The Physiology of Gaming
When Dr. Nora Volkow and I were interviewed by NPR in 2013, she was asked if games were addictive, or being designed with that intent. It's a complex question that she was not ready to answer definitively and I have come to the conclusion that it was premature to ask that question in 2013 or even 2017. Similarly, as governments around the world grapple with questions like “are loot boxes gambling?” I think this question is premature also.
Both of those questions require a variety of prerequisite questions to be answered first and we have not done that so we are getting ahead of ourselves. In order to successfully answer these questions, and even harder ones that are almost upon us, we first need to construct a framework for discussion and dialogue. Many of these questions will have no right or wrong answer and without a system to allow cooperative dialogue, conflicting ideologies will prevent consensus. This consensus is not a luxury. As Dr. Volkow and I agreed in 2013, the ability to control the actions of others electronically existed in 2013 (in rats certainly, and possibly in higher organisms also) and any limitations that may restrict that capability with humans are merely technological hurdles we will clear within the next ten years.
An important first step will be for us to agree that interactive media can trigger physiologic effects in humans, and that we have some control over the quality and quantity of those effects. It is not clear that we have completed this basic first step, as language used in relevant discussions does not reflect that agreement yet. Historically it is clear that anything that might trigger pleasure in humans will be created and sold to consumers whether it is legal or not. A short list of these substances that were sold legally across the world include alcohol, nicotine, natural/artificial/refined sweeteners, sex, a laundry list of opioids, marijuana and THC, and the list truly goes on. Many of these substances were later made illegal and even legal again, and were regulated in some way.
While all humans are motivated by pleasure, there are many world views and ideologies that believe that pleasure is a bad thing. Or perhaps that it should only be accepted if it is scientifically understood. Some of the more serious scientists attempting to study pleasure in the last century were stymied and in some cases imprisoned for their work. Having the author of The Sexual Revolution put in jail until he died, and having his books burned had a chilling effect on others seeking to do such research so it might come as no surprise that there is minimal public research on products being used for pleasure by over a billion consumers world-wide. What we do know is that public health and mental health metrics have been following a downward trend since the advent of mobile interactive media products and we have not done enough research to find out why. Recent public declarations from former Facebook execs (which mirror concerns expressed to me in private by gaming industry directors) are troubling but anecdotal. Determination of causality requires research and a scientific understanding of how these technologies affect us biologically in order for us to be able to develop safer future generations of these technologies.
The Autonomic Nervous System
Most games, all “games as a service”, and social networks work by triggering the Autonomic Nervous System (ANS). [The system is named this way because it controls your cells in the background automatically, typically without conscious thought. Simultaneously, any conscious or unconscious thought can trigger the ANS without you even being aware of it.] Thus effective design of these products is facilitated by having at least one person on the design team having a graduate level education covering human physiology and neuroscience. In the absence of this, statistical methods can be used to do A/B testing to attempt to reverse engineer results. Later in the paper I will detail how this looks so the limitations of this method will be more clear. It is understood that game physiologists don't exist yet, but in 2005 when I began promoting game economics in the industry, game economists did not exist either. Now only 13 years later the idea that every major studio requires a game economist is pretty standard.
The ANS is composed of two competing systems called the Sympathetic Nervous System (SNS) and the Parasympathetic Nervous System (PNS). Competition here means that when one system is stimulated, the other system is suppressed. Normal natural humans have these systems in balance automatically thanks to evolutionary processes. As humans have developed new technologies to pleasure themselves going back hundreds of years to tobacco or even earlier chemicals, there have been more opportunities to disrupt that balance and the result is often an increase in disease. We are so familiar with this pattern that we have come to assume that excesses of pleasure are automatically linked to disease. You can see this with the ideological battle over whether we should allow marijuana to be legally consumed. Many of those arguing that it is dangerous oppose its use even without supporting scientific data, while simultaneously making it difficult for scientists to conduct this very research. Inductive reasoning suggests that if all the other pleasure-inducing drugs before it (even though Marijuana is a very old drug) caused disease, this one will too.
There is some truth to this but as you can imagine, the physiology is not so black and white and a proper understanding of how the ANS works with the SNS and PNS in balance is necessary to weigh consumer needs against public health. The SNS is often called the “Fight or Flight” system. Activation of this system causes increased alertness, strength, and mobility at the cost of tunnel vision and impaired ability to think beyond the present. The immune system is also suppressed by the SNS as all available resources are diverted towards immediate survival needs. When a tiger is chasing you, you don't worry about what you will be doing next Tuesday and certainly not what you will be doing a year from now. On the most basic level, resources from cells are turned into energy during SNS activation, appetite is suppressed, and weight is lost.
