Joshua Tan critiques my recent Scientific American blog post, “The hippies were right: it’s all about vibrations, man!” and the general resonance theory (GRT) of consciousness briefly described in that post.
GRT is a theory developed by Jonathan Schooler and myself over the last decade that views consciousness as a product of various types of resonance at different scales. All things resonate, and a shared resonance allows simpler processes to combine into more complex processes that yield richer types of consciousness.
In his critique Tan relies mostly on my 2011 article in the Journal of Consciousness Sudies, “Kicking the Psychophysical Laws Into Gear: A New Approach to the Combination Problem.”
I don’t agree with Tan’s critiques (unsurprisingly) and here’s why. I’ll respond to just two of his points since they are his main points.
- Penrose and Hameroff’s Orch OR theory has made many predictions that have been empirically supported
Tan suggests that our resonance theory of consciousness is just one in a line of empirically-challenged or, more generally, empirically-ungrounded theories of consciousness. He mentions Penrose and Hameroff’s Orchestrated Objective Reduction (Orch OR) theory of consciousness and the critiques raised against that theory by physicist Max Tegmark.
Tan states that Tegmark objected to Orch OR as being empirically challenged because the quantum effects that Orch OR suggests take place inside neurons at the level of the microtubule (and are thus the seat of massively more computation than would otherwise be available to neurons) couldn’t be relevant to consciousness because neurons and brains are too hot and wet for quantum superposed states to persist long enough. Tegmark wrote:
We find that the decoherence time scales (∼10^−13 to 10^−20 s) are typically much shorter than the relevant dynamical time scales (∼10^−3 to 10^−1 s), both for regular neuron firing and for kinklike polarization excitations in microtubules. This conclusion disagrees with suggestions by Penrose and others that the brain acts as a quantum computer, and that quantum coherence is related to consciousness in a fundamental way.
Tegmark concluded that Orch OR was already falsified by this difficulty. Tegmark essentially repeated his criticisms as part of his 2013 book Our Mathematical Universe.
Tan links to Tegmark’s 2000 paper where he makes this claim and then moves on in his critique.
There is, unsurprisingly, a lot more to this story than Tan lets on, either because he isn’t aware of it (his field of study is physics, not the philosophy of mind or neuroscience, which are the main fields our theory straddles) or because he chooses not to dwell further on a case that he considers, apparently, to be shut.
I’ll dwell a bit further on this debate because the general resonance theory of consciousness (GRT) that Jonathan Schooler and I have developed over the last decade, which Tan critiques, has some significant commonalities with Orch OR. In fact, our approach is, in part, a generalization of Orch OR to a broader framework that is based on resonance as a unifying principle for the combination of simpler conscious entities into more complex conscious entities. Our approach is not in conflict with Orch OR’s more specific focus on microtubules and tubulin dimers as possibly the primary engine of human consciousness (we would argue, however, that it is the fields generated by such activities that are the seat of consciousness, rather than the microtubules themselves). GRT and Orch OR are also both inspired by Whitehead’s process philosophy and his panpsychist starting point. We also raise in our work the possibility that quantum effects may be relevant to human and other mammalian consciousness, but ultimately remain agnostic on this issue.
There has been a long debate about whether quantum phenomena are relevant in biological systems (Tegmark 2000, Hameroff and Penrose 2014, Craddock, et al. 2017, e.g.). Some physicists and neuroscientists adopt the view that quantum phenomena cannot be present in the warm wet systems of mammalian brains, as Tegmark has. However, the field of “quantum biology” is thriving and new examples of established or potential quantum effects in warm wet biological systems are coming to light with increasing regularity (see Lambert, et al. 2013, for a recent review). It is now well-established, despite the commonly-held view that quantum effects can’t be present in mammalian brains, that quantum phenomena are in fact common in biological systems. Lambert et al. states: “Recent evidence suggests that a variety of organisms may harness some of the unique features of quantum mechanics to gain a biological advantage.”
While I am open to the possibility of quantum phenomena in mammalian brains, and also these effects possibly being part of the complex causal phenomena of consciousness, GRT does not require the presence of quantum phenomena for its validity. It is, rather, about resonance chains of whatever causal connections are available.
My main concern in raising the “quantum question” in my work is two-fold: to not rule out quantum phenomena prematurely (to follow the facts where they may lead), but also to not rush to the conclusion that quantum phenomena are in fact necessarily part of the dynamics of consciousness. There is plenty of room between accepting, for example, that some brain dynamics seem to operate independently of traditional electrochemical neuronal pathways, and the notion that quantum phenomena must be invoked to explain such apparent anomalies. There is a vast middle ground that should be (and is) explored before such an alternative explanation is considered to be necessary.
