The Psychology of Video Games

Ah, late December. The time when the gaming press gets its members together and tries to convince each other that one awesome game is more awesome than other awesome games –also known as the Game of the Year Awards. When I worked as part of the creative team on GameSpy.com we would lock ourselves in a conference room and argue literally for hours about the minutia surrounding every big title released that year in order to generate our awards. I’m also listening attentively to the GotY content over on Giant Bomb, which is dedicating a full week of multi-hour podcasts to the raw debates that generated its lists.¹

These podcasts are interesting to me because I keep seeing well established psychological phenomenon coming up, but almost as interesting is when a psychological quirk doesn’t manifest itself because the guys seem to be aware of its danger to the process and have taken steps to avoid it. So in this post I present my list of 2010′s Top 5 Biases That Affect 2010 Game of the Year Discussions. Sponsored by Crest Whitening Tooth Strips.²

#5: The Recency/Primacy Effect

The recency effect describes how it’s often easier for us to recall more information (and more salient information) about things that have happened more recently or items towards the end of a list. Similarly, the primacy effect means the same thing for items at the beginning of a list or that happened towards the beginning of an established time frame. Between the two of these effects, stuff in the middle tends to get forgotten or muddled.

Remember Bayonetta?

The impact on GotY lists should be apparent: If you’re studying a list of games released in the last year, it’s going to be easier to recall stuff about the first and last few games. We’re also more likely to recall details about games we played more recently (like Call fo Duty: Black Ops) or earlier in the year (like Bayonetta). Details and memories of games released toward the middle of the year (like Splinter Cell: Conviction) might not come to mind as easily.³

#4: Confirmation Bias

This is a big one for GotY discussions. Confirmation bias is our tendency to ignore or downplay information that dis-confirms our preconceived decisions or opinions and to pay more attention to and emphasize information that confirms them. If you go into a discussion of the Best Downloadable Game of 2010 thinking that Monday Night Combat should win, you’re less likely to think about its flaws (e.g., limited maps, repetitive comments from the announcer) and more likely to remember its strengths (e.g., class balance, fun character design) relative to someone who didn’t hold the same assumption. What’s more, you’ll probably say that the pros are more important to weighting your decision than the cons.

End of discussion! Wait, what?

Good ways to combat this are to get in the mindset of allowing people to challenge your assumptions and engaging in debate. It can also be helpful to list out the pros/cons (with help from others) so that you see them laid out and from a different perspective.

#3: Over-Emphasizing Salient Features

I wrote at length about this concept earlier, but here’s the quick version: When puny humans are asked to justify a decision, we tend to focus on the most salient or plausible explanations and then give them too much weight. To repeat my example from the previous article: if asked to explain why you favor Red Dead Redemption for the Best Action Game of the 2010, you may think about what should be included in the checklist for evaluating an action game, come up with “the weapons,” and then feel compelled to award or take away credit for how the game’s weapons feel and work. The problem is, the most salient and plausible factors may not be the ones that are really responsible for how much you enjoy the game. The weapons in Red Dead Redemption are largely unremarkable –the game’s appeal lies almost entirely in other areas and any weight given to how cool the weapons are is inappropriate at best.

This gun is irrelevant. Ignore this gun.

I keep seeing this come up in GotY discussions because professional game enthusiasts⁴ tend to hate using vague, worn out descriptors like “fun” or “awesome” or “polished” even though those words may be perfectly appropriate if a bit mundane. But these Internet auteurs are determined to have something more descriptive to say, so they cast about for something else and end up falling for the trap described above.

#2: Social Proof and Groupthink

This one is kind of a twofer since social proof and groupthink are separate but related. Again, I’ve written about social proof before, and the idea is that we will sometimes accept proclamations that are clearly at odds with our own senses just because we often have a desire to conform to the group’s standards. Soloman Asch showed this in a classic study where he got people to say that a long line was shorter than a short line simply by having someone planted in the group who would immediately pipe up and say so. The effect is even stronger with a group of strangers and statements with a less clearly defined correct answer, such as politics or game of the year awards. Which is why someone may not speak up when others in the group immediately jump on World of Warcraft: Cataclysm as the Best Role-Playing Game of the year, even though by most reasonable definitions it’s not a game.⁵

Deathwing thinks Line A is shorter. Are YOU going to argue?

