The True Creator of Everything: How the Human Brain Shaped the Universe as We Know It by Miguel Nicolelis
Miguel Nicolelis seems to be a pretty smart guy. He’s the co-director of the Neuroengineering Center at Duke University, and has clearly forgotten more about superior longitudinal fasciculi and frontoparietal neuronal circuits than most people will learn in their entire lifetimes. He wrote a (seemingly decently received) book in 2011, Beyond Boundaries: The New Neuroscience of Connecting Brains with Machines (Times Books). Most famously, Nicolelis’ work on brain-machine interfaces led to a remarkable scene at the opening ceremony of the 2014 World Cup, when a 29-year-old man paralyzed from the waist down kicked a soccer ball using a robotic exoskeleton.
Now, Nicolelis has written a new book, The True Creator of Everything: How the Human Brain Shaped the Universe as We Know It (Yale University Press, 2020). As titles go, this is, um, a bit bold, but I’m willing to chalk that up to the exigencies of the publishing industry—after all, it did prompt me to pluck this tome from among the four impressively solid racks of science books at my local Barnes & Noble. (I live in Raleigh, NC, and it’s always encouraging to see a bookstore where the science section is more than a fraction of the size of the Bible/Christianity section.)
For this hopefully somewhat ambitious review of this very ambitious book, I’m going to try something different to help drive discussion in the comments thread. I don’t claim to be a PhD-level expert in neuroscience, computational complexity, evolution, the origin of mathematics, or any of the other topics discussed in The True Creator of Everything. I am, however, fairly conversant and well-informed in these and other scientific topics, and have worked in science-adjacent fields for much of my career. I’m going to call out parts of this review where I think Nicolelis’ conclusions are open to debate, based on my understanding (or lack of understanding) of the issue, and throw out specific questions to you, the readers.
A Totally Organic Experience! (With a Tip of the Hat to Clairol)
How does the human brain process information and generate qualia? Nicolelis uses the concept of an “organic computer,” which requires some unpacking. According to him, the brain turns “Shannon information,” the streams of digital bits created by speech, touch, smells, and other sensory inputs, into “Godelian information,” e.g., thoughts, memories, images, etc. that are embedded into brain tissue and thus strictly analog in nature. I will admit this all sounds a bit jargony—I’m still not quite sure why he calls analog information “Godelian”—but the idea gains force with repeated explication:
Rather than being binary and digital, Godelian information is continuous or analog, given that its embedding in organic tissue is fueled by the process of energy dissipation to organisms. As such, Godelian information cannot be digitalized or discretized and treated as binary bits of information flowing in a noisy communication channel. The more complex an organism gets, the more Godelian information is laid out, embedded in the organic matter that forms it.
Basically, I think what Nicolelis is saying amounts to a variant of “a picture is worth a thousand words.” Converting an (analog) mental image or feeling into (digital) words or other exterior representations would require an intractable amount of data. I think this gets pretty close to what most people informally mean by “qualia”: experiences that can only be conveyed approximately by words, painting, music, or any external creation of the human mind. To really understand a quale, you have to directly experience it, and that is only possible via the medium of your first-person-perspective human brain.
Nicolelis also lays great stress on the distributed, non-local, and extremely plastic functioning of the brain, all of which point to an analog rather than digital architecture. One surprising thing I learned from this book is that, according to experiments, different groups of neurons are recruited by the brain to perform the exact same motor functions: every time you raise your right index finger, for example, this action corresponds to the firing of similar (but not identical) groups of neurons. This seems very different from the way a digital computer would be expected to behave. Also, Nicolelis points out, specialized areas of the brain can be “retrained” (by experiment or necessity) to perform other functions, such as the visual neurons of blind people becoming more sensitive to the sense of touch. And, finally, decades of brain scans have shown that widely separated areas of the brain are often involved in the simplest acts and behaviors.
Why does it matter whether the brain processes and stores information in a digital or analog fashion? Well, you can probably see where this is going: Nicolelis is a strong disbeliever in strong AI. For one thing, as he points out, there is no hardware/software distinction in the human brain; he likens the brain to a tree, which converts exterior environmental information into (among other things) the rings that are an inherent part of its structure.
[Question #1 for the SSC commentariat: Does the hardware/software distinction, referenced above, really apply to today’s computer architecture? I have a hard time imagining researchers building a superdupercomputer and then loading up a beta copy of Microsoft Turing Test ™. If this distinction between hardware and software isn’t as unique as Nicolelis thinks it is, then it would tell against his argument for the brain’s noncomputability, but perhaps not totally destroy it.]
In the most memorable line of this book, Nicolelis nails down the essential difference between a brain and a computer:
You can’t reverse-engineer something that was never engineered in the first place.
