Neural Darwinism: Theory, Evidence & Debate

[Tsunami]

I was wondering if the theory of Neural Darwinism

So, I read this popularising book, talking about Neural Darwinism, (Israel Rosenfield, The invention of memory) that was contesting the idea of brain modularity (the idea that the brain is divided into functional regions, interacting which each other but still localised in modules). This book is already a bit outdated; considering all the research done in the field of brain science, I was wondering whether new progress can allow for a more nuanced view of the opposition between full modularity (all brain functions are localised in independent neurons; something I'm convinced cannot be the case) and full emergence (all neurons are identical, and every neuron can 'learn' to interact with another neuron, thus being capable of building functions through cooperation and emergence),

I just made up the term 'emergence', so if anyone knows a better term, be sure to correct me.

What I am interested in specifically, is evidence that shows certain views of full modularity or full emergence to be impossible, - every added hypothesis you have about which view of the brain IS possible, and why so, is of course also very welcome.

(O, and if there's evidence to disprove Neural Darwinism, or arguments why it's a very interesting theory, I'd love to hear as well. )

 

[Moonbear]

I'm afraid I'm unfamiliar with the term "neural Darwinism." I could only guess. You describe briefly what it contests or purports to refute, but what does it say happens instead?

I don't have much time today to discuss this idea, but I think I can offer some answers to at least some of your questions. I would just like more information about what "neural Darwinism" is claiming happens just to be sure I'm understanding your questions in context before I attempt answering them.

 

[Tsunami]

"Neural Darwinism. The Theory of Neuronal Group Selection"(1967) by Gerald E. Edelman, is what the idea is based on. I myself only read the layman's defense of Edelman's ideas by Israel Rosenfield, "The invention of memory".

I'm just going to say how I interpreted this book (but I only read it once, and the concrete content is already slipping from my memory, so take it with a grain of salt):

Basically, because earlier brain research gave impressive demonstrations of the modularity of brain functions (for instance, the localisation of language functions in regions of Broca and Wernicke), in cognitive science there's a very strong fix on localisation of brain functions in fixed places of the brain. So always when you hear neurologists talking, they are talking about which parts of the brain are triggered when performing certain functions. If different parts of the brain are triggered in different people, then this is seen as irrelevant for the task that's being performed. Also, if small regions are triggered, but only with very slight intensity, then this is also seen as irrelevant. MRI scan data is always performed with a certain intensity threshold, yet I've never heard people talking about threshold margin, but that might just be because I only see the 'popular' version of the research.

However, in Rosenfield's book, he tries to present a much more dynamic, less localised idea of brain functions. He talks about certain psychological cases which can't be explained by full modular brain functions, but only by a much more context-based, associative functioning of the brain. Just like in evolution, random mutations are rooted out by natural selection, which is adjusted well enough to the environment, the performing of brain functions according to neural darwinism happens by a form of natural selection of neuron interaction patterns. So in this sense, you get a more intuitive image of the brain as being a set of identical neurons, which 'learn' to interact with each other in such a way, that they are able to perform higher brain functions. So the higher brain functions emerge out of simple neurons and simples synapses between them. That would be an idea of full emergence.

Of course, it's probably something of both. Certain structures have learned to differentiate, and adjust to different functions, and so the modularity of regions certainly exists. However, Neural Darwinism urges to back this modularity up with a whole collection of emergent behaviour, and I believe that functions like memory, consciousness, learning, etc. are all functions that can be best explained through this idea of emergence ; essentially all functions which are context-based, person-based, ... in other words, variable for every single brain. (But also biological adjustments to impairments, ...)

This also means that there are a number of things where very tiny parts of the brain might have an immense impact on certain functions. Which means that it's absolutely wrong of neurologists to assume that slight brain activity can immediately be pointed at as irrelevant, and which also means that context-based information is very important. I've heard people say that there's a certain region which is only active when you're lying, and not when you tell the truth... but I'm wondering what happens to the region when you have to answer a question you don't immediately know the answer to (the brain effort I do then seems very similar to me as lying).

So, I was wondering what happens in the field today, regarding this dilemma. (As said, I'm a layman in the field, but I do have a scientific interest in it.)