≡ Menu

Miguel Nicolelis and Ronald Cicurel: The Singularity Isn’t Near and the Brain Can’t Be Simulated

Miguel Nicolelis and Ronal CicurelThe idea that man is a machine is an old one. And since the brain is a physical part of that machine it is often presumed that it can necessarily be simulated on a digital computer. But very little scientific scrutiny has been given on that presumption. The Relativistic Brain is a new book by Miguel Nicolelis and Ronald Cicurel where they address precisely that question: Can a digital machine simulate the human brain?

the Relativistic BrainIn their book, Nicolelis and Cicurel provide a variety of neurophysiological, evolutionary, mathematical and computational arguments to conclude that “the brain is relativistic and cannot be simulated by a Turing Machine.” And so I thought that we had all the best ingredients for a fantastic Singularity 1on1 podcast episode. I hope you enjoy it as much as I did for I only managed to ask about half of my questions and plan to do a follow up as soon as I can.

During our 90 min conversation with Miguel Nicolelis and Ronald Cicurel we cover a variety of interesting topics such as: their interesting backgrounds in neuroscience and mathematics, brain-machine-interface [BMI] and the Blue Brain Project; neuroplasticity and Miguel’s Walk Again Project; the first Brain-to-Brain communication; the Relativistic Brain Theory [RBT] and why it cannot be simulated on a Turing Machine; the distinction between mechanism and organism; computable vs non-computable [Gödelian] information; their 23 scientific predictions as a way to falsify RBT; computationalism, Plato’s Cave and the mathematical arguments against simulating the brain; Stephen Jay Gould’s “tape of evolution argument”; the collapse of the Human Brain Project;  The Penrose-Hameroff Quantum Theory of Consciousness

(You can listen to/download the audio file above or watch the video interview in full. If you want to help me produce more high-quality episodes like this one please make a donation!)

Who is Miguel Nicolelis?

Miguel Nicolelis 2Miguel Nicolelis, M.D., Ph. D, is the Distinguished Professor of Neurobiology, Biomedical Engineering, and Psychology and Neuroscience at Duke University; and the Co-Director of the Duke Center for Neuroengineering. He is also the founder of the Edmond and Lily Safra International Institute of Neuroscience of Natal, Brazil.

Who is Ronald Cicurel?

Ronald CicurelRonald Cicurel, Ph. D., a fellow of the Edmond and Lily Safra International Institute of Neuroscience of Natal, is a mathematician and philosopher interested in studying mathematical logic and the philosophy of science. He has spent the last decade investigating how mathematics, epistemology, and physics can contribute to the understanding of the brain.

Like this article?

Please help me produce more content:



Please subscribe for free weekly updates:

  • Greg

    At the beginning of the talk you described the speakers views as an “attack on the singularity”. I haven’t read the book yet but I’m wondering if they do indeed they believe a technological singularity is impossible or if instead they just believe that whole brain emulation/simulating the brain in a computer is not a possible pathway of the many proposed pathways to achieve a TS.

  • Nicolelis and Cicurel argue that both simulating the mind and the technological singularity are impossible. But I would say they make much better argument on the first than the latter. So during our next interview I am planning to argue that there are at least 4 different ways to reach a singularity, at best, they have negated only one of those. But not all.

  • Greg

    Great idea. I personally have never thought simulating the brain was the most likely path to strong AI, although I’m far from convinced that it isn’t feasable, perhaps thier book can convince me where thier talk couldn’t. If they plan to debunk every proposed pathway to a singularity it’s going to be a longest Singularity 1 on 1 ever. I have a feeling they’re going to come up short. Either way I can’t wait to read their book.

  • I listened to the entire interview. It was informational and instructional about secular humanism except for the host’s digression about Alan Watts who, while an Episcopalian priest by training, was a proponent of the situation ethics of Zen, and posthumously is not known as a theologist.

