Rethinking Consciousness: A Scientific Theory of Subjective Experience

by Michael S.A. Graziano

We apply the same process to more than just people. We attribute awareness to our pet cats and dogs, and some people even swear that their houseplants are conscious. The ancients felt sure that trees and rivers were sentient; children perceive consciousness in their favorite toys; and heck, the other day I got mad at my computer. So I’m not talking about intellectually figuring out whether something has a mind or cleverly deducing what might be in that mind—although we do that, too. I’m talking about an automatic, gut intuition, which is often wrong but sometimes persuasively potent, that an essence of awareness is emanating from an object.

These numbers may change with new data, but as a plausible, rough estimate, it seems that neurons, the basic cellular components of a nervous system, first appeared in the animal kingdom somewhere between sponges and sea jellies, a little more than half a billion years ago.

After the emergence of the nerve net, however, nervous systems rapidly evolved a second level of complexity: the ability to enhance some signals over others. This simple but powerful trick of signal boosting is one of the basic ways that neurons manipulate information. It is a building block of almost all computations that we know about in the brain.

Signals compete with each other, the winning signals are boosted at the expense of the losing signals, and those winning signals can then go on to influence the animal’s movements. That is the computational essence of attention. Our human attention is merely an elaborated version of it, made of the same building blocks.

Octopuses, squid, and cuttlefish are true aliens with respect to us.15 No other intelligent animal is as far from us on the tree of life. They show us that big-brained smartness is not a one-off event, because it evolved independently at least twice—first among the vertebrates and then again among the invertebrates.

Once you start with the intuition that consciousness arises naturally from complex information processing, it’s hard not to slip into panpsychism, the belief that everything in the universe is conscious to at least some degree.

It may be a simpler, less computationally expensive solution for the zebra to build a schematic model in which the lion has a mind, the mind can take possession of items in the world, and once having done so, the mind can then guide the lion’s actions.

It’s a favorite trope that we humans have a superior consciousness. We tend to think that other animals are not conscious at all or that their consciousness is less developed. This perspective dovetails with the popular idea that consciousness emerges from complexity. Because we have the most complex brains of all animals, we must also have the most vivid consciousness. Yet of all the mental talents that we humans like to brag about—math, language, tool use, and so on—consciousness may be one of the most primitive and least special to us.

But the fact of consciousness itself, our ability to have a subjective experience of anything at all and to attribute subjective experience to others, is of such basic utility that it may be shared across a vast range of the animal kingdom. If the attention schema theory is correct, then it is certainly not unique to humans.

Almost everyone has had the spooky feeling of being stared at from behind, as if a beam of energy is touching the back of your neck. In 1898, the psychologist Edward Titchner thought the belief was widespread enough to be worth testing.14 In controlled experiments, he found that people can’t directly feel each other’s stares, no matter how compelling the psychological impression may be.

The belief that vision involves something beaming out of the eyes is so intuitive—built so deeply into the way we understand active vision—that it is the default belief among children, as the psychologist Jean Piaget discovered.15 We’re only later taught the scientifically correct account.

Maybe there’s no fundamentally inexplicable, nonphysical essence in us. Maybe our task as scientists is to explain why people tend to believe in a hard problem in the first place.

You will never be able to close that gap. You can’t use superficial, intellectual knowledge or a few hours of thought to erase an attention schema that was shaped through millions of years of evolution and is constructed deep within your system.

And yet, even though every educated person now understands that white light is a mixture of all colors, that knowledge doesn’t change the models built into the visual system.

Arguably, science is the gradual process by which the cognitive parts of our brains discover the inaccuracies in our deeper, evolutionarily built-in models of the world.

The answer, I believe, is that people generally don’t have that intuition. We don’t understand consciousness as being physically substanceless. Instead, we understand it as something for which physical attributes are irrelevant. And those two intuitions are very different.

The hard problem derives from assumptions that come from that deep, subsurface model, the attention schema. The attention schema theory is a meta-answer that explains why people believe in a hard problem in the first place.

For example, suppose a complaining friend tells you, “I swear, my manager’s competence is an illusion.” That person doesn’t mean, “He’s a competent manager, but in a slightly different way than you might expect. In fact, he might even be more competent than you expect.” No, your friend means that his manager doesn’t have any competence. In the context of a metaphor, to call something an illusion is to deny every aspect of its existence.

Most vision scientists understand the word illusion in this sense of a glitch or aberration from the normal.

The knowledge presented itself as cognitive certainty. This strange phenomenon is called blindsight, and it is a consistent result of damage to the primary visual cortex.

This experiment was a watershed moment in the study of consciousness. It showed finally that the mechanisms of attention could exist even when the mechanisms of consciousness were broken. Clearly, attention is not just a local concentration of consciousness. It is a different property.

If you can understand that Sally has a mind, that her mind holds beliefs, and that her beliefs can contradict reality, then you can solve the task. Sally will first look in basket A, where she originally put the sandwich. As trivial as the task seems to human adults with years of social experience, children under the age of about 5 can’t solve it, and very few nonhuman animals show evidence that they can.5

In the attention schema theory, to go about determining whether a machine is conscious, we should probe its innards to find out whether it contains an attention schema, and we should read the information within the attention schema. We will then learn, with objective certainty, whether this is a machine that thinks it has a subjective conscious experience in the same way that we think that we do. If it has the requisite information in that internal model, then yes. If not, then no. All of this is, in principle, measurable and confirmable.

My point is that in the attention schema theory, consciousness is not, by definition, always and forever private. It isn’t true that “I know I’m conscious but can never really know if you are.” Finding out whether a brain thinks it has consciousness in the same way that I do is a matter of developing the technology to read the information in it.

But our hypothetical machine contains all the same components that you do, including an attention schema that contains the same information within it. On consulting its attention schema, it, too, would be informed that it has a private, personal, and immaterial feeling. Both you and the machine, constructed in the same way, are stuck in the same logic loop. You know only what you know and can report only the information within you—and that information can, in principle, be probed by someone else.

but at the same time, attention is not at the mercy of the environment—it’s also under your own control, so that an internal directive can shift attention from item to item. Above all, attention can shift easily between very different domains of information. It can be directed to spatial locations (the main property of attention that has been incorporated into artificial systems so far), but can also be focused on color, movement, taste, tactile pressure, or even something internal like a memory or an idea.

But I suspect that the hardest nut to crack will be emotion. It’s the least understood domain of information in the brain. The best that can be said of it is that many brain structures responsible for emotion have been located.11 People sometimes call them an emotional network or an emotional circuit, but the system isn’t understood at the precision of a circuit.

It is still an open question as to whether emotions are a necessary component of consciousness. They are something we deal with and can influence consciousness but that is a different issue than whether emotional states are necessary to be conscious.

That hypothetical map of all the neurons and their synaptic connections is called a connectome,6 a word intentionally analogous to the genome. The idea is that if scientists were able to map the human genome, an accomplishment once thought to be impossible, then they can tackle an even greater technical challenge and map the human connectome. Each person has his or her own unique connectome, defining a unique mind.

The best that can be said at this time is that our knowledge of the processing in the brain has some humbling and gigantic holes.

Maybe the better term is the “foundation” world and the “cloud” world. The foundation world would be full of people who are mere youngsters—less than 80—and who are still accumulating valuable experience. Their unspoken responsibility would be to grow up, tune up, and gain wisdom and experience before joining the ranks of the cloud world. The balance of power and culture would shift rapidly to the cloud world. How could it not? That’s where the knowledge, experience, and political connections will accumulate. In that scenario, the foundation world becomes a kind of larval stage for immature human minds, and the cloud world would be where life really begins.