We do see brain dynamics, such as the rapid achievement of gamma synchrony in areas of mammalian brain that seem to be too far separated for electrochemical effects to be the only cause. (Freeman and Vitiello 2006).
What is happening that seems to allow such rapid communication across large parts of mammalian brains? Freeman and others (Pockett 2012; McFadden 2007) have suggested that the electric field itself, which is created by the brain and supervenes on the brain, becomes a mediator of information. Freeman suggests that cortical neurons that resonate together undergo a phase transition into a unitary quantum state. Freeman and Vitiello 2006 states:
Our field theoretic model leads to our view of the phase transition as a condensation that is comparable to the formation of fog and rain drops from water vapor, and that might serve to model both the gamma and beta phase transitions.… The adoption of such a field theoretic approach enables us to model the whole cerebral hemisphere and its hierarchy of components down to the atomic level as a fully integrated macroscopic quantum system, namely as a macroscopic system which is a quantum system not in the trivial sense that it is made, like all existing matter, by quantum components such as atoms and molecules, but, as already mentioned, in the sense that some of its macroscopic properties can best be described with recourse to quantum dynamics.
Weingarten, et al. 2016 shows plausible pathways for quantum networks to exist alongside classical networks in the brain.
I find this research pretty intriguing without considering itby any means definitive. The current evidence suggests that it is plausible that electrical resonance is mediated by quantum resonance and perhaps prompted/prodded by quantum resonance in such a way that the very rapid achievement of electrical resonance states that we observe in mammalian brains is made possible by the far more rapid quantum resonance states at the subcellular level (in microtubules and related proteins perhaps).
As stated, I am agnostic currently about the likelihood or necessity for quantum mechanisms mediating mammalian consciousness, or for the necessity of quantum mechanisms to be involved in a broader explanation of consciousness beyond the category of mammalian brains. At this juncture there is not strong evidence that an ongoing whole-brain quantum synchrony is the pathway for shared information between the constituent micro-conscious entities. That appears to primarily occur through electrical and electrochemical pathways. That said, there is a growing body of evidence that quantum effects are relevant at the subneuronal level, including in microtubules — the focus of Penrose and Hameroff’s work. We’ve suggested in GRT that mammalian consciousness is a nested hierarchy of different levels of conscious entities that feed up to a single dominant consciousness. As such, each level of resonance may rely on different and perhaps multiple types of resonance chains, including electrochemical, electrical, magnetic, some kind of quantum entanglement (maybe), and even mechanical effects.
Craddock et al. 2017 (including Hameroff as a coauthor) is a recent example of the many empirically-tractable predictions of Orch OR, focusing on terahertz-level oscillations in tubulin molecules (which are the building blocks of microtubules). Tubulin dynamics has been identified previously as a promising area for empirical work by Hameroff himself. Craddock et al. 2017 focuses on the strong correlation between terahertz-level oscillations in tubulin molecules and the effectiveness of anesthetics, confirming some of the predictions made by Penrose and Hameroff in their work.
Regardless of whether quantum mechanisms are required to understand mammalian consciousness, however, the theory and worldview that GRT proposes is consonant with the quantum mechanical worldview and its wave-and-field-based ontology, which is (uncontroversially) the underlying ontology of today’s physics.
In sum: Tan’s comments about Orch OR being empirically-falsified are simply not true and reflect an ignorance about the many developments in this field over the last two decades and about the field of quantum biology more generally.
2. Is GRT empirically-tractable?
The second claim in Tan’s critique that I’ll address concerns GRT rather than Penrose and Hameroff’s Orch OR, or any other theories of consciousness. Tan writes:
As a scientific proposal, at least Penrose and Hammeroff’s [sic] idea has the commendable attribute that it was falsifiable in principle. In contrast, Hunt’s proposal is simply unfalsifiable. It may be an attractive description for some, but it does not provide any means for us to evaluate whether it is true.
By my reckoning, Hunt’s proposal fails, minimally, [the falsifiability] test. In fact, in his journal article [my 2011 paper, “Kicking the Psychophysical Laws Into Gear”] Hunt even seems to acknowledge to this: “..there is no proof possible regarding my proposed framework. But ‘science probes; it does not prove’. We must, each of us, proceed instead on available evidence, inference, and aesthetics.”
It is indeed true that there is never “proof” in science in the sense that there may be in formal logic or mathematics, but, crucially, there is disproof. I submit that Hunt’s idea in its expansive formulation does not provide any means for us to discover that it is incorrect. It is, instead, an example of an ipse-dixitism, a claim that is presented as self-evidently true and unfalsifiable, and, because of this, it crosses over from a scientific claim into one that is something else.
