Are cognitive psychology experiments just bad video games?

Written by . Filed under shop talk. Tagged , , . Bookmark the Permalink. Post a Comment. Leave a Trackback URL.

the most boring video game ever?

 When I describe the experiments we ask people to participate in for our research, I sometimes just say that we ask people to “play very boring video games.” In the “lexical decision” game, for example, you watch the center of the screen, and when a string of letters appears, you press one button if you recognize it as a word, and another button if you don’t, and to do this as quickly and accurately as possible, until you run out of strings. The lexical decision game happens in a defined time and space that is separate from everyday life, it has a well-defined goal, and clear rules. In this way it shares some formal properties of games, although the motivation for “playing” is most often extrinsic (participant reimbursement in the form of cash or course credit from the psychology department).

All cognitive psychologists are in this way game designers, albeit to a limited extent — perhaps to the same extent that we dabble in programming, statistics, philosophy, linguistics, etc. Our games aren’t very good, because they generally aren’t designed with the player’s experience in mind. In those instances when we’ve created something so tedious that it starts to degrade the data quality, or when dealing with uncooperative participants, like babies and toddlers, who need a lot of cajoling to maintain their attention, the solution is often to add something outside the “game” itself to make the medicine go down better. And of course, our experiments tend not to have any artistic, narrative or entertainment aspirations. No, our “games” are instead designed to elicit and measure specific behaviors under a variety of controlled conditions.

Lately I’ve been thinking about how more intrinsically interesting video games might be used to conduct behavioral research. We have a population that is walking around with powerful computers in their pockets, and happy to share enormous amounts of information about themselves with various corporate entities in exchange for relatively little. Perhaps they could be sweet-talked into helping us unlock some mysteries of the human mind while they are busy unlocking achievements in some simple games? In fact one group has already had some success running a lexical decision experiment on people’s iPhones.

I always thought the problem with this would be that, as games go, our experiments are just not that much fun. In my experience, though, games don’t actually have to be that much fun to engender high levels of player engagement. This took me a long time to realize, even though when I play games I tend to gravitate toward things like the Civilization series: contemplative simulations in which you can test out the long term consequences of “a series of interesting choices”. While playing, I slip back and forth between detached hypothesis testing — “Is the benefit of optimal positioning vis à vis key resources worth spending more turns before building a city?” — to emotional responses that make me aware of the extent to which I’ve become invested in an artificial world — “Nooooo! Why are they sacking Persepolis? I’m just about to finish the Oracle!” The fact that the game world draws from history, but progresses according to its own logic can make this very confusing if you step outside the magic circle for a moment: on several occasions I’ve found myself invaded unprovoked by Gandhi during the Iron Age. Or, playing as the Iroquois, I’ve conquered a medieval New York with advanced fighter jets.

Although it provides a lot of enjoyment by allowing you to experiment with a complex system, Civilization often feels more like a job than a game. It is not too hard to spend five or six hours at play, even though “play” largely involves managing a dizzying array of resources and processes, and dealing with external constraints that interfere with pursuing your own goals. For example, I’m never too interested in building and maintaining much of a military force early in the game, and often have to go back in time and make incrementally more investment in my military industrial complex to avoid being overwhelmed by club-wielding Frenchmen while I focus instead on building the Pyramids or the Colossus. As much or more than anything like fun, the experience is characterized by anxiety punctuated with frustration and occasional relief.

Interestingly, though, it is possible to strip away all of what I tend to think of as the “content” of Civilization, and create a simple interactive “game” that is not particularly fun, but highly engaging.

My first experience with this was the original Tamagotchi. This was a little “egg” that you kept on your keychain. It hatched a digital “chick” that needed to be tended to in a variety of ways. It would get hungry, or sick, or sleepy (you had to turn out the lights for it to sleep properly, if I remember correctly) in real time. Thus, you would be in the middle of a meeting, and you’d get an urgent beep from your pocket because your Tamagotchi needed something. Oh yes, I just remembered that you were responsible for cleaning up its poop as well. So, you would break out of whatever meeting or life event you were in the midst of to go and clean up your digital pet’s digital poop, and then maybe check whether it was well fed, or needed to be “played with” (right! it had a happiness meter…BTW, I drafted this without internet access, and the effort of actually remembering these things is gratifying in a way that makes me want to start using Freedom more often).

If we can talk of this first iteration of Tamagotchi as having a goal, it was just to keep the thing alive as long as possible. When it died, of course, you could just start over with an identical one, with the hope of being more responsive to its make-believe needs. Unlike an actual pet, it couldn’t love you, cuddle with you or breed with others of the species. And yet Tamagotchis somehow multiplied explosively. This same basic “negative experience” strategy is key to a whole genre of “social” games that has taken root in the last few years: Zynga’s -Ville series series and the social world-building genre in general basically depends on the capacity of a persistent, real-time fantasy world to yank you out of your life at regular intervals for small doses of reinforcement that you “earn” by just showing up.

Ian Bogost refers to these as “Cow Clicker, in which you are rewarded for clicking on a cow. The game mechanics are a reductio argument about the pointlessness of social games, and yet it became a relatively successful example of the genre. Even after the “cowpocalypse,” some players continued to log in and click the space where their cows had been, in hopes of earning a diamond cowbell.

I had mostly shut out the noise of cow clicker games on Facebook, but in the process of trying to develop ideas for how to use mobile devices as a means of collecting data about human behavior, I stumbled onto their slightly more recent — if not more advanced — mobile cousins. Since the point of embedding experiments into games is to get as many people to participate as possible, I took a look at the Top 25 in the App store. And because I was about to download fifty or so games for, ahem, research purposes, I was particularly interested in free ones. And it turned out the top free games over the past few months had mostly been in this genre.

After experimenting with them, I can report that they are not fun in the same way the Civilization and Tamagotchi are not fun. Like Civilization, in a game like Crime City, or Paradise Cove, or Haypi Kingdom, etc., you are in charge of building a “world.” In the case of Crime City, your options are defined by a mix of different kinds of gangster tropes, so you can build a fish store in case anyone wants to send a “message”, a basketball court where young entrepreneurs can sell drugs, a collection agency, etc. By developing your territory, you can accumulate in-game currency, which you can then use to upgrade your equipment (night vision goggles, anyone?) and increase your ability to beat up on other players and steal money from them. You can also “level up” by taking on various missions in a very thin story line. The only real goal in the game is the accumulation of stuff. There’s no room for conducting diplomacy, or balancing your investment in different kinds of resources, or engaging in any kind of problem-solving at all. There is only the quest for more currency that in turn allows you to accumulate stuff that upgrades your ability to accumulate more currency and defend your property from other players.

The basic game mechanics are also incredibly simple. All tasks are completed by simply double-tapping on a target. The only strategic element here is that you have a limited amount of “energy” you can put toward completing narrative missions, and a limited amount of “stamina” you can put toward interactions with other players. You can think of these as two different forms of capacity for labor. Choosing to invest these in low-value activities like mugging another player who isn’t carrying any money, or attacking someone whose defensive abilities outweigh your offensive abilities are mistakes, but they don’t have any long term consequences. You can always come back and get more stuff later.

When you have used up your “energy” and “stamina,” you have to wait, in real time, for them to replenish. That is, while you are living your real life, out in the world, your avatar is resting up and gathering courage for the next round of action, and your fiefdom of mobbed-up fronts is accumulating cash that you must log in to claim. As with the Tamagotchi, the real-time element is key to making the boundary between game and real life permeable, and encouraging frequent interactions. As Paul Gestwicki points out (here, in the discussion of fixed interval schedules) this is also a blindingly simple solution to the problem of “how do we get the user to play more often?”, i.e., “Put them on a timer, and make them come back at fixed intervals.”

The fact that Crime City is, quite simply not a good game, and yet is wildly popular, opens up a number of potentially distressing questions about our relationships to our devices and the persistent virtual worlds they connect us to. But it also holds some encouragement for the potential to use these games as research tools. It’s not hard to imagine building the capacity for richer interactions between players into a platform like this, and using it to conduct behavioral economics experiments, for example.

Even in its current form, there is scope to ask interesting questions about people’s behavior over relatively long periods of time. Because robbing and assaulting other players is part of the game, I’ve had a chance to see how lots of other people play. There seems to be a wide range of approaches, from those who seem to be ignoring the world-building portion of the game and just leveling up by following the story (they have the bare minimum of buildings), to people who are using their territory to beef up their defensive capacities (they have lots of buildings that don’t earn income, but rather provide a bonus when being attacked by another player) to people who have clearly spent a lot of time decorating and arranging their territory, despite the fact that this confers no advantage at all in terms of game play (a barbershop earns the same whether it is facing the sidewalk, or hemmed in by six other buildings with no way to access it). Are there other types of players? Are these strategies associated with demographic factors? How do these same people behave in different game contexts? What questions would we want to ask them about their real-life behaviors?

Most importantly, could we develop games that people would actually want to play, and that would provide insights about language, attention, and other basic cognitive processes?

Image credit: screen grab from Desert Bus, perhaps the most boring video game ever devised.


  1. Nano
    Posted September 10, 2012 at 2:20 PM | Permalink

    As you point out, these games are an interesting dataset on what is treated as ‘reward’ in real life versus what we consider ‘reward’ in psychology & ethology.

  2. Posted September 12, 2012 at 1:17 PM | Permalink

Post a Comment

Your email is never published nor shared. Required fields are marked *


You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>


Get every new post delivered to your Inbox

Join other followers: