A paper in The British Medical Journal in December reported that cognitive behavioral therapy — a means of coaxing people into changing the way they think — is as effective as Prozac or Zoloft in treating major depression.
In ways no one understands, talk therapy reaches down into the biological plumbing and affects the flow of neurotransmitters in the brain. Other studies have found similar results for “mindfulness” — Buddhist-inspired meditation in which one’s thoughts are allowed to drift gently through the head like clouds reflected in still mountain water. Findings like these have become so commonplace that it’s easy to forget their strange implications. Depression can be treated in two radically different ways: by altering the brain with chemicals, or by altering the mind by talking to a therapist. But we still can’t explain how mind arises from matter or how, in turn, mind acts on the brain. This longstanding conundrum — the mind-body problem — was succinctly described by the philosopher David Chalmers at a recent symposium at The New York Academy of Sciences. “The scientific and philosophical consensus is that there is no nonphysical soul or ego, or at least no evidence for that,” he said. Descartes’s notion of dualism — mind and body as separate things — has long receded from science.
The challenge now is to explain how the inner world of consciousness arises from the flesh of the brain. Michael Graziano, a neuroscientist at Princeton University, suggested to the audience that consciousness is a kind of con game the brain plays with itself. The brain is a computer that evolved to simulate the outside world. Among its internal models is a simulation of itself — a crude approximation of its own neurological processes. The result is an illusion. Instead of neurons and synapses, we sense a ghostly presence — a self — inside the head. But it’s all just data processing. “The machine mistakenly thinks it has magic inside it,” Dr. Graziano said. And it calls the magic consciousness. It’s not the existence of this inner voice he finds mysterious. “The phenomenon to explain,” he said, “is why the brain, as a machine, insists it has this property that is nonphysical.”
The discussion, broadcast online, reminded me of Tom Stoppard’s newest play, The Hard Problem, in which a troubled young psychology researcher named Hilary suffers a severe case of the very affliction Dr. Graziano described. Surely there is more to the brain than biology, she insists to her boyfriend, a hard-core materialist named Spike. There must be “mind stuff that doesn’t show up in a scan.” Mr. Stoppard borrowed his title from a paper by Dr. Chalmers. The “easy problem” is explaining, at least in principle, how thinking, memory, attention and so forth are just neurological computing. But for the hard problem — why all of these processes feel like something — “there is nothing like a consensus theory or even a consensus guess,” Dr. Chalmers said at the symposium. Or, as Hilary puts it in the play, “Every theory proposed for the problem of consciousness has the same degree of demonstrability as divine intervention.” There is a gap in the explanation where suddenly a miracle seems to occur. She rejects the idea of emergence: that if you hook together enough insensate components (neurons, microchips), consciousness will appear. “When you come right down to it,” she says, “the body is made of things, and things don’t have thoughts.”
Proponents of emergence, who have become predominant among scientists studying the mind, try to make their case with metaphors. The qualities of water — wetness, clarity, its shimmering reflectivity — emerge from the interaction of hydrogen and oxygen atoms. Life, in a similar way, arises from molecules. We no longer believe in a numinous life force, an élan vital. So what’s the big deal about consciousness? For lack of a precise mechanism describing how minds are generated by brains, some philosophers and scientists have been driven back to the centuries-old doctrine of panpsychism — the idea that consciousness is universal, existing as some kind of mind stuff inside molecules and atoms. Consciousness doesn’t have to emerge. It’s built into matter, perhaps as some kind of quantum mechanical effect. One of the surprising developments in the last decade is how this idea has moved beyond the fringe. There were three sessions on panpsychism at the Science of Consciousness conference earlier this year in Tucson. This wouldn’t be the first time science has found itself backed into a cul-de-sac where the only way out was proposing some new fundamental ingredient. Dark matter, dark energy — both were conjured forth to solve what seemed like intractable problems.
Max Tegmark, a physicist at the Massachusetts Institute of Technology (he also spoke at the New York event), has proposed that there is a state of matter — like solid, liquid and gas — that he calls perceptronium: atoms arranged so they can process information and give rise to subjectivity. Perceptronium does not have to be biological. Dr. Tegmark’s hypothesis was inspired in part by the neuroscientist Giulio Tononi, whose integrated information theory has become a major force in the science of consciousness. It predicts, with dense mathematics, that devices as simple as a thermostat or a photoelectric diode might have glimmers of consciousness, a subjective self. Not everything is conscious in this view, just stuff like perceptronium that can process information in certain complex ways. Dr. Tononi has even invented a unit, called phi, that is supposed to measure how conscious an entity is. The theory has its critics. Using the phi yardstick, Scott Aaronson, a computer scientist known for razorlike skepticism, has calculated that a relatively simple grid of electronic logic gates — something like the error-correcting circuitry in a DVD player — can be many times more conscious than a human brain.
Dr. Tononi doesn’t dismiss that possibility. What would it be like to be this device? We just don’t know. Understanding consciousness may require an upheaval in how science parses reality. Or maybe not. As computers become ever more complex, one might surprise us someday with intelligent, spontaneous conversation, like the artificial neural net in Richard Powers’s novel Galatea 2.2. We might not understand how this is happening any more than we understand our inner voices. Philosophers will argue over whether the computer is really conscious or just simulating consciousness — and whether there is any difference. If the computer gets depressed, what is the computational equivalent of Prozac? Or how would a therapist, human or artificial, initiate a talking cure? Maybe the machine could compile the counselor’s advice into instructions for reprogramming itself or for recruiting tiny robots to repair its electronic circuitry. Maybe it would find itself flummoxed by its own mind-body problem. We humans may not be of much help.