First, I’ll discuss where Tan is correct, then where he is incorrect.
He is correct that I acknowledged that there is no proof possible with respect to panpsychism as a broad metaphysical view about the nature of consciousness in relation to matter. Indeed, as I write in the quote he includes (from Gregory Bateson): “Science probes, it does not prove.” Tan doesn’t seem to appreciate the import of this quote or of my many related statements regarding my proposed theoretical approach.
Here’s the key point, which Tan agrees with: scientific theories are never proved, they are only supported or falsified. In practice, broad and popular theories are rarely falsified, because there are so many ways that different experiments can be interpreted or the theory at issue modified to accommodate new data. Rather than being falsified, such theories are simply superseded by newer and better theories (if such is the case). (See Kuhn’s The Structure of Scientific Revolutions.)
But I agree with Tan that all scientific theories should in principle be falsifiable. Interestingly, some prominent physicists, Sean Carroll among them, have argued recently that falsifiability should no longer be the primary criterion for judging scientific theories. I disagree with this argument but I mention this controversy to illustrate that many good scientists are broadening their notions of the criteria by which we should judge a good scientific theory.
Tan is incorrect that I present panpsychism as “self-evidently true.” I make the case for panpsychism in my paper and refer the reader to numerous other sources that make the case in more detail, including Griffin’s book Unsnarling the World-Knot and Skrbina’s book Panpsychism in the West, and my own 2014 book Eco, Ego, Eros.
Nor do I present panpsychism as unfalsifiable, as Tan claims. Rather, I state a number of times in my 2011 paper that I will present a more detailed follow up paper that describes various predictions and tests for the framework.
Panpsychism, materialism, and idealism — probably the three most popular philosophical positions with respect to the mind-body problem — should best be viewed as “metaphysical research programs” (a term introduced by philosopher of science Karl Popper) insofar as they intersect with neuroscience, psychology and cognitive science. GRT seeks to unify these various fields with respect to the study of consciousness, so our panpsychist approach certainly does intersect with these scientific fields. Accordingly, I suggest that panpsychism should be viewed as a metaphysical research program insofar as this philosophical position seeks to escape the limits of “mere” philosophy and become the basis for an empirically-grounded theory. This is our intention in grounding GRT in a panpsychist context.
However, falsifying a metaphysical research program is a tough row to hoe indeed. This is the case because, as with natural selection theory in evolutionary biology, the theory is so broad as to make it very hard to falsify. Karl Popper engaged in debates for decades about the appropriate label for natural selection theory and he backed away from his initial “metaphysical research program” label after facing various critiques.
More specifically, and more to the point with respect to Tan’s piece, my 2011 paper provides a mathematical framework for quantifying and characterizing consciousness in any given context, which yields many falsifiable predictions — independent of the broader context of panpsychism. I don’t flesh out specific predictions in my 2011 paper, promising a follow up to that paper (that still hasn’t been completed, unfortunately) that would discuss such predictions and tests.
The skeleton of my promised follow up paper is easy to state in terms of the basic categories for testing the panpsychist approach to the nature of consciousness:
a. Tests on human subjects that can report their experience
b. Tests on human subjects that cannot report their experience
c. Tests on non-human subjects that can report their experience
d. Tests on non-human subjects that cannot report their experience
The central method of the proposed testing framework is to begin with the most obvious cases of conscious experience — people like you and I — and to gradually move down the chain of complexity, and by doing so to gather evidence that will support or refute the panpsychist framework proposed. So rather than proof, we should and can seek support or refutation/falsification.
I thank Tan for helping to push this long overdue paper higher on my to-do list.
Moreover, even though the 2011 paper doesn’t include that specific list of tests, it should be obvious that any mathematical framework may be employed for testing the theory — that’s rather the point of including such details. The central equation of my 2011 paper is (Eq. 3, p. 120):
PI x CI = Omega
PI is the Perceptual Index (or simply sensory bandwidth), referring to external connections; CI is the Connectivity Index, referring to internal connections or processing power; and Omega is a comparative measure of consciousness that is the product of PI and CI.
With such a simple three-term relationship, there are numerous ways to make predictions about one variable by starting with the other two when they are known quantities. This equation provides a powerful predictive tool that is equally susceptible to attempts at falsification of the framework offered. I expand upon these ideas in the paper.
Tan’s is a short piece, so he didn’t have space to include much nuance. However, it’s reasonable to expect that he would have read in full the 2011 paper that he quotes from and critiques. And if he had done so, he should have been able to recognize the distinctions I’ve described here.
I’ll end by asking Tan and other readers a question: how would one falsify the prevailing materialist/emergentist solution to the mind-body problem? Or do we simply take that approach as a priori and self-evidently true?