models – consciousnessandstuff

The Corollary Discharge

Many, many years ago I did a PhD on whether an area of the brain called the frontal eye fields was involved in something called the “corollary discharge”.

A corollary discharge, or “efference copy”, was first postulated by Hermann von Helmholtz in the 19th century to account for the fact that when we move our eyes we don’t experience the visual world moving, even though the image moving over our retina would be expected to generate just such a sensation. Helmholtz’s elegant explanation was that the motor command in the brain to move the eyes is copied as a corollary discharge and used to cancel out the apparent sensory signal coming from the eyes. You can prove that this is what happens by gently pressing on one side of your eyeball – the visual world does appear to move in this case (sensory input but no corollary discharge to cancel it out). I don’t recommend you try the next stage of Helmholtz’s proof which was to immobilise his eye then try and move it, which also caused a sensation of movement in the opposite direction (corollary discharge but no sensory input)!

My research provided zero evidence that the frontal eyefields were involved in this mechanism! However, this area of the brain was clearly involved in selective attention in some way and this is what I concluded: “The function of the frontal eyefields is one of organising co-ordinated shifts of attention … the mechanism underlying this function is that of a repeated testing of an internalised schematic map of the visual environment against afferent sensory information”. This claim was criticised at the time for going beyond the evidence (these were the dark days of behaviourism) but forty years on I still reckon I was on to something. Here’s how.

If instead of the “internalised schematic map of the visual environment” we generalise and call it a comprehensive “mental model”, and similarly generalise the “corollary discharge mechanism” to a fundamental process of “testing hypotheses about the outside world generated by the model against what happens when we interact with the outside world” then I think we have a valuable insight into how the brain might work. Now let’s generalise a bit further to an internal model of the self interacting with an internal model of the outside world and only really paying attention when the sensory evidence from the real outside world is different from what the model predicts. According to this interpretation we normally only need to process a very small fraction of our sensory input – the rest of what we perceive is internally generated. But as soon as something unexpected happens (not predicted by the corollary discharge) we immediately pay attention and concentrate on finding out what’s new. Generalise even further and I can start to envisage a mental model which includes not just a model of me but models of other people I’m interacting with, and once again, I think it’s helpful to think of myself continually testing out the “corollary discharge” of these interactions generated by the model inside my head with what actually happens in the outside world, helping me to plan accordingly and behave in a socially acceptable manner.

Whether any of this is anything to do with the frontal eyefields is besides the point, although, interestingly, the prefrontal cortex does seem to be very much involved with attention, perception, planning, personality, and all the other qualities we associate with higher consciousness.

Finally, what happens in dreams, where we only have the corollary discharge, so to speak, without any sensory feedback? According to the theory, this should result in the internal representation of the outside world acting strangely in some way, by analogy with what Helmholtz saw when he immobilised his eyeball. My personal experience is that my dream self’s interaction with its dreamed environment does indeed tend to be “difficult” in some way. The clearest example of this would be those dreams where I’m trying to move – usually running away from something dangerous – and I find that I’m too weak, or bogged down in quicksand or otherwise constrained. Familiar? So what I think is happening here is that the “corollary discharge” of my action is not cancelled out by what I’d expect in the real world – the visual world streaming past me as I ran through it – so the model then tries to rationalise this by fabricating a narrative about me (weakness) or the world (quicksand) or whatever. With a little imagination (ie going beyond the evidence big time!) it seems to me that a lot of the strange things which happen to me in my dreams make more sense in these terms – failing to complete some goal, getting lost in large cities, being interrupted just as I’m about to have sex (unfair!), people I know acting out of character, a general clumsiness of execution on the part of my dream self. But maybe I’m just weird!

Hermann von Helmholtz 1821 – 1894

Some research into similar theories

To recap, what I’m calling the “Model Theory” is the idea that the way our minds work is by running a model, or simulation, of everything in the world, which, crucially, contains within it a model of our selves. This, it seems to me, explains consciousness, life, the universe and everything!

I’ve now done a little research on the subject and, not surprisingly, it seems several academics have come up with theories of consciousness which are close to what I’ve been blogging about. In fact it seems that consciousness is quite a hot topic again and a respectable subject for investigation and debate (this certainly wasn’t the case when I was a student). But most commentators are focusing only on parts of the issue and no-one really seems to share my extravagant claims for the theory. Maybe this is just the dry manner in which academics express themselves, or maybe I’m just deluded, but anyway, here’s what I’ve learnt, mostly from a site called Scholarpedia, which you can find at http://www.scholarpedia.org/article/Category:Consciousness .

Thomas Metzinger probably comes closest to the mental model theory. His “Self Model of Subjectivity” (SMS) features a “Phenomenal Self Model” (PSM). He makes a couple of particularly good points. “Transparency” refers to us not being aware of the workings of the model. The “Phenomenal Model of the Intentionality Relation” (PMIR) refers to the PSM interacting with a model of an object rather than the real object. Or that’s what I think he’s saying – it’s not always easy to understand these guys :-). But even Metzinger seems curiously diffident about his theory. And oddly, the main conclusion he reaches is that “the self” is an illusion. Well, I think I see what he’s getting at but I’d conclude exactly the opposite. That “I” am real seems to me to be self-evident, haha. Anyway, his 2007 article on “Self Models” http://www.scholarpedia.org/article/Self_models is well worth a read and he’s published a book called The Ego Tunnel which I’ve just ordered from Amazon – watch this space.

Revonsuo talks about a “World Simulation Metaphor” and is particularly interested in dreaming, which he interprets as the safe simulation of threatening situations.

