Consciousness is one of the most heavily debated areas in cognitive science and philosophy. Why is consciousness a problem?

We haven’t found a way to ask the right questions. To ask “What is consciousness?” is just too broad – it’s like asking what memory is, or what attention is. But asking a narrower question doesn’t seem to capture what we’re really interested in. I think this is often because when we ask what consciousness is, we’re looking for some answer beyond the expected scientific answer – and so any kind of scientific explanation seems insufficient. When we’re given evolutionary explanations for the functions that consciousness provides for us, we say, “Fine, that’s why evolution led to consciousness, but how does an individual brain give rise to it?” And when we’re given some brain processes that are responsible for conscious awareness, we still aren’t satisfied as to why those brain processes, and not others, should produce this magical feeling of being conscious.

What kind of explanation do you think would satisfy us when we ask what consciousness is?

I’m not sure. I don’t actually think people want to be told that there’s some magical non-physical substance called consciousness, but I think that’s what it feels like. As long as we remain without any way to bridge the gap between subjective feeling and objective laws of cause and effect, no answer is going to satisfy us. And if we aren’t given such a bridge, then we want to hear something that aligns with our intuitive experience, which is that the laws governing consciousness are entirely different than those governing physical entities.

What do you think is the most intuitive way to think about consciousness from a scientific perspective, then? What’s the best way to bridge the gap (assuming the magical fairy dust explanation of consciousness isn’t true)?

I think it’s important to remember that most of the assumptions we act on are illusions. Our entire visual world is an illusion – the way we instantly group objects, assume continuity of form, project a two-dimensional image into a three-dimensional world – it’s all a bunch of transformations the brain is providing for us. I think consciousness is similar. The way our brain systems interact gives rise to certain perceptions about the way our internal world relates the external world.

You say “the way our brain systems interact” is giving rise to consciousness. Does this mean you don’t think that consciousness is localized to one part of the brain?

Absolutely not. I don’t think you could build a conscious machine just by building some particular “consciousness areas” correctly. It’s true that certain regions have been shown to indicate differences in conscious perception, specifically, regions responsible for object recognition. I won’t deny that those regions are necessary, but it certainly seems insufficient to say that the activation in the object recognition areas leads to conscious experience. Maybe whatever is active in the object recognition regions is what’s visually conscious, but there’s more to consciousness than the pattern of activation in a certain area. An explanation like that is what leads to what I was talking about before, which is the characteristic dissatisfaction with how something as objective as a pattern of activation leads to something as subjective as consciousness.

Let’s imagine a simulation of the visual system. Imagine it has all the steps of the hierarchy – taking in some visual input, processing it gradually more and more broadly until it reaches the top of the hierarchy as a so-called object. It’s not evident how consciousness arises from the recognition of an object, and that’s why there’s often dissatisfaction when we’re provided with neural correlates of consciousness.

So what do you think is missing?

We need to understand the role of the other systems that act on the object recognition system, or that act on any system that is capable of producing conscious representations (for example, of sound or action).

What kind of systems are you talking about? What do you think these systems do?

The systems that plan, direct, and guide perception and action. I think they provide us with an inescapable illusion of free will.

The “illusion” of free will? So I take it you don’t believe in free will?

I don’t believe in it in the absolute sense. Sure, I don’t think there’s something above the physical world that acts on the physical world. But I also think we can’t be conscious without the assumption that we are above the physical world and are acting on it from above. Even if we decide that we don’t have free will, even if we decide that everything is just cause and effect, so we’re not going to take any action – that decision in itself is a choice about how to relate to the external world, and we’re aware of that choice. We’re stuck in the world that the existentialists described – one in which we’re condemned to freedom – except instead of being condemned to freedom, we’re condemned to the illusion of freedom, and no rational understanding of its illusory nature will release us from our conscious perception of free will.

So what exactly is the connection between free will and consciousness? Free will inevitably crops up in any discussion of consciousness, but I think it’s important to delineate what exactly the relationship is between free will and consciousness. How would you characterize it?

I think free will is the basis for consciousness. I think if we explain the illusion of free will, we’ve explained what it means to be conscious.

Conscious action is controlled action, that is, action controlled by the self: the mysterious self with free will. It’s hard to imagine a string of consciously perceived actions that we felt were not being controlled by ourselves, but instead by some physical process of neurons firing. It’s unclear what a conscious perception of uncontrolled action would even be like – perhaps as if our body were possessed by some external force. But as long as we don’t have a pathological condition, we don’t perceive ourselves as being possessed, watching our arm move magically. We feel ourselves moving it.