When the PNS is most dominant (usually during periods of sleep, rest, or just after eating) the body takes available nutrients (from whatever you ate recently, if available) and rebuilds damaged tissues, grows new hair/nails/skin, and stores leftover nutrients as fat. Appetite is increased by PNS activation.
Normally a person cycles between the SNS being most dominant just after waking up in the morning and the PNS being most dominant during sleep. Pre-industrial humans had percent body fat levels similar to modern day Olympic athletes (this is “biological normal” for humans) but were shorter and less muscular. During the three years I worked at The Betty Ford Center, a drug rehabilitation center in California founded by the American First Lady of the same name, I was taught that this normal balance between the two systems (SNS/PNS) can be overridden by outside inputs, usually drugs. At first a person might use something like caffeine to get a “boost” when they are tired or need more awake hours for some task like schoolwork, and then when they have a hard time going to sleep they might drink a beer or a glass of wine to “come down”.
Very occasional use of chemicals does not disrupt the balance of the ANS. As humans are required to work longer and longer hours, raise children with just one parent, and are exposed to an ever increasing array of external stresses (inevitable as population pressure increases) there is tremendous pressure to rely on these chemicals regularly instead of occasionally. Doing so turns off the body's normal rhythm and the cycle becomes drug dependent. The more SNS stimulants a person uses, the more PNS stimulants a person needs to come down in order to sleep. Motor vehicle accidents are the #1 cause of death of teens, and in 20% of accidents a driver that was fatigued or sleepy was the cause.
The following two tables are simplified for brevity and ease of explanation. The stimulants table has abbreviations for what internal chemicals they stimulate (+), emulate(+), or suppress(-). Some of the stimulants affect multiple (mult) systems or are not fully understood at this time (that would apply to marijuana and games) and need more research. All of the adrenal axis chemicals are lumped together as (Adr).
Table of Internal SNS/PNS Chemicals
Adrenaline (Adr) Insulin (Ins)
Norepinephrine (Adr) Oxytocin (Oxy)
Dopamine (Dop) [Growth Hormone (HGH)]
Table of Common SNS/PNS Stimulants
Action Games(+Adr) Social Games(+Oxy)
Cocaine(+Adr) Touch/Social Interaction(+Oxy)
Exercise (Mult) Massage(+Oxy)
Marijuana (+Dop) Hot Bath(-Adr)
Nicotine (cigarettes)(+Dop) Marijuana(*)
Gambling (inc. Loot Boxes)(+Dop) Sweeteners (inc. 0 calorie)(+Ins)
Action/Horror Movies/Books(+Adr) Romance Movies/Books(+Oxy)
Some stimulants affect multiple systems simultaneously, which makes them especially popular forms of recreation even if their affects are less predictable. Since games and sex work in similar ways (though we are still bad at making social games for oxytocin), it's possible they can be social competitors. Many a (usually female) partner has complained to me about how “games ruined our relationship”. Here I mean people in the public space but I have also experienced this first hand.
Catecholamines: All of the SNS stimulants are catecholamines, meaning they trigger the breakdown of stored resources for energy. Adrenaline (the weakest), norepinephrine, and cortisol (the strongest) are progressively released as part of the “fight or flight” response with cortisol being reserved for the periods of greatest or longest stress. These 3 adrenal axis chemicals take longer than 15 minutes to “wash out” so it takes time to come down after they are released. Dopamine is also a key neurotransmitter and like all neurotransmitters it washes out almost immediately.
Exercise and HGH: Exercise triggers the release of all catecholamines and also triggers the release of human growth hormone. HGH isn't itself a PNS stimulant, but it only functions properly when the PNS is dominant. Thus if you train hard but don't rest hard, you don't see the benefit of that exercise in terms of muscle growth. HGH is also responsible for cellular regeneration such that as we get older or more sedentary our levels drop and our skin and many other structures in our body become thin and delicate. Thus our skin shows our biological age which may be very different than our chronological age.
Insulin: This is an anabolic hormone just as HGH is but it is responsible for storing resources as fat, not building muscle and other lean tissues. So insulin makes you larger by making you fatter. Eating any sort of food can cause a release of insulin, but the sweeter or fatter the food, the stronger the release. Modern sweeteners are many times stronger than those that were available during our evolutionary period. High levels of insulin increase appetite, so exposure to very powerful sweeteners, even those without calories, cause appetite increase and weight gain. Chronic exposure, especially when obesity is also present, is a powerful risk factor for adult onset diabetes.
