The latest from SlateStarCodex

[Q. G. Pennyworth]

Updating.

Interesting article about Bayes and neuroscience.  I like how he presents the findings, but still comes away with questions.  The hypothesis makes sense, but that doesn't mean it's right.

http://slatestarcodex.com/2016/09/12/its-bayes-all-the-way-up/

I didn't read it all the way through, but I can tell you that neuroscience-wise this guy reads like a stoner rambling about quantum mechanics with a very shaky baseline comprehension of what that means.

Is the top-down/bottom-up metaphor an okay one to use as a way of describing more complicated systems that are actually functioning, or is there something about it that's so fundamentally fucked that looking through that lens is going to lead me to idiot conclusions? Because I like it as a non-literal model, but I don't want to absorb it if it's going to make me look like an idiot.

[Mesozoic Mister Nigel]

Updating.

Interesting article about Bayes and neuroscience.  I like how he presents the findings, but still comes away with questions.  The hypothesis makes sense, but that doesn't mean it's right.

http://slatestarcodex.com/2016/09/12/its-bayes-all-the-way-up/

I didn't read it all the way through, but I can tell you that neuroscience-wise this guy reads like a stoner rambling about quantum mechanics with a very shaky baseline comprehension of what that means.

Is the top-down/bottom-up metaphor an okay one to use as a way of describing more complicated systems that are actually functioning, or is there something about it that's so fundamentally fucked that looking through that lens is going to lead me to idiot conclusions? Because I like it as a non-literal model, but I don't want to absorb it if it's going to make me look like an idiot.

Top-down/bottom-up is a profoundly flawed model for talking about biological systems, simply because there is no such thing as purely linear directionality in any complex system; there are feedback loops at multiple points along the way, and the information exchange is always bidirectional (if not multi-directional).

[Mesozoic Mister Nigel]

This diagram is a simplified but possibly useful example of the complexity I am trying to describe:

[LMNO]

Yeah, there's a lot of stuff going on.  Our brains really don't intuitively grasp that much complexity, do they?

[Mesozoic Mister Nigel]

Yeah, there's a lot of stuff going on.  Our brains really don't intuitively grasp that much complexity, do they?

We really tend to be binary-seekers, and I'm not sure how much of that is social conditioning and how much of it is neurobiology.