If only one thing can be learned from the idea of pattern recognition challenges, it should be how to drive a player through one. The biggest problem in games these days is a kind of bored ennui - the player moves forward simply because there is no other place more interesting to go. Most of the most lauded games of all time - including System Shock II, which I rarely denegrate - suffer from this problem. Long periods of boredom, where the only thing keeping you going is the inertia of having already played it for five or six or fifteen hours.
The critical thing to remember when designing a game is that the game is IN THE PLAYER'S HEAD. It doesn't matter what algorithms you use. It doesn't matter how good or bad the programming is. The only things that matter are the things that the player perceives - and HOW he PUTS THEM TOGETHER.
Approach EVERY piece of your game like this. Don't think, 'oh, this is interesting because you get a new weapon and see a new bad guy'. Think in terms of the pattern - and the various driving forces that the player is guided by. A new gun is only useful if it actually changes the dynamic of the pattern. Don't think of it as a gun. Think of it as a part of the pattern. How does it change the pattern IN THE MIND OF THE PLAYER? A new gun is a vat of potential in the mind of the player - until it's been fired a few times. That ADAPTATION TO PATTERN CHANGES is the core of pattern challenges, and I'll talk about it later. For right now, just think in terms of what is going through the player's mind in terms of gameplay.
That should ground you solidly in holistic game design - no more fumble-fingered creeping featurism. But I said this was about driving your players, not about manipulating the pattern of the game.
Well, it is. I'll explain in Part II about how to convince your players that they are having a grand old time and want to move forward in the game. I'll cover such topics as: "I want to explore, not move forward" and "god, this is boring!"