Oxytocin: Insulin and oxytocin are the two primary PNS chemicals and SNS counter-chemicals. Unlike insulin which is key to an epidemic of obesity and related diseases that are lowering our lifespan and raising health care costs, oxytocin only has one major down-side: it raises trust levels. In social situations that could make people more gullible and giving. For us as producers of products for consumers, this is all upside. Oxytocin also improves mental health and boosts the immune system. For us the main downside of oxytocin is that it has a very short half life. We can't (yet) put it in a pill and sell it. Sex and massage, traditional ways to get it, are labor intensive services and can't be mass produced. Those services command very high prices for a reason but if we can digitally distribute oxytocin, we can overcome these obstacles. Dr. Paul Zak showed that you can get oxytocin from a digital device and we agreed to do joint public research in 2014 to see how to best generate it in games, but the funding fell through. The potential profits from such development are mind boggling.
*: Opiates emulate the body's natural endorphins, which can be released during extreme game play. This system is not detailed here for simplicity and safety reasons. A more detailed description can be found in “I'm Dying to Play” along with the necessary precautions. Marijuana is a very complex drug that can have unpredictable effects due to its emulation of natural neurotransmitters. No attempt is made here to describe its effects in detail.
Seeing a sexy picture can prime us for sex even without the sex. Seeing a loot box can cause us to release dopamine even before opening the box. Tasting a strong sweetener, even if it has no calories or we don't swallow it, triggers a strong insulin response. Seeing a tiger, gun, or any other threat can trigger the SNS without us actually being harmed. Our body in most cases can react to a stimulus without us even having the stimulus. In other words,
If we can imagine it, then it is real.
When you think about it, our brains are several pounds of meat floating in cerebrospinal fluid. It's dark in there. There's no sound, feel, smell, or any other form of stimulation other than nerve impulses. We have no ability to directly see “reality”. All of those nerve impulses are processed through the filter of our ideology, beliefs, previous experiences, competing nerve impulses, and world view. This means I can show a picture to ten different people and have them tell me ten different stories about what is going on. Putting it a different way,
Reality is personal, or perhaps does not exist at all.
Nonetheless humans are attached to the concept of reality and tend to think their reality is the real one and that other people just “don't get it”. Storytelling, books, movies, games, social media, music, VR, AR, “fake news”, psychometric sandwiches, etc. are all examples of New Realities (NR). Once they become more convincing than what passes as our default reality, advanced NR technologies will allow us to control human behavior much more effectively than the rat experiment that Dr. Volkow described in our NPR interview. Bending human physiology to create the consumer responses we want is a relatively new field, and one that industry is way ahead of academia in. The tobacco and food industries have gotten almost a 50 year head start on interactive media industries in this space, but the potential successes using interactive media are unrivaled in the field of NR. The problem is that industry tends to keep this R&D secret for a number of reasons and thus the public largely doesn't know just how far that science has come.
A Theoretical Example
Let's say that I know that middle aged women are a largely untapped demographic with huge spending potential. There are three single women for every single man in this age group. This is a serious glitch in the Matrix and is causing this demographic a lot of stress. We would expect a little more dopamine and oxytocin would be worth a lot to this group. Animal companionship substitution appears to be occurring, but with at best mixed results. If we can make games that assist this group in meeting their dopamine and oxytocin needs, we could make a lot of money and make people happy at the same time.
We also know that while the adult obesity rate in the USA is almost 40% and climbing, the rate is much higher in this demographic. They consume a lot of augmented sweeteners, often sold to them as diet aids, that raise their appetite and make them gain weight. “Sweetness dependency” is high. So we do some testing and find out that games that revolve around items that are sweet (like candy) are super popular in this demographic. This is just as you would expect if viewing the situation physiologically. [In Asia, where they have less obesity and less animal companions, games with cute animals dominate this demographic just as you would expect. The scarcity of children in economically advanced Asian countries might also suggest the use of children's images might be advantaged].
Now let's throw in a deep, sexy, soothing male voice to spice things up for this demographic. While we could predict this would be popular, our BI guys confirm this is a big hit. But what should we have the voice say?
From a purely physiologic view, the correct answer is obvious but we have our BI guys check just to make sure and they give answer C a big thumbs up. [Perhaps children's voices in Asia?] Okay so this is great for retention but not for spending. Now what? Well I know that spending decisions take place in the pre-frontal cortex which is well developed in adult women but poorly developed in children. Going after kids is poor form so let's stick to options that only include adult women. We know that the pre-frontal cortex is suppressed during SNS stimulation, so let's throw in some of that. Make it so that when they lose a round they “lose a life!”. That should be enough threat. Then for a token amount of premium currency they can keep going and not die. BI again gives me a big thumbs up. To generate more threat we can put a timer on some of the waves, but we have to use caution because adrenal axis chemicals (Adr) take at least 15 minutes to wash out, as opposed to neurotransmitters like dopamine which wash out rapidly and can be repeatedly stimulated. So spacing timed waves at least 15 minutes apart is a good move, and more will reduce the risk of player fatigue and younger age demographic creep.
