For centuries, philosophers have theorised about the mind-body question, debating the relationship between the physical matter of the brain and the conscious mental activity it somehow creates. Even with advances in neuroscience and brain imaging techniques, large parts of that fundamental relationship remain stubbornly mysterious. It was with good reason that, in 1995, the cognitive scientist David Chalmers coined the term “the hard problem” to describe the question of exactly how our brains conjure subjective conscious experience. Some philosophers continue to insist that mind is inherently distinct from matter. Advances in understanding how the brain functions undermine those ideas of dualism, however.

Anil Seth, professor of cognitive and computational neuroscience at the University of Sussex, is at the leading edge of that latter research. His Ted talk on consciousness has been viewed more than 11m times. His new book, Being You, proposes an idea of the human mind as a “highly evolved prediction machine”, rooted in the functions of the body and “constantly hallucinating the world and the self” to create reality.

One of the things that I liked about your approach in the book was the way that many of the phenomena you investigate arise out of your experience. For example, the feeling of returning to consciousness after anaesthesia or how your mother, experiencing delirium, was no longer recognisably herself. Do you think it’s always important to keep that real-world framework in mind?
The reason I’m interested in consciousness is intrinsically personal. I want to understand myself and, by extension, others. But I’m also super-interested for example in developing statistical models and mathematical methods for characterising things such as emergence [behaviour of the mind as a whole that exceeds the capability of its individual parts] and there is no personal component in that.

You’ve set up your team at Sussex as a multidisciplinary group, with pure mathematicians, psychologists and computer scientists as well as cognitive neuroscientists. Why is that?
I was wary of academia because my early experience of education was of progressive specialisation. I still remember having to choose between arts and sciences when I was 15 and that seemed nuts. I was worried that having an academic scientific career would consist of learning a huge amount about something that nobody else cared about. It was a massive relief when that turned out not to be true. We try to keep a question in mind and then use different tools to answer that question without worrying about what discipline they are attached to.

What’s the question that you are all keeping in mind?
At its very broadest, it’s the question of how to develop a satisfying scientific explanation of conscious experience.

Presumably, the mind-body problem is never going to be entirely resolved?
No, but I’d like to make progress. It’s the boring answer of continuing to do rigorous science, rather than proposing some eureka solution to “the hard problem” [the question of why and how our brains create subjective, conscious experience]. My approach is that we risk not understanding the central mystery of life by lurching to one or other form of magical thinking. While science might be a little bit slower, there is much to be done in a straightforward materialist understanding of how the brain relates to conscious experience.

I was interested in your section about memory in the book, in particular about Clive Wearing. Wearing is someone who, as a result of a devastating brain infection, lost all conscious memory and lives in a permanent present tense, as if perpetually waking from coma. Yet the studies show that he demonstrates an abiding love for his wife. How is that explained?
I’ve never met Clive or his wife, only read about the case. But it highlights the fact that some of those things we think are necessary for selfhood are obviously not. There are all sorts of different forms of memory. Explicit conscious recall, autobiographical memory, is just one of them. In neurological patients, you often see how the mind is built of processes that in normal life we never see.

I remember the writer Nicholson Baker suggesting that all thoughts worth having are about the size of a wardrobe and have the complexity of a wheelbarrow. How do you think about thoughts?
The philosopher William James said: “Thoughts themselves are the thinkers.” I think that there’s a truth to that. It’s perhaps always a mistake to think of thoughts being produced or observed by a prior internal self. Thought is foundational to psychology, but it’s one of the things that’s hardest to study. You can’t control thought in the same way you can systematically manipulate perception in the lab. So I’ve tended to avoid investigating how the mind wanders and so on.

But in your studies you begin to observe how some kind of playfulness is built into consciousness?
There’s definitely a sort of internal creative spur to the variation of our mental lives. But where do thoughts come from? I’m left a little cold by psychoanalytic explanations, which suggest there’s a subconscious trying to get in there and give you some thought that would otherwise be repressed. I think, to me, they’re the maximally abstract version of perception.

Your book is full of good aphorisms. One pivotal one in your argument about the how and why of consciousness is the idea that “I predict myself, therefore I am”. What is the “I” in that sentence?
It’s a collection of perceptual predictions. It’s a playful sentence. The “I” is deliberately ambiguous there – it says there is an experience arising of me being a single unified individual, with all these different attributes: memories, emotional bonds, experiences of body. For this piece of flesh and blood here, they seem to be unified – at least if I don’t reflect on it too much.

That first-person feeling is very stubborn. Most of us have a very strong sense of continuity between our childhood experiences and our current self. Is that perceived unity essentially a kind of Darwinian strategy?
There’s a lot of argument about the evolutionary function of consciousness. But the answers you get to that depend on what distinction you’re trying to make. If you’re trying to say why is anything conscious at all, rather than just mechanisms evolving in patterns in the dark?, then you’re simply up against the “hard problem” again. But if you reframe it as what is the evolutionary benefit of the organism having these specific experiences?, then you see that an experience of selfhood is clearly important because it maximises the organism’s chances of survival.

Why is it not possible for artificial intelligence to at least mimic that organising perception and therefore mimic other aspects of conscious selfhood?
I do think it’s very likely possible for AI to mimic that. In fact, in the book I talk about the pace of this ability to mimic being really quite scary, with the combination of “deep fake” things and natural language processing machines. Instantiation is another thing, though.

What do you mean by instantiation?
Building an AI system or a robot that does subjectively experience having a self, as opposed to being a sophisticated machine that gives the appearance of having a self but with nothing actually going on.

But if we take Daniel Dennett’s definition of consciousness as a “trillion mindless robots dancing”, where does the difference lie?
Dan Dennett has been one of my longest-standing inspirations and mentors and the chance over the last few years to argue with him has been a great pleasure. I gave a Ted talk in 2017 and of the 3,000 people in the room – lots of founders and investors and famous people – I was only terrified about Dennett, who I knew was in the audience. And rightly so. At one point in the talk, I described perceptual experiences as a kind of “inner movie”. Afterwards, he said: “Ah, that was all great, apart from the movie. Because: who’s watching the movie?” And that’s a very good criticism. There’s no movie because there is no one watching it.

But there is an inbuilt narrative however we describe it – our internal life is all storytelling?
Dennett is a little equivocal about what he thinks perceptual or phenomenal experience is, if it exists or not – whether, once you’ve explained all the functions of the system and its dispositions to behave in particular ways, there’s anything left over still to explain. I’m on board with that, because I think we can get a very long way to explaining the functions and dispositions of things to behave in particular ways. But I’m agnostic about whether at the end of this programme of trying to account in physical terms for properties of experience, there will still be some residue of mystery left, something more to explain.

Have your thoughts on that ever taken any spiritual swerve – in terms of the why of there being something rather than nothing?
It’s more that I think there’s hubris in assuming that everything will submit to a mechanistic programme of explanation. I think it’s intellectual honesty to acknowledge that the existence of conscious experience as a phenomenon in a universe for which we generally have physicalist accounts seems weird. I want to figure out the ways in which we can undermine this seeming weird.