Thoughts on "Godel Escher Bach" - High Level Chunking

[LMNO]

Been re-reading GEB, and had some thoughts about it.  I'm in the section about chunking, levels, grouping, and chess. 

First thought:  Hofstadter says that when chess masters look at a chess board, they have trained themselves to look only at possible moves.  Further than that, they look only at beneficial possible moves.  I was considering this, and came up with an alternate chess rule:

-After every seven pieces taken from your opponent, you may move one piece any way you wish, only once, with the exception that you can't checkmate the king.  You may do that at any time after collecting the seven pieces

That way you do have to play by the rules, but up to a point: There is always the possibility of a random action.  This way, you are forced to think about all the impossible moves, as well as the possible ones.

That was more of a side thought, though.  The idea I wanted to bring up was from the section on "The Trade off between Chunking and Determinism" (pp 306 of the 1980 edition):

There is, however, perhaps one significant negative feature of a chunked model: it usually does not have exact predictive power. That is, we save ourselves from the impossible task of seeing people as a collection of quarks (or whatever is the lowest level) by using chunked models; but of course such models only give us probabilistic estimates of how other people feel, will react to what we say and do, and so on. In short, in using chunked high-level models, we sacrifice determinism for simplicity. Despite not being sure how people will react to a joke, we tell it with the expectation that they will do something such as laugh, or not laugh – rather than, say, climb the nearest flagpole (Zen masters might well do the latter!).  A chunked model defines a "space" within which behavior is expected to fall, and specifies probabilities of its falling in different parts of that space.

What I was thinking is that it works both ways... That is, in a certain situation, we have "chunked" the possible behavior of others, eliminating the "impossible" or the "highly improbable".  In some instances, we may even eliminate the "unlikely" or the "probably not"!  In most cases, this happens unconsciously, the lower levels of experience and expectation simply taking over.  Unless we really pay attention, we don't even see it happening.

But this same "chunking" also lurks in our own behavior in situations.  We have unconsciously eliminated certain behaviors and reactions from our possible choices, without even noticing.     Through whatever ways our mind sets up (call it 8-Circuit, or Monkey Mind, or Jungian, or Dianetics, et al), we have radically limited our behavior in how we react to certain situations, the limitations becoming less and less obvious the older the structures of our mind get.  It takes a lot of effort to notice these "chunks," and to act outside of them.  Some people never do.

Please keep in mind that "outlandish" behavior can be chunked, too!  That's why some of these pinealists are so predictable, because even though their behavior might be outside the standard observer's expectations, they are often fairly consistent in their outlandishness.

I'm sure you can easily see the BIP looming on the horizon, so I'll leave that to y'all to draw connections.  I would like to say, however, that in light of the above, one of the things that has drawn me to Erisianism is the way it attempts to avoid high-level chunking of models, if even just in a minor way.  It seems to not only teach us to accept experiences that occur outside our "chunks" of others' expected behavior (thus expanding the possibilities in our brain), but it also can teach us to examine our own behaviors, and how we have automatically eliminated certain responses... Which again expands possibilities in our brain.

[AFK]

Interesting stuff.  I liked the example in the excerpt.  It certainly is true, when I tell a joke I do have a certain number of expectations as far as what will result.  Laugh, groan, someone leaving the room, a kick to the groin...but I certainly don't have "climbing up a flagpole" or "baking a choclate cake" as expected results.  I need to ponder this further. 

[Cain]

One the first one, it reminds me of the fictional game of Sava introduced in R A Salvatore's novels.  It is like chess, only once a game, you can roll a die, and if the correct symbol comes up, you can move any one of the enemies pieces, to illustrate the backstabbing and opportunistic nature of the drow, and how conditons of war can help this come about.

Sava was an intricate game representing a war between two noble Houses—at least that was what it currently represented. Pharaun had seen an antique set that recapitulated in miniature the drow's eternal struggle with another race, but such pieces had gone out of fashion long before his birth, probably because no player had wanted to be the dwarves.

With its gridlike board regulating movement and its playing pieces of varying capacities,sava resembled games devised by many cultures, but celebrating the chaos in their blood the drow had found a way to introduce an element of randomness into what would otherwise unfold with a mechanical precision. Once per game, each player could forgo his normal move to throw the sava dice. If the spider came up on each, he could move one of his opponent's pieces to eliminate any man of its own color within its normal reach, a rule that acknowledged the dark elves' propensity for doing down their kin even in the face of a serious external threat.

[LMNO]

Interesting.  I'm gonna have to google R A Salvatore.

to RWHN: this may be a cruical, if overlooked, aspect to the original purpose of O:M -- That something happens that is completely outside anything expected by the target.  The trick, of course, is finding out how the target "chunks" their expectations.

[Cain]

He's a hack fantasy author, if it helps.

[AFK]

I totally, 100% agree with that.  I know the Turkey Curse from the PD is old hat now, but the idea, the motivation behind it is still spot on.  You're right, the trick is finding out what is outside the target and it's a moving target.  The Turkey Curse wouldn't work anymore.  Although, a 50 year old, straight-laced, CEO doing the Turkey Curse probably would.  

[LMNO]

He's a hack fantasy author, if it helps.

It figures, he's from Leominster, MA.

[Cain]

Societal convention.
Tradition.
Status.
Gender.
Age.
Political affiliation.
Past patterns of behaviour.

That's a good start, I would think.

[LMNO]

Good start, indeed.

I "societal convention" sort of like what clique you are in?  Yuppie, Slacker, Junkie, etc.

[AFK]

I would add profession/field of work.

There tends to be certain sets of expected behaviors in different fields.  It's a generalization of course, but in any kind of gathering or meeting you have enough of those types in one place.  

[Cain]

Is "societal convention" sort of like what clique you are in?  Yuppie, Slacker, Junkie, etc.

I was thinking more along the lines of the models of how we understand society works.  For example, the current societal conventions would be game theory, capitalist ownership theory, the nuclear family (successful or failed)...essentially what Marxists would refer to as cultural hegemony, the way in which life is understood to be in the common consensus, the ontology of the social now.

I would say jobs also fall under status, in that jobs are closely linked to income, which often determines the norms for your behaviour, based on informal class groupings.

[LMNO]

It would also seem that the "chunking" of what is percieved as "intention of humor" can easily sink an O:M, in that if they realize it's "only a joke", they can dismiss it.

So in and among the above, you've got to play it totally straight.

[Cain]

Yup. Humour is an understood and designated term with explanatory powers when it comes to behaviour.  It allows for almost any action to be sidelined.  Art is another one.  Once an action is evaluated and appointed an explanatory slot, it can be safely ignored.

[LMNO]

As RWHN says, it's a moving target.

I'm getting this visual, of all these big, round targets swinging like erratic pendulums, and your mission is to shoot an arrow that misses all of them.

[AFK]

You know, this reminds me of some work I did awhile back.  I had to go into a county jail to do some focus groups with inmates.  I have to be totally truthful and honest, when I went in I had some preconceived ideas of how the inmates would react and behave while I was asking them questions.  No, I didn't expect them to jump on me and kill me or anything dramatic like that, but I did, I had a certain "chunk" of behaviors I expected.

As the focus group went on, I quickly realized that I had those preconceptions and recognized how I was feeling when I saw them behaving outside of those expectations.  I in-turn developed quite a bit of empathy for them.  And then of course I had that expected behavior that typically resides in my field of work where I wanted to save them all.  Stupid little monkey that I am.