Hi Torsten

Since I am not conversant with hyperbolic geometry and quantum physics, I cannot pretend to have understood all the fine details present in your modeling of the neural network, but I certainly found it to be intriguing, since my own essay was attempting a different type of modeling, that of the social system. What I found most interesting about your model was that the top-down flow of neural activity is more predictive of conscious awareness than the bottom-up flow of activity. You go on to state in the conclusion, "Top down causation refers to the effects on components of organized systems that cannot be fully analyzed in terms of component-level behavior but instead requires reference to the higher-level system itself".

Speaking purely from what I would expect to be the case for intelligent systems, (without much knowledge of the brain as an organ), I would expect such top down causation to be present if a system is to be considered as intelligent. And although I have not thought about it in precisely those terms in my essay, I think you might be able to recognize this top down behavior in it as well. I think you have written a great essay and I rate it accordingly.

Warm Regards, Willy

    Dear Willy,

    thanks for your words and in particular for reading my essay. I printed your essay today and will read it in the next days.

    I agree with you that top-down causation is one important sign for intelligence but also complex behaviour. Interestingly, at first I also thought about to write the essay about social systems. So, I'm eager to read your ideas.

    Thanks also fo rthe voting (unfortunately destroyed by another down-voting).

    Best wishes (and more soon)

    Torsten

    Hello Torsten..

    I wanted to thank you for the comments left on my essay page, and to let you know I have begun to read and enjoy your essay. There is a lot to digest! I like the way you introduce the need for top-down interactions, to provide a faithful model of perception. I think both the universe and our awareness of it are participatory, even though much is automatic. I think learning about the non-automatic part is the key to understanding perception. Playing with ideas gives more insight than memorizing.

    I appreciate the link to your paper with a no-go result for an n-qubit Spin-2 Hamiltonian simulation. This could be relevant to the matter of quantum computing via a BEC/BH event horizon analogy, I imagine. I apologize for the delay responding. I submerged myself to finish my first ever essay for the Gravity Research Foundation contest, but that is now sent in. Back to you soon, with more on your essay.

    All the Best,

    Jonathan

      Hello Jonathan,

      what a nice coincidence: I also submit my essay for GRG today.

      I'm glad you like the no-go paper (it was never published in a journal. It was rejected because the result is of no interest for physics).

      I'm eager to see your feedback

      All the best

      Torsten

      Torsten,

      Remember you from the last contest.

      I wonder if topology could shed light to the 1.8 billion light-years across supervoid, seemingly an anomaly scientists have discovered, something I reference in my essay.

      Neural network models are a quite intriguing addition to neural studies. I would think that a qualitative approach for the brain is especially effective considering the multitudinous neural pathways of the brain.

      Mindless rules will generate mind! Clever.

      Hope you get a chance to comment on mine.

      Jim Hoover

        Jim,

        thanks for reading my essay and for the comments.

        After the completion of my GRG essay I will have a look into your essay soon (it is on my reading list).

        Therefore More later

        Torsten

        This is excellent work Torsten!

        I have a quibble, that in human brain learning studies it was found that the naive view holding that similarities of structure between table result in similar memory encoding is untrue, but it was found that memory images were grouped by function instead. So there are several places a coffee cup appears represented in the brain, depending on whether it is empty or full, clean or dirty, and so on. It's usage determines how it is stored. So we might find table and couch represented in the same brain areas.

        But this is in keeping with your observation of the importance of top-down influences, because to a living being in the real world objects are meant to be used or to have uses - which affects how we conceive of it. The old Chinese proverb is that the value of an urn is the space it contains. But this is an object fashioned by humans because it can fulfill a particular purpose. It is interesting though, that this purpose orientation is a driver of neurological specialization.

        It was worth taking the time needed to digest the Maths, so I could get a better perspective of your intended meaning. But more digestion is required; you have given me a lot of food for thought, between your essay and comments, so I'll likely have some questions or a few comments of my own. An excellent read overall, and I hope you do well in the contest.

        All the Best,

        Jonathan

          I wanted to cycle back and say more..

          First; I love the notion that consciousness is a fractal, and the fact it is the end result of your test process is very cool. It appears that the definition arises solely by imposing topological conditions and mapping the resulting parameter space; is this correct?

          I've become fond of the idea that fractals are a way for nature to squeeze in more information than would otherwise fit in the constraints of certain geometrical or topological parameters. The folding of space at the boundaries provides an extensive working surface, and the self-similarity assures a consistent rule will emerge for entities exploring that parameter space. So it is, I guess, natural that emergent consciousness would be characterized by fractality. I will continue to ponder what you have written.

          I hope that, at some point, we can expand our conversation beyond the contest topic. My current research carries me into areas where your expertise would be very helpful. The Mandelbrot Set suggests a geometrical route to unifying gravity with the rest of Physics. At (-0.75, 0i); the 5-d black hole --> 4-d spacetime scenario proposed by Pourhasan, Afshordi, and Mann is realized (if embedding M in the octonions is assumed), because Cartan's rolling-ball analogy of G2 symmetries is precisely modeled. And so is the set-up for DGP gravity!