An even more compelling example, though, is that of thoughts. When we have thoughts, we have a distinct feeling that our thoughts are ours. We even feel responsible for our thoughts, feeling guilty for shameful thoughts like wishing something bad on someone else. Just as with actions, we have to ask – how the heck could we have a conscious thought without the sense that it was ours? It would be like some third-party speaking to us inside our brain. Again, without pathology, this is out of alignment with our everyday experience of what it feels like to think.

I should mention that whether we really have the ability to make choices freely and think freely in the most fundamental sense is irrelevant. We can assume either that no one is making choices, and there are only natural laws of cause and effect, or we can assume that there’s some omnipotent being who’s sitting with a remote control and pressing buttons to control each of our actions. The argument remains the same. Choice and control is the stuff of consciousness. Lack of choice – running on autopilot – isn’t. It doesn’t actually matter whether choice is a reality or not.

What do I think? I think the reality of it is that both unconscious and conscious thought – both “running on autopilot” and making “free” choices – can be reduced to the same thing: brain processes giving rise to thought and action. It all comes down to patterns of neuronal firing. On one hand, we have well-practiced, largely unconscious patterns that support unconscious thought and action. On the other hand, we have conscious thoughts and actions which seem to be supported by at least partially “new” patterns of activation. But even if conscious thoughts are entirely “new,” the cognitive control that guides them must be at least partially made up of old, well-practiced patterns.

So by that explanation, unconscious and conscious thought are the same, yet conscious thought feels qualitatively different. What’s the difference?

In the simplest sense, maybe it comes down to whether we’re exerting cognitive control over what we’re thinking or doing. I think the neural explanation for the free will illusion lies in the systems for cognitive control. This is not to say that we are conscious of cognitive control. I don’t think that’s true at all. I don’t think we’re aware when we shut certain things out and let other things in. That would be cognitive control of our cognitive control. That’s something entirely different. What I’m saying is that the cognitive control itself is the stuff of consciousness. Not that the frontal brain regions involved in cognitive control are the location of consciousness, but that their connection to other sensory and motor regions is what gives rise to consciousness.

Judging from your answers, I’m guessing you think there’s no consciousness after death?

Right. Consciousness can’t exist without a functioning brain.

So what do you think about zombies – could there feasibly be a zombie who acted exactly like us but had no consciousness?

No. I think to act like us – to have the kind of “control” over your actions, to have the same kinds of rules guiding action – you have to have the systems that make up consciousness.

But all zombies would have to do is react correctly to the world around them. Does it really matter if they’re really operating according to the same internal laws as us, as long as they always react the right way?

I think that argument would hold only if there were a finite number of possible states of the external world that the zombie would have to react to. Then you could index all the states and the corresponding correct reactions. But this doesn’t work, because the world creates infinite possibilities, so we need some ways of processing information, some ways of interpreting and reacting to it. My argument is that the only way to consistently react like a conscious human is to have the systems that control thought and action – and those are precisely the systems which, when acting on the perceptual and motor systems, make up consciousness.

Do you think non-humans can have consciousness? Many people argue that more complex animals probably have consciousness, but those lower in the hierarchy of complexity probably don’t. What are your thoughts on this? How complex of behavior do you think an animal has to have in order to be conscious? In other words, what are the “minimum requirements” for consciousness?

I think it’s begging the question to say that consciousness is a matter of complex behavior. I’d argue that it’s not about the complexity of behavior, but about the complexity of the systems that control behavior. We have all sorts of systems feeding in to the frontal regions responsible for control – emotion, attention, memory – all playing complex roles in how thought and action are governed. I think you could have an unconscious being with pretty complex behavior (for example, a computer running Windows), or a conscious being with only very simple behavior (but complex control over it). Imagine, for instance, a thermostat, that has only one dimension of behavior: a number which rises and falls. We might at first think that a thermostat is certainly not conscious, because of the simplicity of its so-called behavioral responses. But what if it “decides” whether to change temperature based on some complex principles – say, not just the external temperature as might be expected based on the ultimate goals of the system, but also whether it’s bored because it has stayed the same temperature for a long time, whether it’s changed a lot recently and is tired – and say these factors are often in competition and work against each other, and maybe even work against the system as a whole sometimes. Now we’re getting closer to consciousness, but the behavioral outputs are still about as simple as they come.

So it sounds like you’re saying that a conscious being is necessarily irrational.

Maybe. I think if we were able to act purely rationally, that is, in perfect alignment with “reality” – with all time, all space, all possible outcomes – we wouldn’t need conscious thought. And along the same line of thought, I think it’s hard to imagine building an irrational, yet unconscious machine. It would have to be programmed not just to act against rationality, but to act based on some other set of laws – rooted in emotion, perhaps – which might imply consciousness.

When did you first become interested in consciousness?

When I was sixteen, I started thinking a lot about free will. I started contemplating the nature of human action, and I didn’t even realize that what I was really thinking about was consciousness. What started me thinking about free will was some moral discussions in my high school literature class. It seemed like fate and free will were constantly being pitted against one another, and I was really bothered by this, because I didn’t “believe” in either one. I didn’t think that just because I didn’t believe in God, it meant I believed I was “free.” I was sure that free will didn’t exist, but I also didn’t believe that everything was determined in the sense that some omnipotent being knew about it.

I tried to dissect the forces behind my actions and decisions, and I came to the conclusion that every move I was making aligned with what I believed to be the best possible move for me in that moment, correct or not. If in that moment, I chose to do something I knew I’d regret later – I rationalized that my conscious thoughts must have been telling me that the good now would outweigh the bad later.

To imagine a conscious entity doing something they didn’t believe was the best thing to do didn’t make sense to me – in fact, I was sure that conscious action could only be motivated by the belief that it would maximize my satisfaction in that particular moment. What I think I was getting at was that action is supported by a sense of agency and motivation toward reward.

You’ve talked a lot about action, but what about the simple (or not so simple) experience of qualia? How do you explain how qualia can arise from physical processes?

I don’t think it’s so amazing that conscious experience arises from physical processes. I think the only reason we find it amazing is because we know it to be so functional. Take the analogy of electricity; it seems amazing that something as magical as electricity can arise from a changing magnetic field. But why is it so amazing? Probably because electricity is something we know to be particularly functional, that is, it already holds this special, functional status in our minds and so it seems quite miraculous that it could arise from physical processes. Perhaps the case of consciousness is similar: we find it amazing that it can arise from physical processes, but only because of all the functions we (perhaps implicitly) know it provides for us. This suggests that maybe we should ask ourselves: what are those functions that make consciousness so important? Why is it so amazing that conscious experience arises from physical processes? Rather, why shouldn’t it?

What do you think the link is between language and consciousness? Do you think consciousness requires the ability to report?

It’s true that language is often conscious. Although we’ve all had our automatic conversations (“Yeah. Oh, really? That’s terrible. Yeah. Mmhmm…”) we know for the most part that language is supported by consciousness because we use it to communicate with others. And communication with others is generally a very conscious process. But that doesn’t make it the basis for consciousness. Take the case of Genie, who had virtually no language abilities as an adolescent. I think very few people would argue that she wasn’t conscious just because she had very limited language capacity.

You say communication with others is a conscious process. What do you mean by that?

I mean that communication is an active process of sharing an internally conscious idea with an external, but also conscious being. Could empathy exist without consciousness? Probably not. We can’t empathize with something unconscious (e.g. a tree) but we can empathize with a conscious being, even more so, a conscious being whose consciousness we can “relate to,” whatever that means. It seems like a lot of what we mean when we talk about empathy and communication is tangled up in our ideas about consciousness.

What do you think is the best way to go about studying consciousness? How do we satisfy the broad, currently unanswerable question of what consciousness is?

I know I just spent a long time focusing on the philosophy behind consciousness – but I truly believe we have to just keep getting more data. We can discuss and philosophize for hours, and probably come up with some really great theories, but in the end we just need more data. We should always keep coming back to the big questions and refining them, but it’s really a reciprocal process, where questions lead to data collection, and the subsequent results will inform our future question formulation.

In particular, I think the best place to focus our attention (no pun intended) is on understanding all the systems that direct and control attention. That in itself is a pretty big request – but that’s why we have an entire field of cognitive scientists who can approach it from all angles.

It’s a funny thing.

We’ve got brain regions galore that respond to input about humans in particular. Regions for processing human faces. Regions for processing human body structure. Regions for processing human body motion. Regions for processing human vocal noises.

We are creatures designed to empathize. With our own kind, of course.

This is weird. Think about it. We probably don’t have a special place in your brain for processing houses, clouds, or coat hangers. But humans are a different story.

Super Fun Thought Experiment: Get In Touch With Your Fusiform Face Area.

You see a duck. The duck waddles away, and then another duck shows up. Or is it the same one? Who knows? They all look the same, right?

