards I drew out to study from in my garage someplace. Dopamine (and really all neurotransmitters involved in behavior)  is a topic I studied a lot.

For those of you who aren't following why dopamine kept coming up at the GSummit, here's a little background.  Dopamine is a neurotransmitter, or the chemical juice that lets one neuron (nerve) talk to another.  The neurotransmitters break down into a set of categories, usually based on their chemical structure.  The most common groupings are the amino acids (like glutamate, which is not the same as monosodium glutamate, but that's someone else's blog), peptides (like Substance P, a very science fiction-y name) and monoamines (like dopamine, histamine, and serotonin).  Here's a lovely set of dopamine molecule earrings from our friend at

The brain, and the body in general for that matter, is fairly parsimonious with neurotransmitters and the same neurotransmitter might be involved in multiple very different systems.  Dopamine is a prime example--it's the main neurotransmitter of the motor system (loss of neurons in the nigrostriatal pathway is the primary problem in Parkinson's disease) but it's also one of the main neurotransmitters of the mesolimbic system (the other is serotonin).  The mesolimbic system is sometimes described as a "reward" pathway, which is why it kept coming up at the GSummit.

Why is dopamine maybe involved in reward?  Well, that's an interesting question. Back in the 1950's, a couple of researchers (Olds and Milner) decided to see what would happen if the electrically stimulated a region of the brain in the mesolimbic system called the nucleus accumbens when a rat went into a certain region of their cages.  Turns out, rats liked it.  They liked it a lot.  So much that when the experiment was adjusted so that they could press a lever to get the little zip of electricity, they'd do it as much as 1700 times an hour.  And the nucleus accumbens is one of the structures in the mesolimbic system that uses dopamine to communicate.  Initially, this area was described as a reward center and dopamine was hypothesized to be the neurotransmitter of pleasure. More evidence came from the discoveries that dopamine levels increase in the brain when addicts take drugs like opiates, cocaine and amphetamines.  When dopamine is blocked in the brain by 99%, rats stop eating, which led researchers to hypothesize that they no longer derived any pleasure from eating.

But of course, nothing's ever quite that simple in neuroscience.  Further research on the rats who had all their dopamine suppressed concluded that it wasn't that the rats didn't take pleasure in eating, it was just that they had no desire to eat on their own.  If they pushed food on the rats, they concluded that they seemed to show pleasure while eating (though how they conclude that sort of escapes me) but that left to their own devices, they simply wouldn't eat.  They had lost the motivation to initiate eating. 

"Dopamine-depleted rats still ‘like’ rewards, and still know the rewards they ‘like’. They simply fail to ‘want’ rewards they ‘like’."

Berridge and Robinson, 1998

Further experiments with people confirmed the idea that dopamine might not be about pleasure and reward as much as it created a wanting or motivation to keep seeking out stimulation. There is some evidence that dopamine levels increase when a reward is greater than expected, which then is hypothesized to increase drive or motivation to achieve a reward.  Lack of dopamine meant there was no motivation or drive to do something.

So the role of dopamine might not be in pleasure/reward after all but in motivation and drive.  Still pretty important issues for gamification, just maybe not in the way some folks might have thought.


While I am sort of relieved to see someone with an actual science background bring this up, I am still deeply disturbed that so many people see the gamification/dopamine discussion as a GOOD thing. We are doing nothing more than hacking known vulnerabilities in the brain, in service of getting people to *want* something, and the holy grail of marketers seems to be that rat pressing the button hundreds of times to get the dopamine hit.

I used the word "we" because I was one of the people doing precisely that for many many years. Games, gamification, made no difference to me as long as I could justify it as "providing users with a little more enjoyment." Behavior manipulation through any mechanism is inherently dangerous and potentially unethical. At the least, I wish the conversation around dopamine and it's role in getting people to *want* would be balanced by an equal amount of science-rich discussion on all the potential downsides, side-effects, long-term problems, and yes -- ethical concerns-- around this.

Traditional advertising and marketing was much much weaker than what today's gamification is capable of. Even a poorly-implemented Skinner box carries more behavior influence than a well-designed visual / passive advertisement, no matter how evocative and persistent. If operant conditioning (using +r in particular) were NOT so powerful, it would not be the common thread for conferences on gamification.

We can tell ourselves a story about how this is ultimately good for people, that they "enjoy" it, that you cannot get someone to do something they otherwise would not have *chosen* to do, etc. but anyone making those (false) claims has no business discussing dopamine. That said, I am not directing this at you because you are at least making a start on looking more deeply. But we must quit stopping the conversation at the the technical facts of the dopaminergic system and keep going straight on through to both the short-term and long-term impact.

Given your neuroscience background, then I assume you also know that the reward system (as neuroscientist Eagleman refers to it) is cognitively expensive. It nearly hijacks the ability to do cognitive processing. That works well for marketing (if you don't care about the ethics) but is counter-productive in the long run (and sometimes short) for things that involve deep thinking and learning or... As we now know ... USING WILLPOWER. Anything that burns through cognitive fuel the way the dopamine reward system does is burning scarce, precious, finite cognitive resources. We should think very very carefully about where we want those cognitive resources to be spent, and look at whether dopamine is really what we should be celebrating. But it does work well for giving the short term appearance of "high engagement".

There is a reason almost nobody plays the same games forever. The most enduring, throughout the ages, have been the ones least tied to operant conditioning... Like chess and go. Gamification would be in far better shape if it were to use not VIDEO games (though well designed ones have the right elements, they just also happen to have surface elements that -- when cargo culted -- produce the opposite effect of a good game) but enduring games that don't have those surface dopamine-laden mechanics. Actually, gamification would be in far better shape if it looked to some sports like... martial arts. For example, there's evidence that people who participate in martial arts are more likely to sustain participation in an exercise program than people who are NOT participating in an intrinsically rewarding physical activity but are instead doing extrinsically motivated physical exercise. In the long run, we are not doing people any favors by hacking their brain to drive behaviors. But it is certainly faster to get "engagement", if you are not concerned about the long term implications. And if we think that it's OK because we are simply using it for "onboarding" or "habit development", we are still kidding ourselves. We are playing with people's brains and lives, and I may regret for a long time just how much I did this and in fact just how good I was at doing it.

Posted by Kathy Sierra on June 11, 2013 at 12:25 PM MDT #

Post a Comment:
  • HTML Syntax: NOT allowed

All things gamification, mostly focused on the Enterprise space but occasionally on other issues related to gamification. Thoughts are my own.


« June 2016