Gamasutra: The Art & Business of Making Gamesspacer
Psychology is Fun
View All     RSS
November 23, 2014
arrowPress Releases
November 23, 2014
PR Newswire
View All






If you enjoy reading this site, you might also want to check out these UBM Tech sites:


 
Psychology is Fun

September 23, 2010 Article Start Page 1 of 3 Next
 

[In this thought-provoking piece, psychology researcher and author Clark takes a look at how psychology and can must be applied to game development, to produce works that engage audiences -- offering up concrete examples of the right techniques.]

Gaming's core is fun, and psychology is fun's touchstone. This article restricts itself to psychology's most foundational, most immediately-applicable methods for crafting sticky, captivating experiences. From behaviorism's methods for structuring overpowering rewards, to motivational theories on generating wants and needs, to hybrid theories like flow, it is no longer fiscally responsible for games companies to shun psychology. Let's jump right in.

The Heart Asks Pleasure First

Pleasure first, and then, excuse from pain, shape every move that we will ever make -- so say the behaviorists. It may sound callous, or reductionist, until we realize that the overwhelming majority of life's rewards and punishments are too tiny or timeworn to remember.

Whether we know it or not, whether we like it or not, behavior is shaped by reward. Through behavioral tenets, psychologists have conditioned pigeons to play piano, play ping-pong, and to spot drowning men and women while mounted onto Coast Guard helicopters.

Operant conditioning, often associated with Edward Thorndike, then B.F. Skinner, is the study of how any behavior can be strengthened through reward.

There are two fundamental ways to strengthen behavior through reward: bring pleasure, or excuse pain. Positive reinforcement, and negative reinforcement, respectively. There's also punishment, which introduces sharp consequences but does not necessarily remove what first reinforced a behavior.

Not all rewards are created equal. Primary reinforcers are so named because they are the most automatically powerful, and rely on the innate; some examples would be sleep, visual surprises, or sex.

Let's talk about sex. Whether or not a player realizes, a physically attractive character, when on screen, will serve as an innate positive reinforcer. On the flipside, allowing the player to remove visually aversive characters or imagery is an innately rewarding negative reinforcer. If we care about giving players pleasure, and fun, then we will first work to maximize primary reinforcers.

 

Before Condition

Behavior

After Condition

Impact on Behavior

(+) Reinf.

No reinforcer

Obtains

Reinforcer

Behavior Increase

(Ex: Skinner Box and food)

(-) Reinf.

Aversive

Escapes

No Aversive

Behavior Increase

(Ex: Tylenol and relieving a headache)

(-) Punish.

Reinforcer

Invokes

No Reinforcer

Behavior Decrease

(ex: Remove a kids toy when they are bad)

(+) Punish.

No Aversive

Elicits

Aversive

Behavior Decrease

(ex: Hit a kid when they are bad)

Though it takes little work to reward ourselves with primary reinforcers, we often associate them with other stimuli. Classical conditioning, the precursor to Skinner's work, would call innate pleasures the unconditioned stimulus because they're automatic. No association necessary.

For me, a powerful association comes from that one scene from The Hangover (at the tail of this trailer), when, to the famous movement of Phil Collins' In the Air Tonight, Mike Tyson has his moment with Zach Galifianakis. If you're not expecting it (and for some of us doubly so once we do expect it), this is a novel visual stimulus which provokes an innate physical response.

In the Air Tonight is now for me a conditioned stimulus. I wait the whole song, building up to that movement, which due to the visual association acts as a conditioned response with a very particular reward. I grin. In classical conditioning the level of reward is typically dictated by the timing (the above scene is timed to match the song), intensity of the innate stimuli (Tyson is an intense guy – but some folks might not even notice this scene), and the frequency with which the two stimuli are paired (How many times have you seen The Hangover?)

After enough pairings, many rewards become secondary reinforcers. Money, barter objects, praise, achievements, though there are few limits to what can become attractive by association, we'll want to first take advantage of culturally popular reinforcers. Money is great for developing games in any Western market, because we can literally drown players with money.