          Further down; the Misiurewicz point near (-1.543689, 0i) is an exact model for the quantum critical point of BEC formation, where an analogy can be made with Schwarzschild event horizons. This connection was first suggested by Sakharov, but has been extensively treated in recent papers by Dvali and colleagues. As you know, this has deep connections with topology, exploring degrees of freedom, and so on. But there is much work to be done, to carry this to fruition.

          More later,

          Jonathan

          I was pleased to see I'd boosted your score to par with mine..

          But now I see someone has knocked it back down again. This essay deserves to be in the finals, so that it will receive a review by someone intelligent enough to rate its quality fairly. Again I wish you luck Torsten.

          Regards,

          Jonathan

            Hi Torsten

            Since you haven't yet reviewed, I thought I'll sneak in a comment here. You are probably familiar with Conant's Good Regulator Theorem, since it explicitly deals with brains as an example. I would greatly value your feedback on whether it is applicable to my work.

            http://pespmc1.vub.ac.be/books/Conant_Ashby.pdf

            At present my understanding is that it is applicable. My interpretation of Conant, in the domain of brains, would be that it is the regulational part of the brain that is a model of the brain's basic perturbations. This is a slightly different emphasis from Conant, since in his work he suggested that the entire brain 'must' model its environment.

            On modeling the external world, the brain may not be a perfect at it, but if enough focus is given to a certain area of the external world, the brain could eventually be trained to model that part of the external world very well. Please let me know what you think.

            Warm Regards, Willy

            Dear Jonthan,

            at first I was travelling the last three days and here is my delayed answer..

            Thanks for the voting and for your words. I hope to get a chance for the next round but to have a chance for good discussions is also very good.

            Your interpretation of fractals (as squeezing information) is very good. I never thought in this direction. Fractals is only one method for a more general view, a wild embedding (like Alexanders horned sphere). I wrote about it in the previous FqXi essay contests. I found it by chance during my research on exotic smoothness. The interesting point is the equivalence between wild embeddings and quantum states (and therefore fractals as wild embeddings of the circle must be also correspond to some quantum state).

            Therefore what you wrote about Mandelbrot sets and unification contains maybe a very deep truth. We should start an email exchange about it (torsten.asselmeyer-maluga@dlr.de) if you like....

            As a direct reaction: it was interesting what you wrote about functional memory of our brain. Honestly I don't know it. I was guided by mathematics when I developed this model.

            Thanks again for voting

            More later

            Torsten

            Dear Willy,

            at first I was travelling the last three days and here is my delayed answer...

            Thanks for reading my essay and the comments.

            I agree with you completely as top-down causation is a very important sign for intelligence

            but not the only one (there are also complex systems showing top-down causation without

            being alife).

            With great interest I also read your essay. Before I found this model about the neural

            networks I also thought about social systems (right in your spirit). As you correctly

            stated hierarchical structures played a strong role in this game. (I like the Nelson Mandela cite) I miss a little bit the evolutionary elment in your discussion (my work enforced me to think in this direction).

            I'm quit esure that this regulator theorem is applicable for you theory as well.

            Thanks for writing such a good essay

            Best wishes

            Torsten

            I shall begin an e-mail discussion..

            The term 'mirror neurons' comes to mind, regarding the specificity of activated brain areas being associated with various actions an object might perform, rather than with an object by its qualities of construction or appearance.

            More later,

            Jonathan

            Dear Torsten Asselmeyer-Maluga,

            Thank you for your detailed and colorful essay on neural networks and the fractal nature of the ability to generate intention. While I have had some limited experience with neural networks over the past 20 years, your essay has inspired me with some fresh perspectives. I find the fact that "feedback loops produce the topology of the network" to be a key step in future computational ability too. We are also in agreement about the nature of top-down and bottom up feedback loops being essential in such realizations. Thank you for writing so succinctly and with great examples. I have enjoyed your essay and have rated it too.

            Regards,

            Robert

              I hope you enjoy the paper I sent you Torsten..

              I have been thinking also about a fractal relation involving developed land and forests or other natural spaces. It appears that the quality of life is higher when humans and nature are in close proximity, where conventional planning often eliminates buffer zones in neighborhoods, or isolates them, but interpenetration allows ease of boundary crossing. Anyhow; I think fractal human/nature boundaries are optimal and that it would be good to write about this. It is broadly in line with your essay topic, but this is a question I've pondered for some time.

              All the Best,

              Jonathan

                Hi Torsten,

                Great paper, and I totally agree with what I think is the assumption behind it: that the only way the problem of consciousness will be solved will be to apply ideas from multiple disciplines- including advanced mathematics and physics.

                Best of luck in the contest and om your research,

                Rick Searle

                  Hi Rick,

                  thanks for your words. I like multiple disciplines. One can take the best from each. I already comment your essay, also great work.

                  All the best and good luck in the contest

                  Torsten

                  Hi Jonathan,

                  I have to further think about the paper you sent me. More directly via email.

                  Great idea to write something about ecology and human-nature interactions. There are also so many fractal structures in nature itself. Right I have also to think about it.

                  Best Torsten