The Zynga Abyss

In this exclusive excerpt from the forthcoming debut issue of Distance quarterly, we learn about how too many video games treat players like rats in a Skinner Box, lulling them into easy stimulation but requiring little creativity.

In the 1890s, while studying natural sciences at the University of Saint Petersburg, a Russian mathematician named Ivan Pavlov was analyzing dogs' saliva output over time. Pavlov noticed that dogs tended to salivate more before eating and that merely the sight of a white lab coat would induce salivation -- even if no food was on the way. So he tried ringing a bell before presenting them with food, and found that over time, the dogs would salivate even if a bell was rung with no food presented. Pavlov's research defined classical conditioning, in which a primary reinforcer (one which naturally elicits a response, e.g. food or pain) is associated with a conditioned or secondary reinforcer, such as the lab coat or bell.

Forty years later, Burrhus Frederic Skinner built upon Pavlov's observations as a young psychologist in graduate school. He constructed a soundproof, lightproof chamber that housed a small animal; a lever was placed within the animal's reach, which triggered a primary reinforcer. Called the Skinner box, the device opened up many possibilities for experimentation, leading to breakthroughs in later research: from the relative addictiveness of cocaine in isolation versus in a larger community, to the question of whether rats have empathy.

Skinner is now credited as the father of operant conditioning: a form of learning where a subject is conditioned to respond to a secondary reinforcer through association with some form of primary reinforcement. Not only did Skinner's work show that associations between primary and secondary reinforcers appear in nature, it also demonstrated that new reinforcers can be manufactured.

Skinner and Pavlov proved that primary reinforcers are extremely powerful motivators. After sex and sleep, bacon is one of nature's most powerful primary reinforcers, partly due to its high fat and protein content in comparison to other meats. Bacon has become known as the "gateway meat": the smell triggers intense cravings, even in vegetarians. But in our modern world, our instinctual craving for bacon and other fatty foods can cause significant health problems.

The box also taught us two fundamental lessons, one of which had ramifications that extended far beyond Skinner's experiments. Humans are hardwired to respond to primary reinforcers, just like any other animals. And while primary reinforcers have a diminishing effect once we're satiated, secondary reinforcers, like money or social status, exist outside our biological needs, and these never hit a satiation point. In other words, we are hardwired to seek approval from our peers, and we can never get enough of it.

Many people defend FarmVille as a harmless distraction, arguing that the thousands of hours spent playing the game would still have been wasted on other activities. But there's no question that the social game market, with its virtual currencies and unlimited stock of goods, is a huge cash cow. And it's also clear, when you look more closely at FarmVille, that it was engineered with one goal in mind: to coerce users into tending their virtual plots of land for as long as possible. Using our natural tendency to reciprocate gratitude from our peers, we end up pestering our friends to keep returning. And cleverly-timed crop cycles force players to return to their farms at all times of day. But what about the techniques employed in other games?

Black, White, and 254 Shades of Gray

Moral relativism aside, I think "bad" games exist -- provided we define "bad" in unambiguous terms. In Groundwork of the Metaphysics of Morals, Immanuel Kant tried to specifically define this with the Categorical Imperative: a set of rules that could gauge an action's morality. But regardless of how we choose to evaluate a game's morality, there are certain traits which can push it closer to the "evil" side of the spectrum.

The primary characteristic of unethical games is that they are manipulative, misleading, or both. From a user-experience standpoint, these games display dark patterns, which I define as common design decisions that trick users into doing something against their will. Dark patterns are usually employed to maximize some metric of success, such as email signups, checkouts, or upgrades; they generally test well when they're released to users.

For example, FarmVille, Tap Fish, and Club Penguin play on deep-rooted psychological impulses to make money from their audiences. They take advantage of gamers' completion urge by prominently displaying progress bars that encourage leveling up. They randomly time rewards, much like slot machines time payouts to keep players coming back, even when their net gain is negative. And they spread virally by compelling players to constantly post requests to their friends' walls.