Trehub has a so-called “Retinoid Model” of the brain, which conceives of a self-model at the centre of egocentric space.

Taylor describes a “Corollary Discharge of Attention Movements” (CODAM) model which seems very similar to what I was talking about in my last blog.

All these people and more feature in the Scholarpedia site, which also includes a useful summary of “Models of Conciousness” by Anil Seth. What’s really encouraging is that the problem of consciousness is once again being energetically addressed by a wide range of psychologists, philosophers, physiologists, computer scientists and others, and for the first time they seem to be communicating with each other!

A short history of the universe – and the brain of God!

How time flies. I finished the last post with the words “To be continued” and now I find it’s three years later! But in the meantime I’ve been doing a lot of thinking and reading, especially about quantum physics, and in this respect I’ve just finished a really good book called “Beyond Weird”, by Philip Ball, which has prompted me to revisit this blog.

So, to continue… Having convinced ourselves that consciousness is fundamentally “just information” we were considering the possibility that the whole universe is just information too.

How would this work? Well, starting at the beginning, we might speculate that somehow, suddenly, the following simple mathematical equation comes into being:

0 = 1 – 1

Voila! Out of nothing, we’ve got a something and an anti-something. Call them matter and antimatter if you like. But the important thing is that we’ve got a couple of numbers to play with and out of them we derive the whole of mathematics, already known and yet to be discovered, through a series of increasingly elaborate equations, eg:

1 = 3 – 2; 3 = 12/4; -1 = -4 + 3; i = √ -1; e^(pi*i) + 1 = 0; and so on…

Pretty soon, we’ve got some really meaty mathematical structures such as Mandelbrot’s Set* and, in particular, Schrodinger’s Equation:

$i\hbar {\frac {\partial }{\partial t}}\Psi (\mathbf {r} ,t)=\left[{\frac {-\hbar ^{2}}{2m}}\nabla ^{2}+V(\mathbf {r} ,t)\right]\Psi (\mathbf {r} ,t)$

All this happens in the blink of an eye – more or less instantaneously – let’s call it the “Big Bang”!

Then a miracle happens: all this maths somehow “coalesces” or “crystalizes” into a vast soup of quantum objects which we can think of either as waveforms, defined by various forms of Schrodinger’s equation, or as the potentiality (an infinite network of probabilities) of various elementary particles – photons, electrons, quarks etc. In other words we have created the “quantum world”; (note that, instead of using the term “think of”, I could have said that we can model the the quantum world either in terms of waves or particles).

The rest is fairly straightforward. The particles combine to become protons, neutrons, atoms, molecules and all the elements of what we call matter in the physical world (and presumably other quantum objects become energy: heat, light, electromagnetic radiation etc) ; all this stuff clumps together under gravitational forces into galaxies, stars, planets; on one planet life emerges; the human brain evolves, and we have consciousness. Simples!

Let’s go back to where the miracle happens. There’s actually two rather tricky steps here (for the time being I’m treating the 0 = 1 – 1 bit as more or less an act of faith!):

Step 1: Information becomes the quantum world (strictly speaking I’ve only been talking about mathematics, but if this process works for maths then it seems reasonable to suppose it works for any type of information).

Step 2: The quantum world becomes the physical world, which we know and understand through the laws of classical physics and which exists “out there”.

Having read Philip Ball, I’m now fairly comfortable with Step 2. If I understand him correctly, recent advances in quantum science now mean we can think of the physical world as being a special case of the quantum world. Weird things like superposition and entanglement really are the way things are, and the only reason we don’t notice them at the human scale is because in practice any simple quantum system is surrounded by so much “environment” (heat, light, other particles etc) that its waveform “collapses” almost immediately, and through processes known as “decoherence” and “quantum Darwinism” we see what we see. In particular, it seems we no longer need to get hung up on Schrodinger’s wretched cat and the idea that our observing or being conscious of the physical world somehow brings it into existence. Similarly, we probably don’t need to worry about “spooky action at a distance”, or the quantum world constantly splitting into an infinite number of “multiverses” with different versions of me. So that’s a relief!

Part 1 of the miracle is admittedly more difficult to swallow. The trouble is that information is, or appears to be, extremely slippery, ethereal stuff. It is difficult enough to imagine it floating about, out there, in the absence of anything else, never mind eventually somehow condensing into physical objects we can see and touch. But Philip Ball, like Wheeler and Rovelli (see last post), clearly suspects that information is in some sense at the bottom of the “real” nature of the quantum world, in the same way that I’ve argued that information is fundamental to consciousness. And if, as seems the case, the mathematics of quantum mechanics is an accurate and complete description of the quantum world – in other words, a good model – then isn’t it conceivable that matter, like mind, is fundamentally just information?

Perhaps what we have here is a hierarchy of models, all based on information processing:

Consciousness is a model of the physical world
The physical world is a model of the quantum world
The quantum world is a model of mathematics – in other words pure information.

But I’m still uneasy with the concept of information just being out there, floating about without any obvious substrate, such as the paper it’s written on, or a magnetic disc, or a chip in a computer. As described at length in this blog, I believe consciousness to arise out of an information model running on a substrate comprising the neurons in the human brain. The physical world is in a sense an instance, or model, of the many forms which the Schrodinger waveform might take: its substrate is the quantum world. And the quantum world is mathematics – just information – the purest and most abstract form of model imaginable. But what does it run on, what’s its substrate? The brain of God?

* Mandelbrot’s Set

Mandelbrot set