i heart auditory neuroscience

i've been on a bit of a brian eno kick the last week or so. i just impulse-bought another of his albums: nerve net, from 1992. it's really interesting music. as you should expect from mr. eno. he's better than anyone at creating aural landscapes: check out his "music for airports". it is absolutely just that - listening to it you feel as if you're waiting for a flight or changing planes. "nerve net" is a bit less ambient and a bit more upbeat, but it's still very richly-textured, slowly-evolving ear candy. i'm a fan. more than that, i think that he's probably got a decent grasp on how auditory cortex works and how music really is formulas, and you can use these formulas to create a predictable outcome in the listener's experience.

i've been reading jeff hawkins' on intelligence, and i'm not quite far enough in to do much in-depth analysis, except to say that i think he's missing a huge point, but i will withhold final judgement until i'm finished with the book. but it's got me thinking about computational models of cortex, and ways to simulate how our brain works, or at least to understand it. hawkins holds to the theory that there's a common cortical algorithm; that is to say, all parts of cortex, whether auditory, visual, etc. use the same fundamental method to find patterns in the environment. it's an interesting idea, and i definitely think there's probably some truth to it. but to say that in an adult human, visual cortex is the same as auditory cortex is the same as prefrontal cortex isn't exactly going to be accurate. though it's been shown that developmentally, they are interchangeable. anyway... too soon to know for sure, i'll have to finish the book and then wait 20 years before i can really know whether or not he's right.

(side note: watch hawkins give a talk on how brain science is going to change computing. he's a big-picture thinker, that's for sure.)

the question that remains, of course, is what that cortical algorithm is, how it works. and, of course, whether it's still too early to understand the cortex fully, because we don't completely understand all the underlying structures, and since the cortex is like built on top of and inextricably intertwined with the "lower" brain structures, without a bottom-up understanding our top-down knowledge will be incomplete.

No comments: