Thursday, October 3, 2013

The Fluid Mind and a Problem Dissolved

This post is doing double-duty. In the first place it links to and summarizes another set of notes that I've uploaded to my SSRN page: From Associative Nets to the Fluid Mind. But I'm also asserting that I've dissolved the problem that has, in a sense, guided my intellectual career for four decades.

That is the problem of Coleridge's "Kubla Khan" and its relationship to "This Lime-Tree Bower My Prison." They are very different poems, by the same poet. There's nothing unusual about that. Except that early in my career I decided that I wanted to create one model that would account for both poems, and several others besides (a group known as the Conversation Poems).

I took my first run at that problem in graduate school at SUNY Buffalo and failed. I tried again years later, early in this millennium after I'd finished my book on music, Beethoven's Anvil. That effort resulted in two very long articles which I like very much: "Kubla Khan" and the Embodied Mind (2003), and Talking with Nature in "This Lime-Tree Bower My Prison" (2004). Those clarified how the mind operated in those poems – at least to me. But they shed no light on the underlying model.

Well, I've now decided that that's because the kind of model I'd been looking for doesn't exist. It is, in principle, impossible. That old problem has thus dissolved.

What has happened over the years is that the terms of the problem have changed up on me and I've only just recently realized it. For the moment I'll date the roots of that change to the summer of 1999, when I wrote the notes on associative nets that constitute the bulk of the above linked set of notes (the section, Associative Nets, in the table of contents below). Writing Beethoven's Anvil forced me to differentiate between the brain and the mind and to think of the mind as neural weather. But it took over a decade after that for me to hook things up and realized that the terms in which I'd stated that old problem have now dissipated.

As I've said, that old problem no longer exists. But it's not clear to me just how to state the new problem or problems that have replaced it, much less how to go about resolving them. The first part of these notes – under the heading, Fluid Nets – Or, the Mind is Not the Brain – states those problems, more or less. The last section of this post, Implications for Coleridge, is from those notes.

Abstract

We can think of the mind as a network that’s fluid on several scales of viscosity. Some things change very slowly, on a scale of months to years. Other things change rapidly, in milliseconds or seconds. And other processes are in between. The microscale dynamic properties of the mind at any time are context dependent. Under some conditions it will function as a highly structured cognitive network; the details of the network will of course depend on the exact conditions, both internal (including chemical) and external (what’s the “load” on the mind?). Under other conditions the mind will function more like a loose associative net. These notes explore these notions in a very informal way.

Table of Contents

Fluid Nets – Or, the Mind is Not the Brain
The Mind is a Fixed Cognitive Network
The Mind as Neural Weather
Fluid Nets and a Poet’s Mind
Some References
Associative Nets
1. Introduction
2. Definitions
3. Why An Associative Net?: “US History” As A Gestalt
4. What Does This Get Us?
5. Reception Vs. Production Vocabulary
6. Antonyms And Other Associations
7. The “Real” Abstraction
8. Relations Between Clusters
9. Desire And Emotion
10. Two Forms Of Representation & Their Interaction
11. The Neural Basis
12. On The Overall Scheme Of Things
Some Associative Clusters
Implications for Coleridge

What really hit me, though, was the implications that this has for a problem I’ve been working on for years, that of the relationship between Coleridge’s “Kubla Khan” and “This Lime-Tree Bower My Prison.” Ever since my years at Buffalo with Dave Hays I’ve been looking for a fixed cognitive network such that one path through it would yield “Kubla Khan” and another path would yield “This Lime-Tree Bower My Prison.” Further, paths of that second type would also yield “Frost at Midnight”, “The Nightingale”, and “The Aeolian Harp”.

I figured that, once I’d defined such a network, I’d have the first tiny hint of a sketch of Coileridge’s mind. But I’ve never been able to figure out how to construct such a network. I’ve now decided that such a network doesn’t exist. His mind was in one state when he wrote “This Lime-Tree Bower My Prison” and another, and very different, state when he wrote “Kubla Khan.” Coleridge told us as much when he wrote that preface in which he asserted that the words to “Kubla Khan” came to him while he was intoxicated with opium.

So, it’s not as though there’s anything new in the idea that Coleridge’s mind was in a different state for “Kubla Khan” than for the other poems. That’s always been know. That’s always been part of the romance and mystery of “Kubla Khan.”

But it’s taken me awhile to connect THAT commonplace background information with a lot of other ideas. The important connection is with Kintsch’s distinction between associative and semantic nets. In effect, semantic and associative nets are different states of the same underlying neural structures. Much the same is true of the difference between “Kubla Khan” on the one hand, and “This Lime-Tree Bower My Prison”. These very different poems came from the same brain, but when it was in very different states.

The trick is to specify what’s going on in the operational “region” between associative and semantic nets on the one hand, and “Kubla Khan” and “This Lime-Tree Bower” on the other. All I know is that I’m no longer looking for a single network diagram that will underlie both poems.

To disentangle that one I suspect I’ll have to have recourse to Peter Gärdenfors’ ideas in Conceptual Spaces: The Geometry of Thought (MIT 2000). Those conceptual spaces are geometric object between the complex dynamics of the neural net and the symbolic processing of language. The language processes of “Kubla Khan” are “coupled” to the neural net through a different kind of conceptual space (with different dimensions) than that connecting the language processes of “This Lime-Tree Bower” to the neural net.

And this point, however, we’re in danger of pre-mature reification. It’s best to end here.

No comments:

Post a Comment