What Is the Hard Problem of Consciousness?
It’s a tale of two datasets….
Brad has volunteered to sit in a lab, stare at a screen, and say out loud what he sees. All the while, parts of his brain are being observed and recorded.
Later, in another lab, Amy can look at the recordings of Brad’s brain and know when he saw human faces… or when he felt afraid… or any number of other things Brad experienced while watching the screen, depending on what view of Brad’s brain she has, even though she cannot see what he saw, or feel what he felt. Brad, for this part, had no conscious experience whatsoever of the neural activity in his own brain which allows Amy to deduce what he was experiencing.
So here we have two different sets of observations — Amy’s observations (set A) of what’s going on in Brad’s brain, and Brad’s own observations (set B) generated by his brain. Amy can make correct deductions about set B on the basis of set A because our conscious experiences are tightly connected with the actions of our brains. These consistent and predictable connections are called the neural correlates of consciousness, or NCCs.
The so-called “hard problem of consciousness” is our inability to explain these correlations in a rigorous, scientific way. Specifically:
- Why and how does brain activity produce any conscious experience? Why don’t we simply respond to the world like plants or machines, without our brains creating conscious “percepts” such as colors and odors and sounds?
- Why is some activity in the brain correlated with conscious experience, while other brain activity is not? And why doesn’t any other organ do this?
- Why are the NCCs what they are and not something else? Why does bouncing certain wavelengths of light off our retinas, for example, cause us to see yellow rather than red, or to have some experience other than color. (And no, it’s not because the light itself is yellow, as we shall see.)
Description is not explanation
Before delving into these questions, we need to first understand what exactly a “hard problem” is, and why this is one of them. A hard problem isn’t one that is extremely complex, or involves a lot of work to solve. The solution could turn out to be quite simple. Rather, it’s one that we not only have no way of answering at the moment, but no way of thinking about constructively, no tools with which to craft an answer, or meaningfully search for one other than taking stabs in the dark until, hopefully, we hit something.
For example, if you lived in the Middle Ages, the aurora borealis or “northern lights” would have been a hard problem. Quite literally, you could have known everything there was to know at the time about the northern lights, or indeed everything that possibly could have been known then, and you still would have had no way of explaining what was going on. No amount of observation, no amount of research could have helped you. Nor could you have figured out why you had no way of finding the answer.
That’s because in order to understand the northern lights, you first need to have an understanding of solar plasma, planetary magnetic fields, and the Earth’s ionosphere. Without that scientific foundation, not only can the northern lights not be explained, but there is no way of even framing a potentially correct explanation, no way of steering the investigation in the right direction. And no way of knowing what the missing bits are that would lead to a solution.
The NCCs are not an explanation of consciousness. They are merely a description of what we observe. Our situation is similar to the problem of gravity in the two centuries between the work of Isaac Newton and Albert Einstein, when it was also a hard problem.
Newton provided a revolutionary description of gravity. According to his observations and calculations, gravity was a universal force of attraction among massive objects which was proportional to the product of their masses (mass 1 multiplied by mass 2) and inversely proportional to the square of the distance between their centers, so if you double the distance between two objects the force is one fourth as strong, if you triple the distance the force is one ninth as strong, and so on. Now, that’s all very important, but it doesn’t explain why there is a universal attractive force rather than things just floating around freely, or why that force is proportional to the product of the masses rather than, say, their sum, or why it decreases as the square of the distance rather than the cube or any other value.
Similarly, we can’t point to the behavior of the brain and say, “Well, that explains it, consciousness is the brain activity.” Or put another way, we can’t say “Data set B is data set A” even if we know that one causes the other. Because if they are the same, why are they so different? Why isn’t Brad’s observation exactly the same as Amy’s — why doesn’t he have a conscious experience of neurons firing? And why is it different in the particular way that it is, rather than some other way? Just as Newton’s description of how gravity worked was not an explanation because it failed to tell us why it was like that rather than otherwise, or why it existed in the first place, the NCCs are also merely a description which do not tell us why the NCCs are what they are rather than something else, or why they exist at all.
The truth isn’t “out there”
Now you might be thinking, the reason we see yellow when a certain sort of light hits our eyes is simple — the light is yellow, and we are perceiving the color of the light. But light is not yellow, or any other color. Colors are something done by brains.
Photons, or “particles” of light, are excitations of the electromagnetic field. When measured as waves, we can say that different instances of light have different wavelengths and amplitudes (different widths and heights). The retinas at the backs of our eyes respond differently to different sorts of light by sending varying sorts of neural impulse patterns down the optic nerve. There are two key features of this process to note here.
First, photons are photons and waves are waves. Photons are “quanta” of energy. Waves are essentially patterns of motion. And that is all that they are. There is no way for them to have a color. Where would they have it?
Second, neurons are neurons. They are not light, and they can only do what neurons do. They cannot do what light does. So even if color were a feature of light, it wouldn’t matter, because the light doesn’t get into your head. Nor does the light somehow take color out of its pocket and shove it down your optic nerve toward your visual cortex.
There is an image from old textbooks which exists in many variations and which has likely misinformed millions of young minds. An outline of a human head is shown facing an object. (Trees are quite popular, for some reason.) Two lines extend from the top and bottom of the object and converge at a point on the eye lens. Inside the eye, the lines diverge again, ending on the retina where the object reappears upside down, “projected” in reverse. From there, the image is then re-projected into the brain cavity, appearing rightside up again.
Almost every element of this illustration is wrong.
Not only is there no image of a tree (or anything else) on your retina, it is not even possible to deduce what you visually experience by examining what’s going on with your retinal tissue. A very clear example of this comes from cases of visual rivalry, in which a subject looks at an image that can be consciously seen in one of two ways, but not both at the same time. The information on the retina remains steady, while the conscious percept shifts back and forth. The most well known example of such images is the Necker cube, a “stick figure” box that can be seen as a transparent cube viewed from above or below (or as a flat pattern). There are many other examples, such as competing vertical and horizontal color bars (as seen here) which appear to fade, vanish, and reappear.
Tracing the NCCs involved in these experiences of visual rivalry is complex. No evidence of the shifts in conscious percepts are detectable on the retina or optic nerve. In the thalamus, which contains a sort of routing station between the retina and visual cortex, some small fraction of neurons may show evidence of discrimination, or none may, depending on the stimulus. In primary visual cortex, the percentage of neurons showing differences in firing patterns which correspond to shifts in conscious percepts is quite low. This percentage rises through subsequent stages of visual cortex, until “in the highest echelons of the… visual stream” the great majority of neural activity exhibits a correlation with what the viewer consciously sees. (Kreiman, “Neural Correlates of Consciousness: Perception and Volition”, The Cognitive Neurosciences, 5th ed., 2014)
Plus, there’s a lot going on in the retina that never makes it to conscious awareness at all. For example, our retinas respond to TV monitor refresh rates that are too quick for us to be consciously aware of. And our eyes are constantly making rapid shifts called saccades which we never “see”. And of course, the optic nerve itself creates a well-known retinal “blind spot” which we do not experience. On the flip side, when we dream, our visual cortex produces all sorts of imagery which corresponds to no retinal activation whatsoever. (Kreiman)
The image of the tree is produced by your visual cortex in combination with other areas of your brain. It never exists in your eye. Nor does the tree “look like that” in any objective sense.
The same can be said for all of our senses. The molecules in our noses don’t have any scent inside them, and the molecules on our tongues don’t contain any flavor. The air bouncing off our eardrums doesn’t hold any sound. Even the sensations of our own bodies are produced within our brains, and are not “perceived” by them in the form that we feel them.
A doorbell, not a projector
Rather than imagining a projector and a screen when thinking about vision, we would be better off imagining a doorbell: A finger presses a button which sends electrical impulses down a wire to a bell. Similarly, light strikes the retina which sends biochemical impulses down nerves to… well, to the rest of the brain. We don’t really understand the “bell” part of our biological apparatus yet, and the bits we do understand will have to wait for another article.
That model is oversimplified in some very important ways. For example, our brains aren’t so linear. The neural firings don’t go charging forward in a straight line toward their destination. They’re more like a brigade which sends scouts out in front who circle back with information, while exchanging messages with other brigades on other missions, as well as receiving communications from central command, all of which alters their routes and formations and tactics. It’s a process often described as a “cascade”.
If you saw the 1987 spy thriller No Way Out, then you’ve seen a similar process played out on a fictional computer. In the film, the computer is tasked with enhancing a blurry photo of a man’s face. It does this by making millions of tiny guesses about what an unblurry image of the face might look like, tossing out results that make it look less like a male human face, and keeping those that make it look more like one. Our brains do this in mere fractions of a second, crafting an experience of the world that best matches what the brain expects, based on a combination of what it is born to expect and what experience leads it to expect.
Nevertheless, for our purposes here, the analogy of the doorbell is still useful. Light waves striking the retina — or molecules landing in our noses or on our tongues, or percussion waves in the air striking our ears — are like the finger pressing the button. Nothing about the finger transfers to the button, except the pressure it exerts. The button doesn’t receive any of the characteristics of the finger. By the same token, nothing about the button transfers to the wire. The button does buttony stuff and the wire does wirey stuff. They are in a chain of reaction, and are not carrying anything through from one to the other like relay racers handing off a baton. There is no baton. Just as there is no ringing of a bell moving through the wire, there is no image of a tree or odor of pine needles or sound of birdsong moving through our nerves from our retinas and noses and ear drums.
There is also no way to predict, by examining what’s going on in the wire, what will eventually happen at the other end. Perhaps a bell will ring. Or perhaps not. Perhaps a deaf person lives at this house and the electrical impulses will cause a light to flash. Any number of things could occur. A spray of perfume could be emitted. The floor could be made to vibrate. A bomb could go off. Or nothing could happen at all. Who knows? The wire has no say in the matter.
Similarly, evolution can do all sorts of things at the end of the wire. And understanding this point is crucial to understanding the hard problem of consciousness. If it were true that we experience the world as we do because that is how the world is, then the hard problem would not exist, or at least not in the form that it does. But evolution was not obliged to turn light into colors, or molecules into flavors and odors, or to render anything in the way we consciously experience it. So why and how does it generate the particular percepts that it creates, rather than some other set of percepts, or none at all?
Of sharks and birds and magnets
The basic building blocks of our conscious experience are referred to in some circles as qualia (pronounced like KWA-lee-uh). The human brain produces a particular palette of qualia — sounds, scents, flavors, colors, sensations, emotions, and so forth — which are integrated into a unified conscious experience which seems to us to be the outside world, and our own bodies.
In reality, all of that is inside our heads. And we know this to be true in a variety of interesting ways. Take, for example, the condition of synesthesia, which results from activity in one area of the brain crossing over and triggering another area in an atypical way, so that words have flavors, for example, or numbers have colors, or textures trigger emotions. If it is true that qualia are an inborn function of the brain, and not perceptions learned from experience, then it should be true that people who have synesthesia involving colors and who are also colorblind will experience the “missing” colors in their synesthetic experiences, as the neural regions responsible for producing them are cross-stimulated. While performing research on synesthesia, the neuroscientist V. S. Ramachandran encountered a study volunteer who was red-green colorblind. And in fact “he often saw numbers that were tinged with colors that he never saw in the real world…. He could only see these when looking at numbers.”
[The subject’s] cone receptors are deficient, but the problem is entirely in his eyes. His retinas are unable to send the full normal range of… signals up to the brain…. At the same time, he is a number-color synesthete. Thus the number shapes are processed normally all the way up to his fusiform and then, due to cross-wiring, produce cross-activation of cells in his [visual cortex] color area. Since [he] has never experienced his missing colors in the real world and can do so only by looking at numbers, he finds them incredibly strange. (Ramachandran, The Tell-Tale Brain, 2011)
From human to human, there are naturally occurring variations in the qualia produced by the brain. Other animals have different palettes of qualia, which may or may not overlap with ours. For example, both birds and sharks respond to magnetic fields, which humans have no conscious representation for. Birds navigate by them, and sharks find them repulsive. What we cannot know is how the brains of sharks and birds render magnetic fields as percepts. Whatever it is that these animals consciously experience, we are utterly unable to comprehend it, since our brains are not built to perform the same trick.
