Selfish and Feral Modules (was: Re: Routing and AI algorithmic efficiency)

From: Damien Broderick (
Date: Tue Feb 12 2002 - 19:00:56 MST

At 11:21 AM 2/12/02 +0100, Anders wrote:

>brain seem to be very modular on some levels (lobe interconnections,
>cortical minicolumns) and highly connected on others (intercortical
>connections, inside minicolumns)

Borrowing this observation for my own fell purposes, I raise the question
whether some mental illness might be usefully modeled as an internal
contest between hardwired or firmwired modules, akin to a Dawkins' gene or
meme competition. What's in it for such selfish modules? I don't know.
Maybe everything with a kind of boundary just wants to be free. Well, in
fact, many subprocesses *are* kinda free and bottom-up anarchic, with
emergent coordination; maybe the feral thing is when this gets out of hand.

Relevant to this broad suggestion, perhaps, is the very good article by
Prof. Susan Greenfield in a recent NEW SCIENTIST (02 February 2002 issue,
`Sensational minds') where she argues interestingly for consciousness as
linked to sheer number of neural components activated in concert; I don't
have the article with me, and it isn't downloadable yet from their site,
but she proposes interactions between cognitive structures, affective
modules and the immune system. Depressive illness are linked to activation
patterns that exclude or downplay sensory inputs from body and world,
schizophrenic and bipolar disorders are those where too much unselected
noise drowns out cognitive control. Something like that, in brute summary.

Damien Broderick

This archive was generated by hypermail 2.1.5 : Fri Nov 01 2002 - 13:37:39 MST