Playing Deeper

Jan 16, 2006 13:10

Hypothesis for more thought: Gaming and Perceptual Control Theory

I'll define "Games of Identity" as those games where you have an avatar with which you identify placed in a reactive world. The avatar may be personally created and defined (e.g., any RPG), provided by the game itself (e.g., Tomb Raider), or implied by a recognizable first person point of view (e.g., Doom). This definition would exclude games that lack such an avatar (e.g., Sim City, most RTS games, card games) as well as games where no identification is in place (examples?).

The basic idea: As Games of Identity become better - by which I mean more immersive, more engaging, where the sense of identity becomes stronger - feedback from the game is incorporated into the player's internal control hierarchy with fewer (or no) filters. Thus, "game" goals become "real" goals, "game" events become "real" events - because, due to their immersiveness (quality?), they are perceived as meaningful.

Or, to put it another way, as the sense of "I" gets broader, more feedback is incorporated.

But: Is this just an error in the system, a focus on a too-low-level control system?

Also: If the control systems are hierarchical, what do you call the one at the top?

In Extension: You should be able to take advantage of this and create Games of Identity which reinforce certain behaviors.

A variation of this applies to noninteractive stories - books, movies, plays - where identification with a character takes place. Here, however, the differences between the actions of the character and "what they should have done" tend to reveal higher levels of the user's internal control structures, possibly bringing to light a higher level control system capable of resolving a lower level conflict. We call this catharsis.

I should probably learn more about PCT.

games

Previous post Next post
Up