The flipside is groupthink, which is when members of a cohesive, established group will ignore information, abstain from critical debate and accept otherwise questionable decisions in order to minimize conflict and maintain warm fuzzies. So again, Cataclysm might win, because so-and-so can be such a pedantic jackass about it and nobody wants to harsh the vibe or destroy the atmosphere of friendly discussion.

Interestingly, the Giant Bomb guys seemed to disarm these two biases from the start by joking about how they hate each other and how they anticipate rancorous arguments. This sets the stage that it’s okay –expected, even– to question each others’ decisions and engage in critical analysis.

#1: The Distinction Bias

Many GotY debates in categories like “Best [Genre] Game” come down to two similar contenders, resulting in protracted discussions where the merits of each candidate are obsessively scrutinized. This is a recipie for what’s known as the distinction bias. The idea comes from a theory that people engage in two modes of evaluation when pondering the merits of an experience: joint evaluation and single evaluation mode. The former is done when comparing multiple things at once and the latter when evaluating something individually.⁶

The distinction bias describes how when operating in joint evaluation mode we tend to over-emphasize and over weight otherwise slight differences between the subjects. If debating Need for Speed: Hot Pursuit and Grand Turismo 5 for Driving Game of the Year, we may make a bigger deal about Hot Pursuit’s lower frame rate than we would have if we were evaluating the game by itself. As a result, when operating in this comparison mode we tend to think worse of the loser than we would have if we had evaluated it without resorting to direct comparisons.

We need speed. And also unbounded rationality.

This is perhaps acceptable in GotY debates when we HAVE to pick a winner –it’s often the fine details that act as tie breakers. But the trouble may come when you have a mix of different types of games where two of them are similar. If you aim to trim the initial list to a set of three finalists, a tempting place to start is by comparing the most similar games (c.f., elimination by alternatives). Because of the distinction bias, the loser in that comparison may end up being evaluated worse than before and may end up getting cut from the list even though it was better than the non-similar games.

So there you have it. Five psychological phenomena that drive game of the year debates. Go listen to your favorite GotY podcast (again, I heartily recommend The Giant Bombcast) and see if you can catch them in action. If you do, post about it in the comments section!

1. Someone could write a full dissertation on Brad and Vinny’s debate over Minerva’s Den vs. Lair of the Shadow Broker as best DLC of 2010, by the way. Get on it. [↩]
2. Not really. [↩]
3. Which might actually help in this case, as Splinter Cell: Conviction was awful. You heard me. [↩]
4. God, what a weird phrase. But more accurate than “gaming journalists.” [↩]
5. Shut up. It’s an expansion pack. [↩]
6. You see joint evaluation mode all the time in game discussions, actually. It’s usually given away by comparisons of a game to some benchmark of the genre or platform. “Medal of Honor is not as good as Call of Duty“, for example, even though the discussion is ostensibly supposed to just be about Medal of Honor. [↩]

As I mentioned a while back I wrote an article for GamePro magazine about the psychology of shooters and why interacting with game worlds through the barrel of a gun is often so appealing. I got some great input from people doing real research in this area and I like how the article turned out.

Now, for those of you allergic to paper, GamePro.com has published the article online where you can read it if you haven’t already. Begin clicking and let me know what you think.

I recently wrote about the endowed progress effect, which makes us more likely to complete progress towards a goal if we have the impression that we’ve already begun taking the necessary steps. For example, people who get 2 free stamps on a “buy 10 get 1 free” card are more likely to put in the purchases needed to earn the freebie –even more so than people who got no free stamps but needed two fewer stamps to qualify.

Valve, the masters of digital distribution and the art of Steam, recently illustrated another application of the endowed progress effect when they launched The Great Steam Treasure Hunt. The idea is that you earn Steam-based achievements by completing game-specific tasks such as earning a score of 1,124,400 in Chime, which just happens to be on sale for $1.25 today.¹ Earn any ten of these achievements and you get entered into a lottery for 100 free games. You can also earn progressively cooler Team Fortress 2 hats by completing any 5, 15, and 28 Treasure Hunt achievements.

