Is the Brain a Quantum Device?
In 1989, the eminent Oxford mathematician and cosmologist Roger Penrose published a bestselling tome called The Emperor’s New Mind that was packed with wonderful material on physics, mathematics, and computers. Penrose’s main thesis was that the human brain is not a computer and must operate in some way that cannot be replicated on any computer no matter how powerful. That is, the brain did not follow “algorithms” in solving every problem it dealt with. Fine, so far. But then he went off the deep end with the incredible proposal that the brain’s actual mechanism had something to do with quantum gravity.
Penrose was met with considerable skepticism, especially in the artificial intelligence community, which he was basically attempting to put out of business, and also among physicists who could not see what quantum gravity could possibly have to do with a large, hot structure such as the brain.
Penrose then teamed up with anesthesiologist Stuart Hameroff in proposing a model for how quantum mechanics operates in the brain. Here’s how they explain it:
According to the principles of OR [objective reduction, proposed by Penrose in his 1994 book Shadows of the Mind], superpositioned states each have their own space-time geometries. When the degree of coherent mass-energy difference leads to a sufficient separation of space-time geometry, the system must choose and decay (reduce, collapse) to a single universe state, thus preventing “multiple universes.” In this way, a transient superposition of slightly differing space-time geometries persists until an abrupt quantum classical reduction occurs and one or the other is chosen. Thus consciousness may involve self-perturbations of space-time geometry.
Hameroff was one of the subjects interviewed in the 2004 independent documentary film What the Bleep Do We Know? That film, along with the succeeding 2005 film and still-bestselling book The Secret, exploited the notion that quantum mechanics tells us we make our own reality (see Reality Check September 2007).
In his Scientific American column of January 2005, Michael Shermer gave Bleep a scathing review. Referring to the Penrose-Hameroff model, Shermer references my 1995 book The Unconscious Quantum that discusses their proposal in some detail as well as the general question of whether the brain is a quantum device. In particular, Shermer pointed to a criterion I applied for determining whether a system must be described by quantum mechanics: If the product of a typical mass (m), speed (v), and distance (d) for the particles of the system is on the order of Planck’s constant (h) or less, then you cannot use classical mechanics to describe it but must use quantum mechanics.
Applying the criterion to the brain, I used the typical mass of a neural transmitter molecule, its speed-based thermal motion, and the distance across the synapse to find mvd about two orders of magnitude too large for quantum effects to be necessarily present.
In a letter responding to Shermer’s column, Hameroff wrote:
To debunk our theory Shermer cites an assertion in a book by Victor Stenger that the product of mass, velocity and distance of a quantum system cannot exceed Planck’s constant. I’ve not seen this proposal in a peer reviewed journal, nor listed anywhere as a serious interpretation of quantum mechanics. But in any case Stenger’s assertion is disproven by Anton Zeilinger’s experimental demonstration of quantum wave behavior in fullerenes and biological porphyrin proteins. (Skepticism should cut both ways, Mr. Shermer.) Nonetheless I agree with Stenger that synaptic chemical transmission between neurons is completely classical. The quantum computations we propose are isolated in microtubules within neurons. Classical neurotransmission provides inputs to, and outputs from, microtubule quantum computations mediating consciousness in neuronal dendrites.
First of all, the criterion I proposed is based on textbook quantum mechanics, originating with Niels Bohr in 1913—hardly in need of peer review. Second, I present this as a criterion for the necessary use of quantum mechanics in which you cannot get away with using classical mechanics. I did note that macroscopic quantum systems such as lasers and superconductors exist. They rely on the phenomenon of quantum coherence that can act over large distances.
In any case, Hameroff admits he agrees with me on my conclusion that “synaptic chemical transmission between neurons is completely classical.” He says he and Penrose propose that the quantum effects occur in microtubules within neurons. Microtubules are hollow, cylindrical polymers that are part of the cytoskeleton of all cells. As I noted in my book, I am puzzled that the quantum effects described in this model happen only with brain cells and not, say, the cells of the big toe.
In a 1999 paper, physicist max Tegmark looked at the problem of quantum coherence in the brain and determined that the decoherence timescales would be ten or more orders of magnitude shorter than the timescales for an event in the brain. The brain is simply too large and too hot to be a quantum device, coherent or not.
It is safe to say that the Penrose-Hameroff model has not been supported by the evidence to the satisfaction of the great majority of neuroscientists. However, let us assume Penrose is right about the brain not being a strict algorithmic computer. A simple mechanism exists, well known to complexity theorists, that can enable the brain or an electronic circuit to act in a noncomputable way.
External sources in the environment such as cosmic rays or internal sources such as radioactive potassium (K40) in blood can be expected to induce fluctuations in brain currents. These processes are quantum in origin, which means that they are random—at least in most interpretations of quantum mechanics. Like the fluctuations that provide for mutations in the evolutionary process, these might serve to trigger what complexity theorists call a bifurcation, when a system moves from one quasi-stable state to another.
The brain could operate that way, being basically classical and deterministic but occasionally jolted by a random quantum event. What is interesting is that the decisions made in this fashion would be indistinguishable from creative acts or free will. Is that all there is to it?