Here, Nicolelis is addressing the hope of strong AI researchers that mimicking the architecture of the human brain, using digital components, can bring a self-conscious AI into existence. He makes a strong case that this effort is doomed to failure. Whether you take the starting point as three billion years ago (the origin of life on earth), 400 million years ago (the origin of mammals), 70 million years ago (the origin of primates), three million years ago (the origin of human-like primates), or even 10,000 years ago (the origin of modern civilization), the human brain has been shaped by a near-infinite number of environmental and evolutionary influences over innumerable generations. What’s more, individual human brains are shaped by a lifetime’s worth of interactions with their environments, both experiential (parenting, social and political structures, etc.) and chemical (hormones, neurotransmitters, etc.).
There is simply no way, Nicolelis argues, to replicate these intractable quantities and arrangements of information in a reverse-engineered computer; something essential will always be missing. He drives the point home with another telling analogy: a pool table can be considered almost infinitely simpler than a human brain. Yet, calculating the knock-on effects of a cue ball hitting a second ball, which hits a third ball, etc., by the time you get to the eighth ball you would have to factor in the gravitational field generated by the human player to effectively predicts its trajectory. A real-life analog system, whether it’s a human brain or a pool table, cannot be effectively simulated by a digital computer.
[Question #2: Strangely, in his discussion of whether the brain is a digital or analog computer, Nicolelis does not reference the Penrose/Hameroff theory of quantum computing in brain tissue, though he does reference Penrose elsewhere. However, my somewhat naïve understanding is that a sufficiently powerful quantum computer can model an analog system. Can Nicolelis’ argument that the brain is an analog computer be reconciled with the theory that it’s also a quantum computer?]
Nicolelis does offer a loophole of sorts for supporters of strong AI. One claim he makes, which I am not in a position to judge, is that highly organized patches of white matter in the brain act as “biological solenoids,” emitting electromagnetic fields that synchronize brain activity and the firing of neurons. (He backs up this claim, at least partially, with experimental evidence that transcranial magnetic stimulation, or TMS, can help allay some of the effects of mental illness. I will leave it to our host, Scott, to weigh in on whether this is actually a useful and generally accepted technique.) If a new kind of digital computer architecture can incorporate a magnetic “globalizing” component, Nicolelis implies, it may come closer to duplicating the functionality of a human brain.
Ironically, towards the end of his book, Nicolelis speculates that our current immersion in digital culture (smartphones, virtual reality, the internet, etc.) might actually wind up degrading the analog functioning of our brains. He cites as an example old-school London taxi drivers, MRI studies of whom reveal unusually large hippocampi; what, he wonders, will happen to the hippocampi of drivers who learn to navigate London’s streets not by experience, but by using a GPS system? I myself have wondered about the long-term effects Google will have on human grey matter; after all, there’s no need to memorize (for example) how many Academy Awards a given movie won, in what year, and in which categories, when you can just look that information up instantly, online.
Nicolelis, who is clearly not afraid to coin a phrase, calls this the “digital chameleon hypothesis”: as modern social structures and workplaces encourage and reward responses to digital inputs, the human brain responds by “pretending” to be digital, and loses many of the analog qualities that made us human in the first place.
Anyway, all of this is just the first part of Nicolelis’ book. On to our next section, brainets!
Check out the Big Brainet on Brad!
There is a story, possibly apocryphal, about an abysmally amateurish production of the play The Diary of Anne Frank. Early in the action, an SS man enters stage left, whereupon a member of the audience shouts, “She’s in the attic!”
That’s how I felt reading Nicolelis’s lengthy chapter about what he terms, rather unfortunately, “brainets” (which somehow calls to mind both Skynet and Raisinets):
Basically, a brainet is a distributed organic computer composed of multiple individual brains that become synchronized—in the analog domain—by an external signal such as light, sound, language, chemicals, or radio or electromagnetic waves and, as a result, is capable of producing emergent collective social behaviors.
Does any of this sound familiar? Well, let’s simplify a bit—instead of “radio or electromagnetic waves,” let’s just say the radio broadcast of War of the Worlds or the 5,000-screen release of Avengers: Endgame. Instead of light, sound and language, let’s say that Triumph of the Will rally choreographed by Leni Riefenstahl. Basically, Nicolelis seems to have rediscovered the concept of the meme—which is kind of strange, since he references Richard Dawkins and memes, in passing, later on in the book.
Interpreting this concept as charitably as possible, Nicolelis aspires to add a new layer of meaning to the meme regarding the existence of memes (sorry). From his position as a neuroscientist, he claims that the shared architecture of human brains allows certain memes to take hold and synchronize individuals at the neuronal level, accounting for everything from the Enlightenment to the willingness of British and German soldiers to charge out of their trenches to certain death during the Battle of the Somme. I have no doubt that this true, but I also think that some sort of neuronal link was implicit in Dawkin’s original formulation (where was the meme supposed to gestate, in a box?), and the concept of brainets doesn’t seem to add much to Nicolelis’ overall argument.
