Slashdot is powered by your submissions, so send in your scoop

 



Forgot your password?
typodupeerror
User Journal

Samantha Wright's Journal: Biology Help Desk: Volume Seven 19

Journal by Samantha Wright

Hey, remember these? I do. Vaguely. But vaguely isn't good enough, so here's another one. Since I've discovered that the secret to understanding machine learning problems well enough to implement them is mostly a matter of strategically procrastinating, it seems only fair that I should run one of these. So before cross-entropy actually starts making sense to me, what would you like to know?

This discussion has been archived. No new comments can be posted.

Biology Help Desk: Volume Seven

Comments Filter:
  • I remember them well.
    How about this: biological sensory perception. We have two ears, two eyes, two nostrils and so on, therefore what's called "bilateral symmetry". But would other, non-terrestrial lifeforms exhibit different sensory locations? For example, Arthur C Clarke's octospiders (see "War on Rama") were communicating through visual-only means and were... deaf. Now to me, being deaf is a bit over-the-top, because evolution (at least of an advanced life form) would imply the need to be aware of noise

    • The need for motion in a single direction is one of the most powerful factors involved in the development of symmetry. Worms, fish, sea jellies, et cetera—moving forward is simpler when the effort necessary to do so consists of a symmetrical contraction of muscles. Another factor, which does not always act at a global level, is conservation of body plan: fewer developmental control genes are required to create a symmetrical organism. We see this even in stationary examples like plant leaves. It's poss

    • If you're looking for something alien and weird in form or behaviour, you don't really have to look all that far away from natural examples to find yourself awash with ideas. The microscopic world is full [blogspot.com] of weird crap [youtube.com] all on its own, and you pretty much just need to scale things up to get a good idea of alternatives. (The first link is of the mouth of Caenorhabditis elegans, a very small worm that is frequently studied in biology, and often makes the news because of its well-studied nervous system.)
      • While I agree they're pretty spectacular, none of the examples is intelligent, which brings us all back to the philosophical question: can there be intelligence which looks wildly different from the general human shape?

        • So far the only requirements for intelligence seem to be (a) sufficient competitive pressure, (b) already having a reasonably well-developed brain, and (c) an ability to manipulate the environment so that the brain has something to do. Crows have been shown to be able to bend a piece of metal into a hook in order to fish food out of a bottle, a level of intelligence they developed in order to outcompete similar birds for food. It really does seem that it has to be a competitive pressure, though, not just an

          • First off, thanks for having these journal entries, very inspiring reads always.

            Once the subject is on Brain Infrastructure and perception, relating to higher intelligence I would like to ask:
            Leonard Susskind (A very smart theoretical physicist, one of the fathers of string theory) famously said that "We are all victims of our neural architecture." refering to the difficulty human beings have to natively visualize/comprehend higher dimensionality space (4+ dimensional space).

            What circumstances would aid an

            • The case is, as always, 'because they need to'—we have so much trouble with it simply because we never experience anything analogously, certainly not on a regular basis throughout our history. I think you'd pretty much need to come up with something that existed in more than three spatial dimensions, or that could travel arbitrarily through time before you'd find something like that, though. Maybe you can come up with some other problem an organism would have to solve that involve more than three axes
  • If I watch my toe get pricked with a needle, my brain perceives the visible image of the pricking, and the feeling of being pricked as happening at precisely the same time.

    Yet knowing what we do know about the speed of nerve-signals, the visible signal *does* arrive in the brain earlier than the feeling transmitted by nerves trough the entire body.

    What biological mechanism is responsible for this integration ? How does the fakery work ? Delaying the images from the eyes would "work" but at the cost of makin

    • It takes about 70 ms for a signal to propagate from perception to response (e.g. compulsively moving something out of the way in response to touch.) Pain takes longer, because what it really tells us is that an injury has occurred, not that we're being exposed to a source of injury. If a response to a stimulus occurs immediately after an action, we may even perceive it preceding that action. I believe the entire brain is responsible for this phenomenon, but I'm far from an expert on the matter; this [wikipedia.org] might b

A debugged program is one for which you have not yet found the conditions that make it fail. -- Jerry Ogdin

Working...