This is an example of where a BI team could come up with the same results as a game physiologist, but it just would take more time and money. It is even cheaper if you just copy what has proven to be successful already. But BI can't tell you if there are any side effects to any of these changes. BI can't tell you “why” something works. It also has a difficult time telling you if a seemingly unrelated metric is going down for a related reason.
The key to health and wellness is balance, especially when it comes to the ANS. Balance is the default state but when it is disrupted then morbidity and mortality rates increase. In America anxiety, depression, suicide, obesity, diabetes, and hypertension rates are all skyrocketing. I would attribute all of these trends to progressive society-wide ANS disruption. I previously disclosed the secret to Florence Griffith-Joyner's “unbeatable” world record. She nailed the ANS balance during the 1988 Olympic Games preparation.
Today humans in America sleep an hour less per night than they did a generation ago. That is a massive shift in our physiology in a very short period of time. Since SNS activation causes alertness and PNS activation causes drowsiness, it seems rational to assume that our ANS is out of balance, with the SNS being stimulated a lot more than the PNS. Less sleep means less cellular regeneration and healing, leading to an increase in “death from all causes”. More SNS stimulation means more alertness and stress. A PNS suppression could also make social interaction more difficult. These are all effects we are witnessing now with teens waiting 3 additional years before dating being just one related metric.
Where is all this SNS stimulation coming from? Could the recent alarming language coming from former Facebook execs saying they fear they made a big mistake by optimizing their platform for dopamine production be related? What about the coffee shops on every corner, many with drive-throughs? For the rest of this article, I'm not going to worry about where it came from.
Let's forget about ethics, health, happiness, sustainability, crime, productivity, military readiness or any of those other niceties we might consider that might be negatively impacted by this imbalance. Let's just talk about money. Let your greed-flag fly proud for a moment! In an environment where we are suffering ill health from too much SNS stimulation, what sort of products would you expect to be big winners? Right! PNS stimulation! PNS and SNS systems are competitive so stimulating one reduces the activity of the other.
In an environment where people aren't sleeping enough, and dying from that and are unhappy before dying, what's going to sell is things that help them sleep. PNS stimulants. Things like food, sex, opiates, marijuana, and alcohol. All of these are selling amazingly well right now, especially the middle three since the first and last ones have nasty side effects that have been well researched. Opiates have nasty side effects but that doesn't seem to be crimping demand at all, possibly because the price point has dropped significantly with the introduction of fentanyl (this is also raising lethality).
We have a huge opportunity to take advantage of the situation by selling pro-PNS products. As an industry, we are horrible at making these products and the resources dedicated to making pro-PNS games is less than 10% of total investments. Your BI team doesn't need to know why threat generation improves (short term) conversion rates. They know it does and they have the stats to prove it. They also know that gambling mechanics like loot boxes (which trigger dopamine release, an SNS stimulant) lead to conversions. I'm not disagreeing with any of this because it's all true. But BI can't tell you why, and the why is important here. Not knowing the why is costing us a lot of money because we are making products that consumers don't want. Yes they are physiologically vulnerable to these methods but this leads to consumer adaptation as their inductive reasoning links your product to their last bad experience with something similar. Only the youngest gamers (the ones with relatively small budgets) want more stress (SNS stimulation) in their games. Supposedly we are not targeting those gamers anyways, at least this is what all the major developers (and their lawyers) keep telling me.
Some readers may be concerned that adding physiologists to the game design process might be a dangerous escalation of the game design paradigm. If you consider that we already use data science to gather the metrics that benefit us in data driven design, but are unable to gather metrics related to player health using these means, we are already putting our consumers at higher health risk than we would be using a physiologist. Thus seeing this as a negative in terms of consumer well being is not a rational response. Liability issues I cannot address sufficiently as I am not an attorney, but this genie is already out of the bottle and history will repeat.
So how do we go about making products that better meet the needs of consumers? Part two of this two part series, “How to Make Healthy Games” will focus on answering this question. While making gamers healthier and happier is an added bonus of making healthy games, the bottom line is that there is huge unmet consumer demand for these products and this is why we should be making them.