My own private I dunno

February 15, 2012

I have been reading The Connectome: How The Brain’s Wiring Makes Us Who We Are, by Sebastian Seung, a Professor of Computational Neuroscience and Physics at MIT, and formerly a theoretical physicist.

One of his talks on the web is called “I am my connectome.” My own private connectome usually has a sort of uncontrollable synaptic response to statements like that, which seems to deny itself, but that would be quibbling with PR, and so, having functioned my whole life while knowing nothing at all about my insides, I have enjoyed the first 60 pages.

But one paragraph on pages 63-64 did throw me off:

If I could observe the activity of all your neurons, I would be able to decode what your are perceiving or thinking. This kind of mind reading would require knowing the “neural code,” which you can picture as a huge dictionary. Each entry of the dictionary lists a distinct perception and its corresponding pattern of neural activity. In principle, we could compile this dictionary by recording the activity patterns generate by a huge number of stimuli.

Now, just prior to reading this paragraph I had read Peter Woit’s latest blog post about the absence of signs of supersymmetry at the LHC in Switzerland.

And then my connectome wondered: is there a pattern in my connectome’s neurons that corresponds to my connectome’s discovery of the correct Theory of Everything?

Or can that pattern only be there after other people have found the right concepts?

Put another way, was there a neuronal pattern residing in Aristotle’s neuronal dictionary that corresponded to Newton’s Laws?

If so, then somewhere in my connectome is the answer to everything that can be known, even if the steps towards it and the concepts necessary have not been articulated yet.

Does a statement like that really mean anything at all?




We welcome comments that advance the story through relevant opinion, anecdotes, links and data. If you see a comment that you believe is irrelevant or inappropriate, you can flag it to our editors by using the report abuse links. Views expressed in the comments do not represent those of Reuters. For more information on our comment policy, see

I think Seung is wrong. He seems to just be saying that the brain is equivalent to a Turing machine or a cellular automaton. Fine. Then the connectome sounds like a description of what’s currently on the tape or the states of the cells. But without knowing how the neuronal activity evolves (ie the state machine inside the TM or the rule behind the CA), you don’t know what I will be thinking next. Right? Maybe I’m missing something. Can you tell looking at the memory space of a computer what the underlying algorithm is?

His more general point, that every human brain is at least in principle capable of computing anything that is computable, seems reasonable. Nothing is impossible! :)

Posted by pmaymin | Report as abusive

I don’t think there isn’t a correlation between physical and mental states. But I do think that to understand the mapping (even ignoring quantum mechanics) you need to understand the state of the entire world around you AND how it got there.

Posted by EmanuelDerman | Report as abusive

Nah, I think this means that the premise of a dictionary-like neural code is either nonsense of oversimplified.

Or at least that the dictionary changes its entries all the time, so observationally identical patterns at two different times may correspond to different perceptions and/or thoughts. And the neurochemical consequences of a set of perceptions at one time, along with everything else, change the dictionary constantly. The verb “is” isn’t good to apply to the connectome.

So the Theory of Everything is likely not latent in your brain (I suppose it could be)- it needs to be developed, or the pathways to make a latent Theory come out need to be developed, and I’m not sure these processes can really be distinguished.

Posted by GaryinDC | Report as abusive

Yes, the pattern corresponding to Newton’s laws was in Aristotle’s mind, and the pattern corresponding to the correct TOE (if it exists) resides in your mind. However, similar to Borges’ Library of Babel, the patterns to countless incorrect TOEs also reside in your mind, and there’s no way to tell which one is correct just by looking at the neuronal patterns.

Posted by DaveKavanagh | Report as abusive

A very interesting, short but comprehensive post. As you may know, the central question here is an ancient one. Can Socrates’s/Plato’s theory of recollection (all knowledge that can be known is already within us and we can not truly know/learn anything than the the innate knowledge that can be recollected) work for a brain in a vat? Can the brain be simulated with the representation of the universe and than state “I am my connectome” (what a name)? We can even go further. If we by simulation represent in the brain in the vat that it is a man in a Chinese room (Searle, Putnam, computational theory of mind) and give him a note on which a TOE equation is written, which is only correct for the room but not outside the room. What is than its use for the man to know the TOE is only true for the room if he will never know that it is false outside? I think it means nothing. There was in this case no external truth!

In this sense I agree with the critique of Aristotle about the necessity of interaction with reality, not to be confused with the platonic realism of universals, and it is only in this aspect that Plato’s idealism is shortcoming.

In this regard it is interesting to know about the exact extend of innate knowledge for the necessarily, essential, interaction with reality. For example, is the Cicadas prime number based reproductive cycles formed existentially, by experience? If yes, why can’t we unravel prime numbers? I think it is because they are not innate, in a sense that it is beyond the necessity for our conscious functioning. Even if we were able to understand the role of prime numbers in the universe, any application of it would be universally fragile and not robust, like how Cicada’s use is.

You have elegantly demonstrated in your philosophy of science how false materialistic perceptions lead to illusions. These are the same illusions that Plato/Socrates warned us about! Yes, indeed, and I too firmly believe, and maybe one day proof, that we can function, even more elegant and consistent, without consciously knowing anything from our insides and more importantly without creating possibilities for our extinction.

Humbleness is the virtue for perceiving the anarchy that lies beyond necessity.

PS: My apologizes for my English.

Posted by MahdiAbdulrazak | Report as abusive

Maybe Seung’s model has something in common with the immune system. We have millions of immunoglobulins which are capable of ‘fitting’ an extensive variety of three dimensional shapes. These catalog “every possible shape” in the sense that:
i) they have a good chance of binding to every non-self shape (viral particle, protein fragment, transmembrane protein on a bacterium, etc.) that appears in the body, and
ii) the immune system has some ability to ‘learn’ new non-self shapes and increase its repertoire.

Our connectome is highly versatile based on the common genetic and cultural heritage (spoken languages, family structure, etc.) that most human brains are exposed to in early development. Then it continues to adapt as our experiences specialize.

But it is probably meaningless to speculate about “where in your connectome” the ToE is “currently” residing. If nothing is currently activating it, does it exist?

Posted by SethOnReuters | Report as abusive