When we have an unsolved question in science, sometimes we find a clear answer, but other times we watch the question slowly dissolve before our eyes.

A big part of my motivation for studying the brain these last few years has been to find a “robust” theory of the mind and brain: a theory which gives us a clear, unified picture of how the brain works, and precisely how mind and brain come together to generate all the things we’re used to—things like our coherent perception of the world, our ability to communicate via language, our capacity to think, imagine, or conjure new ideas, our capacity to remember and understand events of the past, or to plan and coordinate complex projects.

A question I’ve been thinking about recently, though, is whether it’s even possible to find such a theory. Is the kind of unification I’m looking for effectively impossible, because I’m trying to understand several distinct things that would better be understood in isolation? Is this the kind of question that slowly dissolves as we get a more detailed understanding of the mechanics of the brain?

Alternatively, is such a unifying theory already out there, and it just hasn’t been widely accepted yet? As far as I’ve seen, we haven’t yet found such a theory, or at least we don’t agree on whether we’ve found it, see e.g. John et al 2022 which claims that “In many ways, neuroscience is relatively preparadigmatic (Kuhn, 1962), akin to the field of biology before the insights of Charles Darwin, or chemistry before atomic theory.”

What I’m subjectively looking for: an “aha” moment when I read a particular theory of memory, consciousness, or emotion, where it makes sense why our experience has the particular structure that it does, and where the functions of the brain become a necessary consequence of its structure. A theory of the higher-level constructs of the mind. There are a few theories that vaguely give me this feeling, like Anil Seth’s idea about perception being a controlled hallucination whose function is to reduce prediction error. Or the idea of a Hebbian assembly that I mentioned in my current understanding of neuroscience.

But as I was reading through the entirety of Bear’s intro neuroscience textbook, I never really got such a theory. Again, there are some hints, some rough outlines, e.g. long-term potentiation in the hippocampus as a mechanism for memory consolidation, but much of it felt quite vague, e.g. the chapters on language and emotion, which mostly talked about correlational claims like “X emotion is associated with Y brain region” (and as I wrote earlier, I find such claims unsatisfactory).

The Lichtman lectures on connectomics were a further demonstration to me of how complex the brain is and how crude our current picture of it is—e.g. we currently cannot map the connections of all the neurons in even rat brains, let alone human brains. (As far as I know the only nervous system we’ve fully mapped is that of the C. elegans worm, which only has a few hundred neurons.)1

With all this said, what are some concrete signs I’d be looking for to indicate that we’ve indeed arrived at a unified theory of the mind? There are a few things a robust theory of the mind would have to do in order to count as the kind of explanation I’m looking for:

  • It would inform a clear measure of consciousness, that can tell us “definitively” whether or not a person or a given organism is presently conscious.2
  • It would inform the construction of an artificial, human-like mind3—i.e. tell us exactly what about the brain we need to replicate to produce such a mind, and what about the brain is an unnecessary relic of our evolutionary past.
  • It would give us a theoretical picture of the landscape of possible intelligences, so that we could e.g. answer questions about what the limits of intelligence are.
  • It would allow us to build highly targeted interventions for disorders of the mind that have a primarily “mental” or “experiential” etiology (as opposed to a purely physical cause like encephalitis)—depression and anxiety for example.4
  • It would allow us to determine whether there is such a thing as a “crucial neuron” – an individual neuron which, if killed, would result in an observable and predictable change in behavior.5

Such a theory would also give us a clear research program: it would give us a context in which there are a bunch of subproblems we need to solve. It would tell us what we don’t know and what we need to find out.

If such a theory has not yet been found, and we want to try to find it, where do we go looking? Do we look very closely at the properties of neurons? Do we formulate a mathematical theory of consciousness and intelligence? Is such a theory somewhere to be found in graph theory, or signal processing? Do we need to look within neurons?

  1. This begs the question: do we need to map all the neurons of the brain to have the unified theory I’m looking for? Not necessarily, but I imagine it could help. ↩︎

  2. I put “definitively” in quotes because I’m not sure we could ever definitively ascertain the consciousness of another being. Unless we’re literally able to project their conscious experience from their mind to someone else’s. I also use the word “definitively” because I know some people have created speculative measures of consciousness like the PCI, but in the absence of any agreement about what consciousness actually is, these measures remain firmly in the realm of speculation. ↩︎

  3. With the rise of AI language models like GPT-4, we’re debatably seeing the development of entities that have a “proto” mind. Not to suggest that they’re conscious (I personally don’t think they are) but that they do conduct some of the functions that human brains do, namely processing written language as well as an average human. With that said, these “proto-minds” are very much not human-like—they contain a much broader array of “knowledge” than the human mind does, but they also require enormous energy resources to compute things that human brains do with a tiny fraction of the energy expenditure. They also run on fixed hardware, while the hardware of the brain is constantly changing as a result of its activity. And there are a number of other obvious differences along the same lines. ↩︎

  4. Adding the nuance that sometimes depression is purely biological, while other times it can be better described as psychological in its cause. ↩︎

  5. Kandel’s Principles of Neural Science alludes to this:

    In invertebrate animals, and in some lower-order vertebrates, a single cell (a so-called command cell) can initiate a complex behavioral sequence. But as far as we know no complex human behavior is initiated by a single neuron. Rather, each behavior is generated by the actions of many cells." (Chapter 2, p37)

    The question seems to be somewhat open, and I’m wondering if a robust theory could give us a principled answer to it. ↩︎