« Getting Nearer | Main | In Praise of Boredom »

January 17, 2009


It seems to me that most near-thought has to do with private property: what's mine, what's yours (and when) etc...which, by the way, is the source of most disagreements. Far-thought is necessary to "elevate" oneself from this basic way of thinking, but too often it applies 'talis qualis' near-thought patterns to general abstract matters that have nothing to do with private property.

I worry that if you're allowed to postulate "detached detail", then you're allowing arbitrary thoughts to be tagged as ideology-fodder regardless of whether they're processed in fine detail or huge abstractions.

I would suggest, rather, that while church rituals were invented by people who took their God much more seriously than modern theologians, people who now sit in a church and passively experience the religion, may process it as Far.

In other words, the author's experience is not the reader's experience. (In my own experience, writing a story seems to involve at least ten times as much emotional energy as reading a story.)

The empirical predictable here would be something like: "People who invent fine details for their religion usually take it more seriously and show a noticeably greater tendency, albeit an imperfect one, to act as if the religion is actually true (even when this would be somewhat costly)." In other words, a breakdown of the ideology-isolator.

"[...] our intentions tend to be much more real to us than our actions, and this can lead to a great deal of misunderstandings with other people, to whom our actions tend to be much more real than our intentions."
— E.F. Schumacher

Most religious rituals (incense, rhytmic music, chanting, synchronized movement, boring rational thinking with sermon, vertically positioning priest in high status position relative to churchgoers, architecture etc.) have very practical psychological utility. Their purpose is to help people reach an altered state of mind, to help people have "religious experiences". Religion is organized mass hypnosis, opiate for the masses. This does not refute your point, but reinforces it: altered state of mind increases the vividness of the stories you tell yourself, it makes them more real.

There is also an upside for making distant thoughts concrete. All new creative endeavors start by visualizing the end result. In the beginning this vision is hazy, but it gets more concrete as you add detail. More detail increases the likelihood that you will in fact succeed in producing the end result you envisioned.

What is the point of distinguishing abstract and specific thoughts, and postulating corresponding near and far subsystems (even brain regions!), as you originally introduced the divide, if the actual criterion cuts across that and simply states that far thoughts are those less constrained by reality? As I noted before, it looks like distinguishing individual thoughts is enough, with weaker thoughts more easily seduced by emotions, while stronger thoughts more connected to reality. In the context of this post, strong thoughts are more indicative of reality, of more value if you want to learn about reality, and so you'd want to distinguish when other people declare their strong thoughts as opposed to deranged weak thoughts.

Beware weak thoughts, which in themselves may look just like the strong ones, but there is no reason for them to reflect reality.

Eliezer, yes I am introducing a new distinction in this post, which reduces the simplicity of the overall picture. This must pay for itself in enough added understanding, or we should reject it. We must have lots more data on fiction authors than religion inventors, so testing would have to focus there.

Yes fiction authors elaborate their world view in more detail than fiction consumers, and for that reason would take it somewhat more seriously. And authors using near systems to imagine their worlds have to be and usually are realistic in terms of habits of thought ingrained into their near side systems. E.g., footsteps are spaced like heartbeats in time, and people can see chairs but not bacteria ten feet away. But it still seems to me that a fictional story can be more detached from image-relevant social reality than a true story, such as of your Aunt Frieda's wedding. Authors seem able to start with far-thoughts about what social reality should be like, and then fill in more details while remaining consistent with that far-vision. The data seem to support this.

Mikko, I'm saying that adding detail to your plan does make it more realizable if those details are attached enough to the right sort of other ordinary details, but not if you just fill them in mainly via vivid-wish mode. Imagining the location, size, and furniture of all the rooms of your mansion, and your daily routine in passing through those rooms, doesn't help much in actually getting you a mansion.

Vladimir, the point of making the near-far distinction is the impressive psych lit showing it really is there.

Robin, near-far correlates with strong-weak divide a great deal, and so for this reason if nothing else experiments support them both. On the other hand, there are examples that support strong-weak but oppose specific-abstract divide, ones you list here, and planning fallacy. You can redefine near-far, reducing its simplicity, to incorporate those cases, but this problem doesn't seem to affect strong-weak view.

So my question should read: What are the strong points of near-far compared to strong-weak? It looks like after adjustments near-far becomes strong-weak, you care only about how detached from reality thought content is, whether detailed or abstract.

SAI_2100: There are three general modes of thought not two: very near, near and far.

Me: What is the very near mode?

SAI_2100: Not only do we need a means of thinking about the details of external input for decision making (near) and a means of general social abstraction (far), we also need a means of thinking about our own identity (very near). Nothing is closer than the self.

Me: And the trade-offs between the near and far modes? The far mode is for social image and the near for accurate decisions?

SAI_2100: This is too simplistic; its true there are trade-offs in the modes of thought, but they are not totally inconsistent, nor is hypocrisy necessary. It would be more accurate to say that the far mode focuses on 'social communication', rather than 'social image' per se. Both modes involve effective reasoning. This is shown by the fact that Hanson's posts themselves are examples of 'far' mode thinking.

Me: But a good post by Hanson showing even the near mode of thinking has its own biases?

SAI_2100: Indeed. Note Yudkowsky's unconvincing attempts to counter. Far mode thinking is closely related to analogy formation and is actually a more powerful mode of thought than the near!

Me: Give me an example of an analogy which effectively illustrates the power of far mode thinking. Make it on-topic.

SAI_2100: Certainly. Divide reality into three categories: Physical, Social and Mathematical. In the physical domain, note the existence of physical-status uncertainty (Heisenberg uncertainty principle). In the mathematical domain note the existence of truth-status uncertainty (Godel's theorem). Form an analogy between quantum uncertainty (physical) and Godel uncertainty (mathematical) and then map to the social domain.

Me: What is this, and how it related to the topic of Hanson's post?

SAI_2100: For goodness sake it's obvious! The aforementioned analogy enables one to derive a new, as yet undiscovered (by humans) form of uncertainty in reflective decision making. Technically, this shows that it is impossible to consistently translate the high-level terminal values of any goal-seeking system with general reflective capabilities to precise lower-level mathematical utility functions and bounded physical actions.

In a nut-shell, that's exactly why there evolved three separate brain systems for far, near and very far thinking. Because of irreducible meta-uncertainty about the motives of both yourself and others!

Me: Wow! Thanks for the insight SAI.

SAI_2100: A pleasure little human.

After listening to this talk linked by Tyler Cowen, I think I understand this post better,


Dutton mentions how (wasteful) details can be important in art: "he didn't miss a note" (for a music performer), or "he didn't miss a hair" (for a painter). In essence because we value skill. The parallel is with the peacock tail, which actually makes the peacock more vulnerable. The message should then be to the peacock: overcome your tail? (Maybe just: beware of your tail). Also interesting was the difference between natural selection (sperm) and sexual selection (charm).

Interestingly, metaphors are often detailed, but they are used to convey abstract ideas.

Metaphors do not always convey abstract ideas, metaphors are analogies.

The comments to this entry are closed.

Less Wrong (sister site)

May 2009

Sun Mon Tue Wed Thu Fri Sat
          1 2
3 4 5 6 7 8 9
10 11 12 13 14 15 16
17 18 19 20 21 22 23
24 25 26 27 28 29 30