Sometimes you will see photographs altered to show “what a flower looks like to a bee” or something like that, by rendering ultraviolet or infrared wavelengths of light as some kind of human-perceptible color. Similarly, images from the Hubble deep space telescope are crafted by assigning human-palette colors to various wavelengths of radiation. But all of that is artificial and arbitrary, a translation of the world into human-consciousness-ese.
In truth, the universe doesn’t look like anything at all. It doesn’t sound like anything. It doesn’t smell like anything. It doesn’t taste like anything or feel like anything. All of that goes on in the mind and is produced by the brain. The hard problem of consciousness asks how the brain does that, and why it does it in the particular way that it does, and not in any other way.
What the hard problem is not
The term “hard problem of consciousness” was coined a quarter-century ago by David Chalmers. And it is often described as the problem of how the physical processes of the brain can give rise to the non-physical mind. But such loose philosophical descriptions are, at best, not useful and, at worst, profoundly misleading. It is tempting to go back to Chalmers’ original conception of the problem in an attempt to understand it, but neuroscience has made great strides since the 1990s, and it makes no more sense to refer to Chalmers’ formulation of the hard problem as the basis for what it means today than it would to refer to Lemaître’s work in the 1920s to understand the current science of the Big Bang, or to Wheeler’s writings of the 1960s to understand the current science of black holes.
Another common misunderstanding is that the “hard problem” involves the inherent “subjectivity” of conscious experience. In reality, consciousness is the stuff of observation itself, and all observations are inherently and equally subjective. Amy cannot have Brad’s experience of seeing the yellow light, but by the same token, Brad cannot have Amy’s experience of observing the measurements of his brain activity. That said, the experiment can be repeated with the roles reversed and Amy can have her own experience of yellow while Brad can have his own experience of viewing the results of the neural measurements. Or Brad can have his own experience of viewing the scans of his brain. But that still will not be Amy’s experience of it.
A third misconception of the problem is that it doesn’t exist, that the problem is solved by the NCCs themselves. But as we have seen, knowing that neural activity is correlated with conscious experience does not tell us how or why this is so, and that is the heart of the problem. It is a tale of two datasets, and our inability not only to answer those questions of how and why, but also our utter befuddlement at what qualia even are, or how to begin designing studies or experiments that might lead to an answer.
Where to from here?
One thing we do know — following the chain of neural impulses that produces other sorts of brain-driven functions like moving our muscles is not going to produce an answer. Consciousness is qualitatively different, a new trick that some animal brains learned to perform somewhere along the evolutionary path.
Some have fallen into the temptation of assuming that all the stuff of the universe itself is conscious, a position known as panpsychism. That is certainly a convenient way of removing the problem, but the existing research into conscious percepts indicates that it simply isn’t so. We know, for example, that consciousness involves the integration of synchronized neural activity in disparate regions of the brain. (Tononi and Balduzzi, “Toward a Theory of Consciousness”, The Cognitive Neurosciences, 4th ed., 2009; Carter et al, “Consciousness”, The Human Brain, 2009) It is an evolved, complex bodily function. There is no reason to believe, and every reason not to, that either the universe as a whole or life forms which do not possess complex brains can generate conscious experience.
Because consciousness is still a hard problem, nobody knows or can know right now where the answer will come from. But there are functions and features of the brain to examine other than the neural chain. For example, conscious electromagnetic field theory, or CEMI for short, proposes that evolution used the electromagnetic “noise” produced by the brain to generate conscious percepts in the form of a hologram-like phenomenon inside the skull which was (and is) useful for navigation of the environment. It’s impossible to say if anything will come of that research, but for the moment we are going to have to continue to be open-minded, to rethink the problem in new ways, and continue to take shots in the dark until something says “Ouch!”
And if we are to value and properly understand research into our own conscious minds, and those of other animals, it is imperative that we understand what the hard problem is, as well as what it isn’t.