Game Design Theory -- Pt 1: Autonomy, Competence, and Relatedness

[metroid composite]

Well, I figure as long as this forum says "Game Design" I might as well throw in a game design topic or two.  If I actually finish what I start, I hope to gradually cover several topics in such a way as to be a useful reference.

-----------------------

TOPIC: Autonomy, Competence, Relatedness

PRACTICAL APPLICATIONS: These are used to make a game more compulsive and compelling.  In psychology labs, they are statistically more effective than any other measure at getting people to keep playing a game.  MMO developers and makers of Facebook games have put particular emphasis on these above everything else.  (Most likely you won't want to go that extreme, but these are still handy design tools).

REFERENCES: These are pretty hot topics right now, so...there's lots.

This one goes into a lot of good detail about the theory:
http://www.jonselin.com/uploads/Immersyve%20SRigby%20Talk_AGDC%202007.pdf

This has extremely little detail, but it's a TED talk and thus automatically entertaining:
http://www.ted.com/talks/dan_pink_on_motivation.html
Bits and pieces of it have even hit classical news sources
http://www.guardian.co.uk/technology/2011/may/15/video-game-design-psychology


-----------------------


The theory holds that motivation in the human brain is driven by three fundamental needs.  Competence (feeling like you're capable), Autonomy (feeling like you have choices), and Relatedness (receiving recognition from others).

COMPETENCE:
How often is the player's intended action the actual outcome?
How quickly do players feel they're on top of the in-game mechanics?  The in-game controls?
This extends to "my avatar was walking in a 3D space and fell off a cliff".  This extends to "I'm playing Portal 2, and it really looks like I should be able to make a portal on that wall, but I can't.  Screw this game!"
And yes, to be honest, the simpler and easier your game, the better you're likely to do on competence.

Examples of non-Competence:
Final Fantasy Tactics is a big offender here.  You need to use L and R keys to insert characters into the fight, which is non-obvious.  It's a 3/4 perspective where it's not clear which direction of the D-pad corresponds to which diagonal axis.  It throws a preponderance of mechanics at new players, and forces them to use many of them (like setting your orientation after you move--it's not like this is an optional extra that experts can turn on; even a brand new player has to do this after every move).

Note that different audiences are going to come in with different prior competencies.  For a fun example, I've heard it argued that if the Atari E.T. were released five years later, it would have been considered not that bad.  What E.T. did was context sensitive actions (i.e. "A button opens doors, talks, extends E.T.'s neck, uses tape recorder, etc depending on E.T.'s location") which at the time was horrifying inexplicable madness--no one could understand these controls!


AUTONOMY:
Players feeling like they have real choice.  Not only that, but that they are the cause of their actions.  (As opposed to, say, "I stepped forward and was entered into a cutscene where the main character punched the boss").  And measurably, the number of different game objects you can interact with (including levers, treasure chests, NPCs), the number of different modes of gameplay you can go into.  (This is why GTA offers 50 dumb minigames rather than doing one thing very well).

Examples of non-Autonomy
I'm going to skip the obvious "games with lots of FMV and forced linear paths without customization" examples, and jump straight to Portal.  (minor spoilers).  In Portal, there's a scene where GLaDOS forces you to do something that you emotionally really doesn't want to do--destroy an inanimate cube to which the player has grown attached.  There's no alternate path; no sidequest you can embark on to delay the inevitable, you're just forced to do it.  And this is actually brilliant design--taking away player autonomy makes the player upset, but it is framed in such a way that the anger is focused towards the villain, not the game designer.


RELATEDNESS:
Humans are wired to seek recognition and connection with others.  The good news for everyone working on single-player games is that this stimulus can be faked with NPCs (often more consistently than real humans, because NPCs will be more consistently grateful and positive than a random 14-year-old on the internet).  Statistically, people respond better to positive feedback, than to belittling feedback (so in Katamary Damacy, the King of All Cosmos telling you your efforts are pathetic when you complete a level may have been part of why the game never got that large in popularity).

Examples of non-Relatedness
I'm just picturing "Super Luigi RPG" here, where you walk up to NPCs and they say "I'm sorry, what was your name again?"  You stop a volcano from destroying an entire town, go talk to the townspeople, and they're like "Hey, did you hear?  Mario saved a kitten out of a tree!  The poor kitten couldn't climb down...."  You perform a super ultimate charged crouching jump, and some kid says "That's not impressive.  Mario can jump way higher!  That's why Mario's a hero!"


-----------------------


Note that there isn't necessarily an automatic right-answer to any of these.  I mean, there is if you want to make people compulsively return to your game without understanding why they like wasting time playing it.  But in a forum about made-for-fun ROM hacks, that's not necessarily the goal....

[Dark Holy Elf]

I certainly am not posting this to doubt anything posted here, since it all makes a fair deal of sense to me. I'm mostly wondering at how the opposite of some of those categories seems to be extremely popular at times.

Regarding competence, there is clearly this huge market for games that are quite challenging, usually killing the player repeatedly, which seems the opposite of what competence would suggest. Though you can argue it's all for the sake of making the player feel -that much better- when she finally defeats Phantom overcomes the challenge of the game.

As far as autonomy goes, I think this quote sums up my issue with it best:

Examples of non-Autonomy
I'm going to skip the obvious "games with lots of FMV and forced linear paths without customization" examples, and jump straight to Portal.

So uh things that have been extremely popular and successful basically? I've seen that particular Portal sequence get serious hype, and the FMV-intensive, largely-linear FF7 and its ilk are huge. I could point at other examples (e.g. 2D Mario games, which are about as linear as platforming gets, being the most popular games in the genre). The majority (though not all) successful games have a story which proceeds to a certain fixed end goal (rescuing the princess, beating the villain, saving the world, etc.). This isn't to say that some autonomy isn't good (those Sega CD games that were basically interactive movies weren't especially successful) but it very much seems to be something that needs to be used in serious moderation. It's not just about linear plotlines versus sandboxes, even: it's extremely easy for games to give players too many choices and be overwhelming.

[metroid composite]

A lot of these are things that I'll cover in future topics.  For instance, these three categories literally make no allowance for "fun"--I'll probably cover the psychological state of "play" next.  But yes, games that focus on competence/autonomy/relatedness above all else...well you end up with Farmville, and admittedly Farmville is more popular than...just about everything, but also not well-liked by "gamers" for precisely what it lacks.

I will dispute your claims of low autonomy, though.  Mario, for instance offers tons of choices: do I collect these coins by touching them?  By hitting a block underneath them?  By hitting something into them?  Do I not collect them because I might die by trying?  Do I not collect them in case there is a POW switch around here somewhere?  Do I not collect them to save time?  Similarly for enemies: Jump on them?  Jump over them?  Avoid them completely by taking the high platform?  Use them as springboards?  Kill them with fireballs?  Kill them with a shell?  Try to get a 1-up by chaining jumps?  Skip them completely with a warp?

Autonomy is actually something Mario does quite well (somewhere I have an article that tallies up all the different things you can do in the first 5 minutes of various platformers.  SMB1 was first with about 40, SMB3 was second with about 37, and none of the other platformers tested came close).

One of the things the linked powerpoint above notes is that sprawling is often actually bad for autonomy.  A compact room with 20 things you can interact with triggers autonomy better than a giant world with 25 things you can interact with, but which are separated by long distances.

As for competence in "Lunatic" style gameplay, note that Lunatic style games are very, very careful to preserve player intent.  There's no ice level in Touhou; no sense of acceleration.  No special moves caused by doing a fireball motion.  Just "you press the arrow key for X time, and you move X pixels".  I've never seen someone scream at a Touhou game "I pressed up, damnit!"  In short, what you want to happen is what happens on-screen, and this is a big component of competence.  Or, to say it the way Touhou fans would: "If you die, it's your own fault."

[SageAcrin]

I've never seen someone scream at a Touhou game "I pressed up, damnit!"  In short, what you want to happen is what happens on-screen, and this is a big component of competence.  Or, to say it the way Touhou fans would: "If you die, it's your own fault."

An interesting point that highlights this: Shmup players are spectacularly protective of this, to the point of near-insanity by any other standard. Fighting games are similar.

In damn near any other genre of gaming in existance, two frames of delay would be considered unnoticable. In both genres...well, I remember mauve talking about coding a rollback-based netplay caster for a fighting game he liked because, IIRC, "four frames was unplayable"(Might have the exact frames wrong, maybe it was five). A version of MAME has been recoded to shave off one frame of input delay from many major shmups-added because it added a lot of compatibility to other games, so said version of MAME would break most things besides these shmups, as I understand it. Many of ZUN's shmups actually do have noticable(if variable) frame delay(usually one or two frames, though PCB seems to vary a lot on the system, presumably same with EoSD), and damned if people haven't coded vsync patches for them(Which, as near as anyone can tell, he looked at and used code from, as his latest games are tighter.).

For the highest levels of competence based gaming, the highest levels of controls are absolutely required for many of the players, from what I've seen. A good parallel I'd like to make is Super Meat Boy, which does have weirdly Sonic-like physics and is not incredibly tight about it. Many competence-oriented players do seem to attach to it, but only if they get absolutely perfect FPS(as it has added input delay otherwise, on top of iffy controls) and even then, the market entertained by it seems to be smaller than, say, IWBTG, despite Meat Boy being objectively better even at controls(so I gather).

In other words, both seem to have a "I enjoy this game...from a distance, where I don't expect to beat it" along the same lines as traditional Angband/Nethack/Dungeon Crawl style roguelikes(A fundamentally luck-based form of challenge where decision making counters more and more luck the better you are at it, but where luck still reigns.). IWBTG is fundamentally more cruelly humor oriented about it's killing your ass, and is easier to beat through sheer persistence, so naturally it seems to generate stronger fans. (Though I dunno, this could be a false impression.)

Note that none of these comments are talking about the overall markets of such games-as you pointed out, Gamers are not the same thing as a casual market. I don't think any of them appeal to casual players. My brother is about as casual as it gets-massive Drake's fan, consumes FPSes in general like crazy, won't play standard RPGs anymore because they don't move fast enough for him, etc., very standard-and I couldn't talk him into playing IN Easy because it intimidated him too much. Chew on that for a bit.

[metroid composite]

I couldn't talk him into playing IN Easy because it intimidated him too much. Chew on that for a bit.

Yeah, I've totally seen that reaction when trying to evangelize Touhou.

As for IWBTG, it's almost a meta-commentary game.  Not quite to the level of something like Achievement Unlocked but I'm really not sure it would work if you're not already exposed to the stuff it's referencing and thus can "get" the joke.  I'm curious what would happen if you found someone who never played games before (but had excellent reflexes) and introduced them to IWBTG as their first game.  Would it flop horribly?

I've seen that particular Portal sequence get serious hype

Yeah, honestly, I think that scene is good, precisely because they use predictable negative emotions caused by loss of autonomy, and focus those emotions within the story (intensifying emotions the player feels towards the antagonist).  Basically, I think that you can make a very good design where you intentionally go against autonomy/competence/relatedness, as long as your goal of that particular section is to produce a negative emotion.

and the FMV-intensive, largely-linear FF7 and its ilk are huge.

Ok, let me go into a little more detail on this--while FF7 is FMV-intensive, it's not like there isn't a sense of "I did that".  It would be a huge design mistake if, for instance, the Sephiroth fight was just a big FMV with no gameplay--in that event, the player would not feel like they really beat Sephiroth.  But I mean...FF7 has a ridiculous number of options for in-game customization.  And even in terms of cutscenes you can often change which cutscene you get (both in terms of optional cutscenes, and "one of three paths" cutscenes).

Digging out a few other articles from my archive that I didn't realize related to Autonomy the first time I saw them...

Here's the Platformer Analysis article I mentioned, where holy crap there's a lot going on in SMB games:
http://www.gamasutra.com/view/feature/1851/a_detailed_crossexamination_of_.php?page=27
And here's an interesting note from the Penny Arcade people about Rhythm games, how they only like the ones where you can do some "musical improvisation":
http://www.penny-arcade.com/2008/3/3/