Tuesday, October 12, 2010

Mirror's Edge, Player Psychology and the Implied Designer

(screenshot by Dead End Thrills)
Over on Wesley Tack's blog, he's posted some screenshots / comments about how he approached the Mall chapter of Mirror's Edge. Check it out.

(NOTE: I criticize the level a lot in this post. It's not because I dislike Wesley or DICE or the game. It's because I think it represents a popular way of thinking about levels, a way that I dislike. No offense is intended. Let's all be super friends, criticizing each other constructively.)

I think his commentary is most insightful when talking about the development process, like when he moved the highway portion to the upper part to save himself the time of actually building the highway / when he replaces all the BSP to minimize seams. I have to say though, some of his player psychology assumptions don't really strike me as correct. For instance, when he says:
[the mall logo] is visible in almost every section of the level and guides the player in the correct direction at all times
... That wasn't my experience playing at all. I only went that direction because there was nowhere else to go; it wasn't because the narrative told me I was supposed to go to the mall (I was disinterested in the developer-authored story by then) or because I was "guided" by the logo. I never even looked up for long enough to notice the red "Ne" logo there. I was too busy looking for red props to follow, or trying not to die arbitrarily from the gunfire.

And then I ended up at the mall somehow...?

In general, I'm very skeptical when designers say things like this -- the implication that a designed feature has that great of an effect on gameplay. That's not what's really going on here, in my opinion. "I extruded the bricks and made them darker to guide the player here." Riiight. Sure.

I mean, YES, on an extremely basic level we can guide the player. In Left 4 Dead, players gravitate to light because the shadows might hide boss infected. In Half-Life 2, put a battery somewhere and a player will run off to grab it. These all rely on relatively mechanical impulses, reflexes tied closely to "not dying" in the game; we follow these breadcrumbs because of a first person view that requires light to see, or a health / armor mechanic that protects us from hostile gunfire.

What mechanic does a "look up at this beautiful skyline and see the small red logo on this slightly different skyscraper" breadcrumb rely on? It's probably the least likely thing a player will do in this level, especially when they're stuck in the innards of the bridge and can't even see the sky. Or maybe the player's as bad as me and dies 10 times trying to climb up into the catwalks.

The only way you could direct the player's visual attention reliably, I think, would be like the pre-battle vista before the Nexus siege in Half-Life 2; featureless walls, a locked door, a window, and Barney walking over and telling the player to look through the window -- or similarly, a locked elevator door, featureless Combine walls and Alyx walking over and telling you to look at the unstable core of the Citadel.

Those work because they're extremely explicit and artificial; I see a Combine door lock, so I know I can't progress without Alyx, so I'll probably pay attention to what she says because I don't want to miss the voice acting and facial animation and gee-whiz-technology. "Games, wow!"

In this sense, I think players are always aware they're playing a video game. They don't actually believe the train on the movie screen is actually headed towards them. I pay attention only when I'm locked inside a room because then I know that something cool is going to happen and I don't want to miss it because I paid $30 or whatever to see it happen.

I was reminded of this in March 2010 by something Robin Walker said to me, during his crazed drug-trance stream-of-consciousness playtesting of Handle With Care: he basically stopped playing for a second, looked at me, and said something incredibly profound like, "At this point, I would blow up all the crates until I had one left, because I would want to see all the possible cutscenes and content and make sure I didn't miss anything. Then I would quick save and quick load as necessary."

... And he was right. That's what many hardcore gamers do. Who didn't save their game, near the end of Deus Ex, to play through all the different endings? Who didn't read the pages out of order in the "Choose Your Own Adventure" books, to read endings you otherwise never would've seen? Assuming it's a good, interesting game -- players want to explore the entire space of the game.

The tradition of "finding secrets" is alive and prospering, even more so with the achievement-philia strangling the design of Starcraft II. We want to get our money's worth, even if we didn't pay anything for Black Mesa Source -- but doggone it those bastards still owe us something, right?

We want to see everything, as much as possible, as efficiently as possible. Collect the Zerg research points, the skulls, the packages, the Nirnroot. Optimize your army, your tech-tree, your equipment, your attributes, your skill points. Maximize your agency so you can see everything and do everything.

Sirlin talks about how much he loves the secrets in Donkey Kong Country -- he has even invented some sort of designer personality to help him create a heuristic that finds the secret bonus coins. "Where would He put the big gold coin in this level?" There may or may not be a single designer who hid those coins everywhere, but it doesn't matter because Sirlin made one up anyway. To some extent, we internalize an implied designer and strategize accordingly, so desperate our desire to collect these coins!

Valve has effectively automated this to create a sense of intentionality. "The AI Director is fucking us over." Well, no, it's technically just a bunch of random seeds and presets that balances itself to what Valve's considers is "good pacing" in a video game. There's no personality there, but the construct is useful for us as players to understand what's going on.

So, to summarize:
  • I am skeptical of a theory of hardcore player psychology that makes heavy use of high-level art direction.
  • Some aesthetic effects seem real to me, but those are largely tied to functions of culture and genre tradition.
  • I favor a model that presupposes a partial suspension of disbelief as well as the infallibility of behaviorism, that the player's actions betray the player's intentions -- and these actions are guided by the player's conception of a designer, an "implied designer" (much like an "implied author")
  • In the absence of traditional mechanics that are made obvious and explicit, like "notgames" by Tale of Tales, the traditional hardcore audience will fabricate mechanics.