Chapter 11: Exploiting addiction

Consider some of the following criteria against your own behaviour, and your friends and family:

1. Using something for longer than you intended.
2. Trying to cut down, but not quite managing it.
3. Usage requires you to spend a lot of time recovering (from sleep deprivation, for instance).
4. Feeling cravings when you’re not using it.
5. Usage causing you to fail to fulfil your obligations (at work, school or home, and so on).
6. Continuing to use it even though it interferes with your relationships (for example, recurring arguments with or resentment towards your partner).
7. Using it in hazardous situations (such as while driving).
8. Choosing not to attend social or recreational events so you can use it instead.
9. Feeling withdrawal symptoms when you can’t use it.

This list is paraphrased from the DSM-5, the book used by US psychiatrists to diagnose mental health disorders – in this case, substance abuse disorder.¹ But it’s easy to look at this list and think about compulsive behaviours in games and other digital products. It must look familiar if you know someone who has developed an unhealthy habit with League of Legends, Call of Duty, Twitter, Facebook or whatever else. This is known as behavioural addiction – where there are no drugs involved – but the phenomena are very similar.²

In fact, some neuroscientists claim they have found evidence that all addictions share the same biological origins in the human brain.³ Whether you’re abusing a drug or just can’t stop playing League of Legends, they argue that it’s the brain’s reward system that’s being tapped into – specifically, the mesolimbic dopamine pathway.⁴

It’s easy to exploit addiction for profit if you’re willing to overlook the negative impacts. Some products are designed from the ground up to be as addictive as possible, and they use various forms of manipulation and deception to do this. To quote the first president of Facebook, Sean Parker:⁵

How do we consume as much of your time and conscious attention as possible? […] we needed to sort of give you a little dopamine hit every once in a while because someone liked or commented on a photo or a post or whatever… It’s a social validation feedback loop… You’re exploiting a vulnerability in human psychology… [The inventors] understood this, consciously, and we did it anyway.

— Sean Parker, 2017

Surprisingly, one of the most effective ways to make something addictive is to design it so that recipients don’t always get what they want, and to only let them get it in an unpredictable way. This is called a ‘variable reward schedule’ or ‘intermittent reinforcement schedule’.

This was famously demonstrated by psychologist BF Skinner in 1938.⁶ He conducted experiments with rats and pigeons in which he presented them with a lever to press. When the animal pressed the lever, they received a food pellet or some other reward. Skinner discovered that when the reward was given on a variable schedule (that is, in an unpredictable way), the animals were more likely to keep pressing the lever than when the reward was given on a fixed schedule (in a predictable way). This is best understood in the context of the dopamine cycle, shown below.⁷

It’s believed that the variable schedule works because the animals never know when the next reward is coming. This creates a sense of excitement and anticipation that keeps them engaged in the behaviour for longer periods of time, which is now understood to relate to the dopamine cycle (shown above). A predictable schedule breaks the cycle because it doesn’t give this sense of excitement and anticipation. Given how profitable addiction can be, it is very common to see addiction models being recommended as good practice in marketing and product design literature, even though the harms to society are well known. Eyal and Hoover’s ‘hook model’ (2014) is perhaps one of the most famous examples, though if you look at it closely you’ll see that it’s really just a rebranded version of the dopamine cycle.⁸ Instead of addiction, it’s ‘hooked’, which appears to be an effort to make the exploitation of addiction more socially acceptable.

Behavioural addiction is often associated with the user going into ‘the zone’, a state of flow where they lose track of time and their worldly concerns, becoming deeply immersed in the activity. Anthropologist Natasha Dow Schüll argues that it’s the zone that many people seek when they become addicted to gambling, as it gives them an escape from their day to day problems.⁹

It is possible to exploit this insight by designing products to avoid interrupting users when they are in the zone. For example, infinite scroll is employed in most social media products, and it serves to avoid interrupting users with pagination buttons, nor reminding them with how many pages of content they have browsed through. The inventor of infinite scrolling, Aza Raskin, regrets creating it. In a 2019 interview he said, ‘I regret that I didn’t think more about how this thing would be used […] I know as a designer that by taking away the stopping cue, I can make you do what I want you to do.’¹⁰

Autoplay is another feature that works in a similar way. When the user finishes streaming a video, another video is automatically selected and played without the user needing to take an action, which can lead to them spending more time than they intended using the product. Similarly, notifications and gamification features (like streaks) can be used to drive users back into a product. When combined with a compulsion loop, this creates a potent mix. Together, these features have drawn much concern, leading to proposals for regulation, such as the proposed US Social Media Addiction Reduction Technology (SMART) Act.¹¹

Video games are one of the biggest areas of tech that exploit the principles of addiction for profit. One of the most controversial techniques is the ‘loot box’: a device that uses a random number generator in order to let users purchase the chance to win virtual items that offer a gameplay advantage or social status within the game world. This is essentially a form of gambling, addictive because of the variable reward schedule which taps into the dopamine cycle, as described earlier. This has led numerous experts to call for it to be regulated more extensively.¹²

But how did we get here? Up until the 1990s, the normal revenue model for video games was pretty simple: money was made when the game was purchased in a shop. This meant that the strategy for success involved maximising gameplay quality and using traditional marketing to increase sales.

In those days, loot boxes were included in many games but without any sort of revenue model attached. For example, if you play the 1996 Nintendo racing game Mario Kart, you’ll find that if you drive over an item box, you’ll pick up an unknown item. This ranges from low value, such as a banana (with a bit of skill you can throw this in front of a competitor and cause them to spin off the race track), to high value, such as a star, which gives you invulnerability to all obstacles. What’s interesting is that the game designers tweaked the item box algorithm so that it was not truly random; instead, they designed it to be very kind and player friendly. If you were in first place of the race, you were more likely to receive a low-value item, and if you were near the back, you were more likely to receive a high-value item. This served to balance the gameplay experience – it stopped novice players from being left too far behind, and it stopped good players from getting too far ahead, so a very mixed group of people (for example, parents and kids) could play together and still enjoy the thrill of a competitive race. In those days, gameplay was primarily optimised for maximum fun.

The advent of the web, mobile apps and integrated payment systems brought about a new revenue model involving in-game purchases and microtransactions, leading gameplay to be optimised for maximum profit. This rapidly became a dominant force in the gaming industry, and it opened the door to predatory practices and deceptive patterns. Game designers started to spend a lot more effort optimising their game algorithms to create a compulsion loop (so people kept playing) and to drive in-game purchases. When combined with paid loot boxes, this became known as ‘gamblification’.¹³

In 2017, game designer Manveer Heir was interviewed about the use of these practices by his former employer EA (Electronic Arts), one of the largest game producers in the world.¹⁴ He said about microtransactions and loot boxes:

‘It’s the same reason we added card packs to Mass Effect 3: how do you get people to keep coming back to a thing instead of “just” playing for 60 to 100 hours? […] EA and those big publishers in general only care about the highest return on investment. […] They don’t actually care about what the players want, they care about what the players will pay for. You need to understand the amount of money that’s at play with microtransactions. . . . I’ve seen people literally spend $15,000 on Mass Effect multiplayer cards.’

– Game designer Manveer Heir (2017)

Loot boxes are becoming increasingly regulated. Games in the Apple App Store¹⁵ or Google Play¹⁶ are now required to disclose the odds of winning in a loot box interaction; and in 2018, the Belgian Gambling Commission banned loot boxes for any businesses that do not have gambling licences, though this has been criticised as being weakly enforced.¹⁷

Even without loot boxes, game designers can find many ways to create recurrent consumer spending hooks.¹⁸ For example, a game can be designed to be fun at the start, but then it can become very hard work (aka a grind) or very difficult. From that point, the user can be offered the means to ‘pay to skip’ the hard work, or ‘pay to win’ and get past a difficult part of the game.¹⁹ When deeply involved in a game, players can find themselves getting carried away and spending large amounts of money, owing to the carefully designed compulsion loops and deceptive patterns. Manipulation, deception and addiction in game design is a thriving area of research, and if you wish to read more, good starting points are the report ‘Insert Coin’ by the Norwegian Consumer Council²⁰ and ‘When the Cat’s Away’ (2021) by Scott Goodstein.²¹