Science Centric has a report on a new, more effective approach to simulating neural networks:
In their doctoral theses, Arvind Kumar and Sven Schrader have simulated large neuronal networks that, for the first time, take this neuronal feature into account. Especially in the neocortex, neurones are intensely interconnected, i.e. they receive many input signals that can modify the integration of subsequent signals. Taking the special features of such highly interconnected networks into account yields simulations that are in excellent agreement with recordings from biological nerve cells in the intact brain. The new virtual network thus reflects reality better than previous models.
This is of course relevant to the mind-brain question we’ve looked at often here. The article’s headline is “How thoughts arise.” For a dualist like myself, a better headline would be, “How neural aspects of thoughts arise,” or “How neural correlates of thoughts arise.” For it seems quite obvious to me that viewing thoughts as strictly neural events will not work. I’ll re-state one problem with it here.
We know that thoughts cause things. Suppose someone says to me:
All men are mortal.
Socrates is a man.
Is Socrates mortal?
My thoughts will run through what I know about men, and I will agree that all men are mortal. I will assume for the sake of discussion that Socrates is a man (he was once, at any rate). I will consider whether the answer is logically entailed by the statements I’ve been given. So I will think, yes, Socrates is mortal. Then I will answer aloud, “Yes, Socrates is mortal.” (Or I could be a wise guy and say, “Who wants to know, and what do I get if I answer right?”)
I have thoughts, which cause other thoughts, which ultimately cause physical actions in the world. Note that the progression of causes is in virtue of the content of the thoughts. It’s what the thoughts are about. The scientists quoted in this article want to address that question:
‘But it does not suffice that the brain is just active,’ adds Rotter. ‘The activity pattern must somehow be connected to a meaning.’ When we remember, our brain has to make associations and has to produce meaningful behaviour. How meaningful patterns arise in the ocean of neuronal network activity will be subject of new investigations by Rotter and his colleagues at the Bernstein Centre.
But there indeed is the rub for those who would say thoughts are just neural activity. That activity must “somehow be connected to a meaning.” The activity must be about a meaning. It is the meaning of the Socrates question above that causes the subsequent thoughts and spoken words.
Nobody has yet determined how a physical state, condition, or action can be about something else. How can a rock be about a seashore? It can be on it, it can be near it, but it cannot be about it. How can ink on a page be about love? That’s a little tougher, in the case of a love sonnet, for instance; but the answer is it can only be about love if there is a translation to that in the mind of the reader. The ink itself has no aboutness to it. The rock on the seashore cannot mean anything in itself, unless there is a mind to make a link of meaning with it.
There is good reason to believe this aboutness problem for physical entities cannot be solved; that a physical thing could never, in principle, be about something else. Thus neurons, no matter how complexly interrelated, cannot by themselves by connected to a meaning.
There is even better reason to believe that even if there were a solution to this, it wouldn’t be discovered through science. It’s a philosophical question, not an observational question. Still, it can’t hurt for them to work on it, and I wish them great success in learning more as they proceed.