Of course, many of these are game-specific achievements, so you have to buy the game to earn them. But the clever thing Valve has done, at least on this first day, is to insert some easy to earn achievements to get you started. In fact, the first one is simply “Setup an avatar,” which is so easy that many people currently using Steam will have already started their progress towards 10 achievements:

Bam. Endowed progress effect. Well played, Valve. Well played.

1. Coincidence? I THINK NOT. [↩]

Apparently the Xbox Kinect is a retail success despite the fact that I haven’t personally bought one. Enough people seem to enjoy flailing their extremities about and barking simple commands that Microsoft has sold 1.5 hoojillion of the devices and the holiday shopping season has only just begun. I’ve written before about how motion controls can create more immersion in players by engaging our sense of body location, but there might also be another vector in play. Over on his blog, author Jonah Lehrer has some interesting thoughts about how buttons free controllers like the Kinect affect our emotional reaction to games given that physiological and mental states present psychologists with a bit of a “chicken or the egg” problem:

Let’s say we are playing a shooter on the Kinect. Unlike other game consoles, which leave us stranded on the couch, this console (like the Wii before it) actually makes us move. If we want to kill off the bad guys, we need to run around and break a sweat. We are no longer just twiddling our thumbs.

In order to prepare for all this combat, the brain automatically triggers a wave of changes in our “physical viscera,” such as quickening the pulse, flooding the bloodstream with adrenaline, and contracting our intestines. While even stationary entertainment can lead to corporeal changes – that’s why the heart rate quickens when watching a Hitchcock movie – the physical activity of the Kinect exaggerates these effects. Although we might look a little foolish flailing around the living room, the game has managed to excite our flesh, and that means our emotions aren’t far behind. As a result, we are more scared by the possibility of virtual death (and more thrilled by the virtual victory) because our body is fully engaged with the game.

Lehrer argues that high definition graphics and surround sound offer diminishing returns, so kinetic movement is the next big win for game designers wishing to engage us in their game. This is hardly an unprecedented idea. In his book, The Science of Happiness Stefan Klein¹ notes that “As [neuroscientist Antonio Damasio] reminds us, our mind is, in the true sense of the word, embodied, not ‘embrained.’ A disembodied being would feel neither happiness nor sadness.”²

Xbox, dashboard! Xbox, smile! Xbox, love! LOVE, XBOX, LOVE!

In 1993, researchers Paul Ekman and Richard Davidson even studied this question scientifically by testing to see if simply smiling can make you happy.³ All of us can fake a smile of one sort or another even when we’re pissed or bored, but it turns out that “true” smiles –those that erupt whenever we’re genuinely happy– involve a specific muscle: the obicularis oculi. This is the muscle around the eyes that causes us to make that particular, gleeful face during moments of unmitigated merriment. Some people can fake using the obicularis oculi to make apparently genuine smiles⁴ and Ekman and Richardson screened potential subjects for their study based on this criteria and then trained them further on how to do it at will. After taking some baseline measures, the researchers found out that faking a “real” smile led not only to higher self-reports of good moods, but brain activity as measured by EEG⁵ during fake smiles was practically identical to activity measured during genuine amusement.

But it’s important to note that the subjects had to smile the “right” way. Those who didn’t manipulate the obicularis oculi and related muscles didn’t become happier; they just looked a little bit like it. If the Kinect and other motion control game devices are going to trick our bodies into making us feel more engaged or emotional, they’ve got to do it convincingly and really mimic those genuine physiological reactions. They also need to either put out some games that will entice us to play, or offer us $5 and 10 extra credit points for our Psychology 101 class.

1. Klein, S. (2002). The Science of Happiness. De Capo Press. [↩]
2. Lehrer also references this idea from Demasio in the blog post I liked to, but grad school taught me nothing if not how to pad out my references. [↩]
3. Ekman, P. & Davidson, R. (1993). Voluntary Smiling Changes Regional Brain Activity. Psychological Science, 4 342-345. [↩]
4. Stefan Klein seems to think it’s hereditary [↩]
5. Or “electroencephalography” for those with more time on their hands to pronounce really long words [↩]