Roger Penrose does not think a computer will ever be conscious, and he does not think a brain running on ordinary neural computation would be conscious either. For close to forty years, the Oxford mathematical physicist has argued that consciousness cannot be the product of algorithmic processing at all, and that explaining it will require physics we have not yet discovered.
That is a stronger and stranger claim than it might first sound. Penrose is not simply saying the brain is complicated, or that neuroscience has further to go. He is saying that no arrangement of switches, however intricate, produces the kind of understanding a mathematician has when they see why a proof is true. Something else has to be happening, and in his view that something is rooted in quantum mechanics and, more specifically, in an unresolved question about how quantum states collapse into definite outcomes.
What Orch-OR actually proposes
Penrose developed the idea, called Orchestrated Objective Reduction, with the anaesthesiologist Stuart Hameroff, whose interest came from watching what general anaesthetics do to consciousness. Their joint theory, laid out most fully in a 2014 review in Physics of Life Reviews, proposes that structures called microtubules, tube-shaped scaffolding found inside almost every cell in the body including neurons, are able to sustain a form of quantum superposition for a brief window before that superposition collapses.1
The “objective reduction” half of the name refers to Penrose’s own long-standing proposal, separate from Hameroff’s biology, that quantum superpositions do not collapse only when observed or measured, as standard quantum mechanics is usually read, but collapse on their own once a threshold connected to spacetime geometry is reached. The “orchestrated” half is Hameroff’s contribution: the claim that this collapse, inside microtubules, happens in a structured rather than random way, and that the pattern of those collapses is what we experience as a moment of conscious awareness.
Put together, the theory says something specific and testable in principle: that consciousness arises not from information processing in the ordinary computational sense, but from a physical event, gravity-linked quantum collapse, occurring inside a particular cellular structure. It is this claim, that the seat of experience is a physical process tied to the deep structure of spacetime rather than a pattern of neural firing, that gives the theory its appeal to people frustrated with purely computational accounts of mind. It is also the claim that most of the field rejects.
Why most neuroscientists have not been persuaded
The standard objection is thermal. The brain is warm, wet and constantly jostled by surrounding molecules, an environment in which quantum superpositions are expected to collapse, or decohere, within an extremely short span of time, far too short to be useful for anything resembling a thought. A formal rebuttal published alongside the 2014 review, by the chemist Jeffrey Reimers and colleagues, argued that the specific mechanism Hameroff and Penrose proposed for shielding microtubules from this thermal noise is not scientifically supported.2
Penrose and Hameroff have pointed to a 2013 study from a group in Japan reporting vibrations in microtubules, which they read as evidence that the structures can support some form of coherent quantum behaviour even in a warm environment. Critics have countered that detecting vibration is not the same as demonstrating the specific, long-lived entangled superposition the theory requires. The dispute has not been resolved by either side conceding the point, which is the normal state of an active scientific disagreement rather than a sign that one side has been quietly dismissed.
Where the theory has actually been tested
Orch-OR is unusual among consciousness theories in making a claim precise enough to test outside the brain altogether. If gravity really does trigger the collapse of quantum superpositions at the scale Penrose describes, that collapse should show up in careful physics experiments that have nothing to do with biology.
One such test was run by a group including the physicist Catalina Curceanu, using an underground laboratory beneath the Gran Sasso mountains in Italy, shielded from the cosmic radiation that would otherwise swamp the measurement. Their result, published in 2022, found no evidence of the gravity-related collapse Penrose’s model predicts at the scale relevant to microtubules, and described the simplest version of the Orch-OR collapse mechanism as “highly implausible,” while noting that more complex variants of the collapse model were not ruled out.3 That is a narrower finding than a refutation of the theory. It closes off one specific version of the mechanism without closing off the broader idea that objective collapse of this kind occurs.
More recently, an independent 2025 preprint by the engineer James Tagg reported an experiment on a superconducting quantum computer that the author describes as consistent with objective reduction as opposed to standard quantum measurement.4 The paper has not been through peer review, does not touch on biology or microtubules, and demonstrates a physical mechanism rather than anything about the brain. It is worth naming because it illustrates why the theory keeps attracting serious attention despite the scepticism: parts of it can, in principle, be checked against hard physics, one experiment at a time, rather than argued about in the abstract indefinitely.
Why a theory built on doubt keeps getting taken seriously
Penrose’s standing helps explain why. He shared the 2020 Nobel Prize in Physics for work, done in the 1960s, proving that black holes are a robust prediction of general relativity, work with no connection to his consciousness theory and never in dispute.5 That track record buys Orch-OR a hearing it might not otherwise get. As the science writer Steve Paulson put it after interviewing Penrose for Nautilus, the reaction among physicists tends to run along similar lines: the theory is almost certainly wrong, but Penrose is serious enough that dismissing it outright feels premature.6
The deeper draw, though, is what the theory is reaching for rather than the mechanism itself. Penrose’s starting point is that human mathematical insight, the capacity to see that a formal statement is true rather than merely to follow rules that generate it, cannot itself be the output of a rule-following system, an argument he ties to Gödel’s incompleteness theorems. Whether or not that argument holds, and many logicians think it does not, it points at something real: current computational accounts of mind do not explain why understanding feels like anything from the inside. Orch-OR is one answer to that gap. It is not the only one, and on the available evidence it is not the leading one. But it is among the few that says plainly where a physicist could go looking for proof.
What to watch next
The theory’s fate now rests less on argument than on instruments. Further underground collapse experiments, refinements of the Gran Sasso approach, and continued work on whether microtubules can sustain coherence at biologically relevant timescales will each chip away at, or shore up, specific pieces of the model.
None of that will settle, on its own, whether Penrose is right that consciousness lies outside computation altogether. That is a question physics experiments can constrain but not fully answer.