Now imagine the same scenario – but instead of ducks, people. (Of similar ethnic background to you, in particular… but that’s another story.) If you see two people, are you really going to have any trouble telling them apart, assuming they aren’t identical twins?

Of course you aren’t. People are just so distinct, it seems. That subjective feeling of seeming “distinctness” of faces is because your brain has an entirely different way of processing faces. It has nothing to do with human faces actually being any more distinct than duck faces or houses or clouds. It has to do with the way your brain responds. Your mind is your reality – so as far as you’re concerned, humans are distinct and everything else just blurs together.

Now, to be fair, apparently this region responds not just to faces, but also to general areas of expertise. So if you know a ton about cars, this region might be activated when you see pictures of cars.

Fine. Some people are car experts.

But we’re ALL human experts.

That’s pretty cool. And it makes sense evolutionarily, right? Humans are what we care about. Humans are the entities with which we have to cooperate, compete, and coexist. Humans are the entities in which we ultimately find companionship. Sure, if we work hard at it every day, we can become experts at pretty much anything. But to become a human expert – you don’t even have to try. You just have to be alive and in the social world.

Social pressure is unavoidable. And by “pressure” I don’t mean it in the “peer pressure” sense. I mean that we have no choice but to live in a world full of complex social structure, patterns, and forces. These forces aren’t good, bad, or evil. They just are, and they define human experience. And these forces are what drive us to become “human experts” from the day we’re born.

So maybe we aren’t designed to empathize. Maybe we’re forced to.

I’m not. I’m not particularly passionate about linguistics. I’m not particularly passionate about visual cognition. I’m not particularly passionate about neuroscience or artificial intelligence or music cognition.

Okay, maybe I am.

But what I really live for – is discovering any thread that ties them all together. I live for the times when I stumble across some abstract concept that has meaning in every discipline. Those are the times where I feel like I am intimately acquainted with the human mind. Social cognition, music cognition, linguistics, neuroscience, learning, visual cognition – in that moment, they are all the same. They all fit perfectly into this all-encompassing, big-picture framework.

So, what’s tonight’s all-encompassing big-picture concept?

Expectations.

Our thoughts are centered around them. We live to predict. Or we predict to live, really. Every subfield of cognitive science runs into this concept. Linguistics, music cognition, visual cognition, and social cognition alike.

Let me talk about linguistics for a second. I promise to keep it accessible.

The basics: Say the consonant /n/. It’s articulated by your tongue touching that hard ridge behind your teeth, the alveolar ridge. Now say the consonants /b/ and /p/. They’re articulated with both lips.

I say the words “green boat.” The sounds change a little from what you’d expect. The /n/ becomes more like an /m/ – as if I’m saying “greem boat.” Why? Because there’s a /b/ after the /n/ – greeN Boat. Our articulators are lazy and sloppy and like to make everything as easy as possible to say.

Greem boat. Greeng car. Greem bear. Greeng cat. Yes?

Now here’s where expectations come into play: when I say “greem” – BEFORE I even say “boat” or “pear” – you’ve already predicted it. You know greem isn’t a word. So in that instant – that ten millisecond period of /m/ – you’ve decided that the /m/ you hear is really an /n/ AND that there’s some consonant articulated with the lips – probably /b/ or /p/ – looming in the near future.

Just a few milliseconds down the road, actually.

Why can’t you just wait? It’s only a few milliseconds away. Sure, you could wait until you actually heard the upcoming /p/ or /b/ before making any assumptions. But here’s the problem: there are zillions more consonants and vowels that’ll be zooming past you at the speed of light. You can’t afford to waste any time. You have to be efficient. Predict, predict, predict.

Speech perception is all about prediction. Moment by moment, millisecond by millisecond.

Guess what? So is physical motion perception. Social perception. Music perception. They’re all rooted in moment by moment fluctuations of expectations, far below the surface of conscious awareness.

I love this. So much.

And I love that it’s all the same, even across such different domains.

Cognitive scientists are the polyamorists of academia. We’ve all got about seven love affairs going on. But we’re not that interested in any of our lovers. We’re just fascinated by their similarities.

Just imagine this: You have to find an article – just one article – for a class. You’re searching the online archives of its journal, and you can’t help but see all the other titles of the articles in that issue. Before you know it, there are twenty-five new tabs open in your browser window. Why? Because, well, there are twenty-five papers that you just HAVE to read. NOW.

Sometimes I call it Academic Mania. It happens.

You know what else happens? Needing desperately to gush about the amazingness of science.