Dynamic Models of the Mind
George Kampis
Japan Advanced Institute for Science and Technology
Tatsunokuchi, Ishikawa, Japan
I am approaching the problem of the mind from two different angles that I hope to show you to converge to a common point.
First, there is the problem of representation. Both neural networks (which are purely fictional and arbitrary) and neurodynamic models (which are committed to electrophysiology) develop models of learning. Learning is commonsensically understood as a process that brings forth a learned state, one that contains information about the enviroment's regularities. In other words, a learned state contains a representation. What are representations good for? Representations are important only insofar as they have a causal effect, or in other words, if they are active. However, representations in neural networks and neurodynamic models – or to take a more general virewpoint: representations in all kinds of dynamicsal systems – are anything but active. At best, they act as filters of perception, or input, but lack an own causal power.
Second, most models of the mind face the issue of ontology in some form or another. Machine translation is an example. In order to properly translate, it might be necessary to ‘understand’, and a necessary condition for understanding is probably the ability to possess a model, an ontology, a whole world within the world. How can an ontology be defined in the mind, understood as a dynamical system? As I will discuss, an ontology requires the free combination of elements into new classes, something which a dynamical system cannot permit, for reasons of having a fixed dynamics..
In the main part of the lecture I will discuss causality as an alternative tool for understanding active representations and ontologies. In particular, I will discuss why I believe that causal systems are different from dynamical systems. The bottom line is that as a result of these convergent considerartions I believe dynamics is a result, rather than a vehicle, of cognition.
Finally, in a speculative outlook, I will contemplate the possibility that the brain with all its synapses, electric potentials, and firing patters is at best a detector for the activity of a causal mind, as is consciousness, another unecessary ingredient for causal processing.