This trend is not just limited to social games, though: many combat games, like America's Army, are funded by the U.S. military and serve as thinly-veiled recruitment tools. Some brands have launched Facebook games like Cheez-It's Swap-It!, and they serve as tools to sell more products. These techniques can be used in any sort of game, in any context.

Of course, not all games fall neatly into a clear division between good and evil. "Good" games can ask little in the way of critical thinking and problem solving, but still be fulfilling. Tetris is a complex, enriching game that is just as addictive today as when it debuted in 1988; and yet many would argue that it lacks the problem solving challenges or rich plots that make adventure games like Sword & Sworcery and The Legend of Zelda so compelling. Different people find meaning and fulfillment in different aspects of gameplay; that's what makes gaming so fun for players and game designers.

Hard Fun

One of the best-selling independent iOS games is a side-scrolling jumper called Canabalt. It consists of a tiny man running over the rooftops of a dystopian cityscape. Players tap to jump, and the gameplay speeds up over time. There is no end to Canabalt, only longer runs: the longest run yet recorded, at 8 minutes and 16 seconds, takes the player through more than 22 kilometers of crumbling buildings, falling obstacles, and inconveniently placed windows.

At IndieCade in October 2011, Adam Saltsman, Canabalt's creator, discussed the notion of "time until death." All of us have a finite amount of time on earth, and any time we spend on a particular activity is time that we can't spend doing something else. This means that the time we spend gaming represents most of a game's cost of ownership, far more than any money that we spend. If that time is enjoyable (or rather, if its benefits outweigh its costs), then the game was worth our time.

Value is created in different ways for different people, but the most immediate is through generating engagement until players achieve mastery. In a panel held at Seattle's Casual Connect in 2011, game designer and consultant Nicole Lazzarodescribed two types of fun: easy fun and hard fun. Games that don't challenge players beyond a certain point -- "easy fun" -- will never allow them to achieve mastery, which could deprive them of a highly rewarding part of playing.

The panel also included Demetri Detsaridis, the general manager of Zynga's New York office. Zynga has its own ideas of what constitutes the "real fun" in FarmVille and similar games, which align neatly with the company's business interests. His answer of how they approach "easy" vs. "hard" fun was telling:

You know, while we don't necessarily have this framework in particular in mind... you know, while I was looking at this chart yesterday, while we were talking, I was thinking "Well, Zynga does a lot, if not most, of its development work in this kind of infinity symbol loop here between 'people' fun and 'easy' fun... there's sort of an overlap here that isn't maybe entirely clear on the chart, but a lot of... I think a lot of social games kind of are really quite close to the top, you know three-quarters of this, that the 'people' fun and the 'easy' fun are really sort of mushed together, and that where you see the hard fun coming in is in perhaps surprising places, like thinking about your social graph and how you, in real life, are managing that - well, am I sending, you know, friend requests to these, these people? Is that - you know, so that's actually part of the game, and designers know that, so that sort of is an interesting almost meta-game layer of almost 'hard' fun on top of what otherwise might seem in this structure to be really people- and easy-centric.

Detsardis nodded in approval as Rob Tercek, the panel's moderator, summed it up:

The games themselves aren't where the action happens; the strategy component is: when do you reach out into your social graph? When are you going to spam that list? How frequently are you gonna do that?

I'll reiterate this in plainer language, just in case the quote wasn't clear: Detsaridis said that one of the most compelling parts of playing Zynga's games is deciding when and how to spam your friends with reminders to play Zynga's games.

Creating hard fun isn't an easy task. It requires thinking deeply about the gamer's experience, not just using cheap tricks to drive engagement. FarmVille, Tap Fish, and Club Penguin all employ Skinner-like techniques to persuade people to spend more time and money. But there are plenty of honest ways to create real engagement, and it's our responsibility as creators and consumers of games to demand more honest and fulfilling fun from our entertainment.