There is a Gamasutra article about the psychology behind games. For those of you that read this sort of stuff fairly regularly, you won't find anything astounding here. It does offer some nice insights, but nothing that you can't find elsewhere (and, mostly likely, have already found elsewhere).
It always strikes me when they list their 'types of play' or 'needs' or anything else they feel like enumerating, they're ALWAYS simply listing 'pattern recognition challenges on THIS level' or 'pattern recognition challenges on THAT level'.
For example, this article has four kinds of play: competition (overcoming challenges), chance (minimizing chance and guessing results), vertigo (seeing the world in a new and unusual way), and make-believe (seeing a new and unusual world). All four of those kinds of play are simply pattern recognition challenges.
The other things they talk about - positive and negative rewards or difficulty levels, for example - are just ways of setting up the challenge. You can't succeed if there is nothing to succeed AT, or if it's too hard.
It strikes me that far too few people actually LOOK at what they are saying. Now, I'm a huge geek, so I like unified theories. And the unified theory of play seems to always boil down to OFFERING PATTERN RECOGNITION CHALLENGES. Play is simply an opportunity to INTERACT with a pattern. In what ways the pattern acts makes the different kinds of games - some of these challenges are fast and simple, some are slow and complex. Some are social, some are physical, some are mental, some are rotational. Some are based on real-life patterns, some establish a pattern in-game from scratch, some are a combination.
WHY people INSIST on categorizing this into different 'kinds of play' never made much sense to me. At first I thought that it was for ease of use. You know, "if you're making a game, think about these kinds of pattern challenges". It seemed helpful. But their lists are always incomplete, because there's a trillion kinds of pattern challenges. And once I got past these 'tutorial' methods, I expected to find them saying, "okay, that was just to get you started. What play REALLY is is pattern challenges, and those are just some common kinds. So how you really design these things is..."
But they never do. Not anywhere. Is it just fundamentally understood? Or fundamentally MISunderstood?
It gets mentioned in this article. He mentions how games are all about interacting with patterns. Yaaaaay!
So why doesn't he DO anything with that? He just MENTIONS it and then starts BABBLING this outdated rhetoric about how games are escapism. Yes, he does it very nicely. He brings up a lot of solid commentary. But it's like having the keys to the car and commenting for hours on how they feel in your hand. DRIVE, damn it!
Maybe I'm missing some fundamental key as to why people never take this tack. So I'm gonna take it. In the near future, I'll start posting musings on OFFERING PATTERN RECOGNITION CHALLENGES. Keep your eye on me!