  • I have no idea what you are referring to JCJ – I am a total fan of Alan Watts and he is one of my most favorite teachers of all time…

  • You can expect them to weasel out of that one. Even in this podcast Nicolelis already says that it may well be possible to arrive at a machine intelligence. He only asserts that it will be a fundamentally different type of intelligence than what humans have. So really, no, they are not arguing that the singularity is impossible at all (although they may think it is a long way off for various other reasons).

  • Alê GM

    I think you didn’t get the point.
    They are not saying that it is impossible to understand, through science, how the brain works. They believe that it is possible, and they are trying to do so. But they think that (at least some of) the processes that happen in the brain are impossible to translate into digital computation. It is known that some physical processes are incomputable. What they failed to demonstrate (at least in the interview – though I haven’t read their book) is that the brain uses any incomputable physical processes in it’s workings. If it is demonstrated, then it will be proven that you can’t simulate the brain in a computer (which does not mean that the brain will forever be “inexplicable”).

  • Pingback: Singularity 1on1: Ramez Naam on Apex, Science, Politics and Ethics()

  • Unfortunately I don’t have the time or the resources to produce transcripts at this time so there isn’t one ;-(

  • Pingback: Unanimous AI CEO Dr. Louis Rosenberg on Human Swarming()

  • Pingback: Can Consciousness Be Uploaded?()

  • Robert Strohfeldt

    If you have studied mathematics, you would know your comment that “scientists can use methodologies to test, understand and predict the complexities/behaviors (predict complexities?) of atoms, molecule and cells” is not true. Well the prediction of atoms and molecules. Electron movement, boiling water and synapses are 3 areas maths faces great challenges and least success in modelling.

  • Gary Feierbach

    Total nonsense! This is like free energy, based on a lack of understanding of quantum science and relativistic physics. Tis another foray into pseudo science not unlike Deepak Chopra’s purposeful misunderstanding of quantum mechanics to develop his points. I won’t swear that the singularity will arrive on time, but it will arrive!

  • Gary Feierbach

    By the way, Penrose and Hameroff have very little support for their theories as well and those theories are largely discredited by most in the field. I don’t believe any of those involved in these theories are scam artists, but simply people that have fallen in love with their own theories.

  • Travis

    I’m surprised someone hasn’t invented the voice to text feed. A website where you upload your mp3 and it gives back the text version. 🙂

  • Yes, they have. But I am yet to find one that actually works well ;-(

  • Travis

    Ok, so someone needs to invent wiki-transcript. 🙂

  • phoenixxl

    Digital computing makes it hard indeed to make a straight “copy” of a neural structure and emulate it in real time. This does not prevent strong AI from being developed on current digital computers. Identical problems can be resolved in different ways.
    We are also taking our first steps into an ( albeit limited for now ) quantum computing world. Qubits beat neurons when it comes to parallelism. A meatbrain “bloom” based on multiple inputs and current brain state moving at 100m/s is pretty slow compared to a few cycles in a quantum computer which would factor x bits worth of results in 1 cycle. Compared to a simple digital Von Neuman based computer which would of course need to loop and time share between all splits of the tree-like “bloom” and “decide” which ends get enforced it would be very fast . Again, design wise making new strong AI out of a quantum computer or current design computer ( not the thing on your desk ) would be different than what an actual brain does now.

    Abut the “never” … I remember the phone company telling me with firm assurances in the early nineties it would never be possible for anything to go faster than 1200 baud over their copper phone lines..(i was using a courier HST back then) we’re at 300000000 baud today.

  • phoenixxl

    Today we have bacteria that can make microscopic wires. We will have computers on a very small scale soon. We may not need to emulate anything. Imagine axons being replaced by coiled up wires and braincells being grafted with nano sized computers mirroring the signals the neurons are making.

    Imagine your brain as planet full of buildings, full of rooms , full of orchestras. If one violin has it’s strings broken, and gets replaced. it’s still the same violin. If a building gets old and gets renovated brick by brick, it’s still the same building. In the end, while the biological brain breaks down the neural mesh takes over until nothing biological is left.
    For me this is the most plausible way of keeping the current state going on in digital form.

    In a lot of scifi this is a recurring concept.

Over 3,000 super smart people have subscribed to my newsletter: