Here’s a great keynote speech on how we learn and why we play video games. Pac-man gets horribly abused in the middle of the speech, but Koster makes some interesting points anyway:

“People are really good at pattern-matching. I’m going to offer the vast oversimplification that what we think of as ‘thinking’ or consciousness is really just a big memory game. Matching things into sets. Moving things into the right place, then moving on… Once we see a pattern, we delight in tracing it, and in seeing it reoccur. That’s meaning, all of a sudden. The brain doesn’t learn something the first time it sees it, it takes a while. You have to practice it… Building those patterns is necessary for our survival. If you don’t have a pattern library, you are going to die. You won’t be able to tell an apple from Draino. Fun is the feedback the brain gives while successfully absorbing a pattern.”

I think Koster is right here, and this is pretty much how the brain works all the time. We don’t have knowledge, we have constructed pictures. Trying to figure out if those constructs accurately correspond with reality is a task for epistemology, but attempting to answer that question forces you into an epistemic tailspin, because you need an epistemology before you can tell whether or not your pattern corresponds with reality, but you have to do pattern matching before you can come up with an epistemology. Which, in one sentence, is why I’m an antifoundationalist. But anyway.

“Here’s the thing: ask a gamer about grand theft auto’s hooker moment, they see this: pac-man eating a cherry. They’ve grokked it: it’s a power up.”

He’s trying to prove — by abstracting the in-game consequences away from the story –that gaining extra health by sleeping with and then killing hookers in the Grand Theft Auto games isn’t really as morally hazardous as it seems. Fine, but he’s chosen some unfortunate phrasing there. My childhood memories of pac-man will never be innocent again.

He goes on to talk about how the industry can make better, more engaging games:

“There’s a fundamental tragic flaw in games: the need to have one right answer without interpretation. We need puzzles when there is more than one right answer, games that can be interpreted; if we want games to become art and not mere craft, we have to get beyond this kind of simple thing [cartoon of a child’s drawing]. We need games with interpretation [cartoon of a master painting]. There are a lot of endeavours in human life like this. Writing a book. You come to it thinking you knew what you were going to say, but you learn a lot in the process. Music too. Things not expressed in the bare notation [cartoon of score with notes] - music is a finite set. Music is very mathematical. All possible combinations could be computed. Thank god for interpretation! All these challenges involve communication. Talking to your SO. One of the great cognitive challenges of life. It is a puzzle with no right answer. Perhaps that’s why we find it one of the most rewarding things life can offer. Fundamentally we have to regard games as being communicative objects, as media. They say something.”

No. This is where the Koster goes off the rails. He’s mixed up gameplay and story.

Gameplay is about pattern matching, as the author says, and it’s fun precisely because there is one right answer. There’s always a sense of progress, of challenges that you know you can solve if you try. That’s what makes games different than life. Why would you want to change that? Make games messy, open-ended, and unclear, and why would I play? Grinding away at Everquest is already periously close to grinding away at a nine-to-five, but (unlike the real world) at least the game is fair and I’m sure to be rewarded for my effort.

Games shouldn’t be “open to interpretation” like real life is! Games are fun precisely because they are artificial - meaning that they can be trusted to reliably inhabit the “medium space” between boringly simple and unbearably complex. The trouble with interpretation is that we have to work at it, and there’s no guaranteed payoff. Real life has more than enough of that, thank you.

Here’s my take on the issue: games need better stories. Desperately.

Gameplay may be about pattern matching, but stories are about making meaning. Stories give us drama and connection and context, and they make us feel that the effort we put into the game matters. Sure, puzzles can be fun in and of themselves, but they’re much more fun when they have an extrinsic purpose as well. But modern game designers tend to skimp on the story, especially compared to other mediums of expression (like books). With rare exceptions — interactive fiction comes to mind — story writing for games is awful.

So, in my view*, that’s the next step for game designers who want to “make art”. The industry has gotten pretty good at turning out gameplay-driven games. We could always use some innovation there, sure, but mostly variations on the theme will be enough to keep us interested. Where games usually fall down is in their writing. They need to hire better writers.

We started off writing stories on cave walls, then we progressed to parchment and paper, and finally we moved on to making movies - and now we have computers. But right now the comic book industry hires better writers than the gaming industry, and that’s just pathetic. Fortunately, it’s also easy to fix.

* Being the Media Mogul that I am, I’m quite sure that my views will be heeded. I look forward to playing Wallace Stegner Presents: Angle of Repose next Christmas.

via Sua Sponte.