Alright, so a lot of the stuff I'm going to talk about is quite cutting-edge in psychological game design, and the research is literally still in-progress. (In fact, I was just a research guinea pig a couple of hours ago).
Short version--I know a lot of you are familiar with stuff like the Myers Briggs (y'know, ENTJ and all that). Now, it turns out that the Myers Briggs test is owned by a corporation, who won't let outside research psychologists make changes to it. Most research psychologists in motivational psychology today are working on a model called The Big Five--and there's mountains of research done on it, and how it varies across genders and across cultures. Which means...well, I know how popular personalty tests are on the DL, so I'm just going to link this before people TL;DR
http://www.personal.psu.edu/~j5j/IPIP/Take test, post scores, whee!
------------------------
So...here's where it gets interesting.
http://www.gamasutra.com/view/news/165358/gdc_2012_applying_psychology_.phpAnd if you want to look through the slides:
http://www.darklorde.com/wp-content/uploads/2012/03/5_Domains_of_Play_GDC2012.pdfhttp://www.darklorde.com/wp-content/uploads/2012/03/5_Domains_of_Play_GDC2012.pptxJason Vandenberghe started trying to see if there were some correlations between this model and the kinds of games people play and what kind of play they enjoy. And he found out that yes: an extremely strong correlation exists. Something like three or four test subjects in he was finding that he could predict what kind of games they would like to play and what aspects they would enjoy based on their test scores.
Interesting things too, like...people whose personality skews towards being an asshole, but that's not really socially acceptable, were saying things like "Well...in real life I'm actually really nice, but when I go online I grief and troll people all night." Jason's metaphor is that basically "we're the Sims". We have psychological meters that need filling, and sometimes games can fill these in ways that real life cannot.
So...the big five is five scores, and then 30 sub-scores. The interesting thing is that they are all bell curves and have two extremes, and the people at each extreme are looking for different things. This is to contrast with earlier player models, which looked something like...this:
This would be the famous Bartle types for MMO games, which probably many of you are familiar with. Ever wondered--what's the opposite of an achiever? Well...now we have an answer. (One of the 30 sub-facets has one end of the spectrum as achievement-striving, and the other as contentment). In fact, one way in which this is already changing the way we talk about game design, is that when someone says "players want X" the response becomes "well, yes, half of them do, but the other half want Y"
Now, this post is going to be a little devoid on specifics, because I've yet to use and apply the Big Five at all (just looked at it long enough to realize it plugs a lot of the holes apparent in other theories I've been using). Like I said, relating the Big Five to game design is brand new and shiny and only barely tested research, and I'll be posting more as I figure stuff out. I CAN, however, talk about player categorization systems, why they are useful, and how to apply them.
Player Categorization SystemsYou can actually take people, in a lab setting, watching them play games, film their faces, and detect different emotional states. Nicole Lazzaro's excellent research is definitely worth a read if you're interested in this:
http://www.xeodesign.com/whyweplaygames/xeodesign_whyweplaygames.pdfAnd there's a big advantage if your game can achieve multiple different emotional states. Because different people actually seek and enjoy different kinds of emotion. Magic the Gathering has been doing this for years:
http://www.wizards.com/magic/magazine/article.aspx?x=mtgcom/daily/mr11bIt has three player categories, and they actively intentionally design cards for each of the three categories.
Spike: people who care about winning, and showing off their skill
Johnny: people who want to make crazy card combos
Timmy: people who are looking for an interesting experience, and want a crazy story to tell
And...three years ago, this was the system I used when designing LFT (the FFT mod) with Laggy, as it was the best system I knew at the time. Actually, let me go on a tangent to a tangent.
1.3 vs LFT (OMG, Hal's going to ban me for reopening this discussion)
FFT 1.3 does its best to appeal to the challenge crowd and does so explicitly at the expense of all the other archetypes. Most of the combos from the original are dismantled (MP-Switch and Move MP Up were put one on Bard, one on Dancer, stuff like that). A lot of the skills are homogenized as an experience--swordskills are given charge time and MP cost, for instance.
For LFT...we specifically aimed to target the other magic the gathering player types. Instead of removing combos, we kept damn near every combo from the original, and intentionally added several more. We left Orlandu basically in-tact, because while neither of us understood the appeal of Orlandu, when we observed other people talking about the game, they would say "Oh man, remember how awesome Orlandu was?" We wanted to buff Knights, and neither of us were really into tanky melee characters, so we sought out people who WERE and had them tell us what kind of buffs felt satisfying to them. At the same time we didn't completely throw challenge-seekers under the bus--we made the game a little harder and added some optional superbosses.
Now, there's nothing wrong with being laser-focused on a certain kind of player. People who really skew towards that kind of psychological archetype are going to love your game. There are people who think 1.3 is much better than LFT, because of their psychological inclinations. At the same time, people have IMed me claiming that on the internet as a whole, LFT has spread further with broader appeal than 1.3--which is to be expected based on the psychological design decisions. When you do a pretty good job of appealing to multiple psychological archetypes, you tend to reach a wider audience than when you do an excellent job of appealing to one psychological archetype. The reverse of that is that I've seen people express the opinion that 1.3 is not only the best game ever but that it's lightyears ahead of anything on the market. That's the kind of evangelism you can only get by laser focusing on a narrow audience. (The highest praise LFT tends to get is "Well...FFT was already a 10/10 game for me, and LFT is generally an improvement, so I guess it's my new favourite?")
Player Categorization SystemsWhich is to say, there isn't a one -correct- way to use these models. Unless your goal is to make lots of money--in which case you want to appeal to the widest number of players possible. But it is perfectly valid to laser-focus on one audience. Where these models ARE really useful is helping you to understand players that aren't you. Pretty much anyone can design a game that they would like. But designing a game for a different audience is hard. For instance, there are games designed for little girls, and since most developers are adult male computer nerds, presumably the people who made the game had no interest in playing the game. How do you understand what's a good game design decision for someone that is not you? It's hard. And it's especially hard because a lot of the time people don't even REALIZE that not everyone thinks the same way they do. But not only that, if your goal is to design for a wide audience, then you need to know a LOT of people who aren't you, and not only that but know that these people actually exist and what they want.
The Big FiveWhich brings me back to one of the reasons I'm really quite excited about the big five, and how it's going to mesh with game design. The 5 categories have been broken down into 30 subcategories, and some of these are very much off the radar for game designers. When Jason interviewed me, one of the things we discussed was that I like silly characters--Yumei from Valkyrie Profile because she does the chicken dance, and Lilka Eleniak from Wild Arms 2 because she beats people up with an umbrella. So he asks me "Would you anticipate a game that does something like this?" And my response was "Well, I can only think of a few games with stuff like this, so if I did I wouldn't be anticipating very many games at all." And he was like "Yeah, I think we as an industry have been overlooking some audiences."
Work in ProgressI'd love to now tell you exactly how each of the 30 facets of the big five relates to game design, but...that's still being figured out. Jason's got some good ideas, but they're still changing and adapting rapidly as he gets more data. I will, however, probably have several follow-up posts with my
speculation about a lot of these categories. (Well...either that or I'll go into crunch and you won't see me for several months
).