Though not every monster is fun to kill, or working for the equipment upgrade delivery service, each may carry the universal symbol for reward. It's as Madonna once said: "the boy with the cold hard cash is always Mr. Right." Though some designs will be too abstract for some secondary reinforcers, working with the grain of cultural assumptions helps to make reinforcement obvious and powerful. It makes the game fun.

But a fun game won't just throw money, goods, or innate rewards at players; it will deliver those on a weighted and considered schedule. Foundationally, behaviorism offers us five foundational ingredients for a healthy and balanced reward schedule. Firstly, continuous reinforcement operates just as it sounds. We reinforce a player every single time they perform the behaviors that we'd like to see. We may even reinforce behaviors that get incrementally closer to what we'd like to see, what behaviorists call shaping.

A great example for both continuous reinforcement and shaping is level one in World of Warcraft, regardless of race or class selected. Learn to walk properly, kill efficiently, use skills, loot, sell, etc., and there's no dearth of praise, experience, and cash value. Yet, continuous reinforcement is the first to wear off, because players immediately notice once you've staunched the flow of reward.

So we also distribute rewards in ratios and intervals. Sometimes these are fixed ratios, or fixed intervals, rewarding only after a set number of correct responses, or rewarding after a set amount of time, respectively. We can also use variable ratios, or variable intervals, rewards that come after roundabouts a number of correct responses, or roundabouts a certain amount of time, respectively. Along with continuous reinforcement, these first five are the most common of the simple schedules.


Variable ratio is widely noted for its effectiveness and consistency. Note that such charts vary where FR and FI are concerned. Some paint FR as inciting more responses than portrayed, some paint FI as inciting fewer than portrayed. (CC: Public Domain)

Most of us have probably played through compound reward schedules (though perhaps not noticed). Say you need 100x ghoul extremities (it's for a quest, okay?) At first you'd find extremities on one of every three ghoul corpses. Before long, however, you'd only find extremities once every two minutes, and then only on one of about every five ghouls, and then (zounds!) on every damn ghoul you'd marauded, but that didn't last long.

Finally, once you had about ninety-five helpings of ghoul extremities, you could only seem to find one every, say, three minutes or so. Noticing this is not purely paranoia. Using simple schedules in tandem, as with the example above, is common.

If a quest-giver actually provides us some indication of the reward conditions, for instance, "just kill six ghouls," then, "gooooood, now kill one every minute for fifteen minutes, that shall work quite nicely," and so on, until there's an extremity-themed reward at the tail end, behaviorism also calls this a chain. If you're looking to go further with compound schedules, Wikipedia (as usual) is a great place to start, though (as usual) is not always scholarly.

As with game design, behaviorism encourages seeding concurrent reward schedules. They let our brain pick and choose the best way to reward itself. The key to generating fun in the brain of the player is to cater to them. They should always have options for how they want to stimulate themselves. Don't bother them with aversive situations. We already know about the world we're escaping from.

It's as the great Dr. Frank-N-Furter once said, "Give yourself over to absolute pleasure."

Though do so with savvy. It may, for instance, help pleasurability to take short breaks now and again, as pointed out by this short Gamasutra piece on hedonic adaptation. Well-structured rewards may make it hard for us to do this ourselves, as our chief neurochemical for motivation, dopamine, has been shown to have less to do with pleasure than with appetite, or "seeking."

Our brain changes itself structurally, over time, in a process called plasticity. Though we may feel the delightful spritz of dopamine the first few times we encounter a new stimulus, before long we rewire to feel the need of these once-novel stimuli. Rewards then begin to trigger the same motivational neurocircuitry as food, sex, stress, and so forth. Games, therefore, must further their understanding of this neurocircuitry, as well as behaviorism, should they desire to keep up the pace.


Article Start Page 1 of 3 Next

Related Jobs

Infinity Ward / Activision
Infinity Ward / Activision — Woodland Hills, California, United States
[11.22.14]

Lead Game Designer - Infinity Ward
GREE International
GREE International — Vancouver, British Columbia, Canada
[11.21.14]

Sr. Game Designer
Yoh
Yoh — Vancouver, British Columbia, Canada
[11.21.14]

Vehicle / Weapons Artist
Yoh
Yoh — Vancouver, British Columbia, Canada
[11.21.14]

Special FX Artist





Loading Comments

loader image