This is disappointing, because earlier in the book, Nicolelis hints at a much more original definition of brainets. This stems from his experiments with actual brain-to-brain coupling in test animals, so that, for example, one rhesus monkey (with titanium pins in its brain) can help another rhesus monkey (also with titanium pins in its brain) navigate an improvised wheelchair toward a pre-established goal, so both can reap the reward of a tasty fruit drink. If this sounds creepy, that’s because it is. To paraphrase another critic of Nicolelis’ work, it’s like the opening scene of a Paul Verhoeven movie the studio decided not to release.
Why does Nicolelis lay so much stress on memes (I mean, brainets?) Because he’s building toward his main theme, that human brains, both individually and collectively, determine the reality of our shared existence. First, though, we have to deal with the small issue of life, the universe, and everything.
[Question #3: How do you feel, personally, about inserting titanium pins into the brains of live rhesus monkeys? I have to admit, I’m a little freaked out about it.
What is Mind? No Matter. What is Matter? Never Mind
Remember that analogy I quoted above, where you have to take into account a pool player’s personal gravitational field in order to calculate the knock-on effects of his opening break? That is how I feel about books like The True Creator of Everything. The author starts with one-reasonable sounding concept, then adds another reasonable-sounding concept on top of it, and by the end of the book his inexorable chain of reasonable-sounding arguments lands you somewhere that…doesn’t seem quite so reasonable. But the scenery is interesting, so at least let’s talk about it.
After an interesting discussion about how the introduction of clocks in the Middle Ages permanently altered human beings’ relationship with and perception of time, and how Newton and Einstein’s scientific discoveries permanently altered human beings’ relationship to and perception of both time and space, Nicolelis offers this:
The very primitive concepts of space and time are also mental abstractions created by the human mind in order to reduce the dimensionality of complex potential information obtained from the outside world. Furthermore, I propose that as basic mental abstractions, time and space emerge as a result of natural selection…as a way of enhancing our evolutionary fitness. By filling the human universe with a continuous scaffold made of time and space, our brains enhance our chances to survive the contingencies imposed by the environment in which we have been immersed since the origin of our species.
This is not the first time I’ve heard this idea, but it may be the first time I’ve heard it expressed by a reputable scientist. Nicolelis adduces, as evidence for our continuing “creation” of space and time, the ability of the human mind to fill in information that isn’t actually there: perceiving a solid triangle in an optical illusion, filling in missing gaps in conversation, basically making sense of a world where solid information is at a premium. Nicolelis goes so far as to speculate that hallucinogens work not by impairing brain tissues that ordinarily process normal reality, but by reducing the ability of these tissues to manufacture the illusion of space and time out of the chaos that lies outside human perception.
But wait, it gets even weirder. Over and above organic computers, brainets, and the manufacturing of space and time, the most thought-provoking idea in The True Creator of Everything (in my opinion) concerns the true nature of mathematics. Here, Nicolelis relies heavily on Where Mathematics Comes From: How the Embedded Mind Brings Mathematics into Being, by George Lakoff and Rafel Nunez. This book, which I recommend, makes a strong case that human mathematics is based on metaphorical aspects of everyday human existence—we talk about objects residing within a set, being located a certain distance along a number line, having certain geometrical properties based on our experience of three dimensions. There’s a strong case to be made that mathematics is not platonic in nature, but is simply another human creation, like music or architecture. As Nicolelis puts it, “the whole body of accumulated mathematical knowledge could be seen as another type of emergent property produced by a human brainet, dispersed in time and space, over the entire history of humankind.”
Nicolelis proposes—I’m not saying I agree with him, I’m just saying he proposes—that any intelligent species arising in any universe would necessarily have a different evolutionary history and brain structure than us human beings, and, to mirror his sentence in my own words, “the whole body of accumulated mathematical knowledge could be seen as another type of emergent property produced by a Phelgmorphian brainet, dispersed in time and space, over the entire history of Phlegmorphkind.”
The implications of this are stunning, provided (with a big P) you accept Nicolelis’ arguments. Why should we have any expectation that, with their different brain structure and shared evolutionary history, Phlegmorphians would perceive time and space the same way we do, or have the same kind of mathematics? What Nicolelis seems to be proposing is that not only would it be *very difficult* for humans to communicate with another intelligent life form in the universe (think of those octopus beings in the movie Arrival), but it might actually be *logically impossible* to do so.
[Question #4: Is there any ontological/epistemological/phenomenological etc. content to the notion that another intelligent species in the universe could evolve a type of mathematics or cosmology that is completely incommensurable with our own? This sounds to me like trying to run away from your own shadow—I literally can’t conceive that it’s possible, but does that mean it’s impossible?]
This leads me to a final big idea, which is not in Nicolelis’s book, but which is mine because I thought of it and it is my theory and no one else’s. What if the solution to the Fermi Paradox is that intelligent observers actually do create their own reality, as Nicolelis says, and this reality is incommensurable with the reality of any other intelligent observers? Then, by definition, our universe could only be inhabited by one intelligent species—us, and our big brains.