Neuroscientist Anil Seth shares insights from his new book and what to expect from the upcoming CIFAR Neuroscience of Consciousness Winter School
Anil Seth believes the human brain is a prediction engine, shaping our perceptions of the world, and of the self.
Seth, co-director of CIFAR’s Brain, Mind & Consciousness program and a professor at the University of Sussex, explains this theory — the basis behind his new book, Being You: A New Science of Consciousness — and previews what to expect from the CIFAR Neuroscience of Consciousness Winter School.
Your new book, Being You, has been described as not only “deeply profound,” but “radical.” Can you explain the theory behind the book? What makes it radical?
It’s really my chance to write down and pull together the ways in which I’ve been thinking about consciousness over the last 20 years, which have been developed along with many of my colleagues, including many of my colleagues at CIFAR.
The overall approach to understanding consciousness that is described in the book, is not to treat it as one big scary mystery that requires one eureka moment of a solution – finding the “magic dust” – that conjures consciousness from mere mechanism.
To do this would be to address what, in philosophy, people call “the hard problem of consciousness.” How is it possible that any kind of subjective experience can emerge or be identical to any physical mechanism whatsoever? I prefer to approach this challenge indirectly and say, look, consciousness exists, we all have conscious experiences. And conscious experiences have different and describable properties.
There are properties that are shared by all conscious experiences: for example, all experiences are both unified and informative. They also each have distinctive properties. Visual experiences are different from emotional experiences, are different from memories, are different from feelings of ‘free will.’
My approach is to treat all of these different kinds of conscious experiences as perceptions of one sort or another and look for a common mechanism that underlies them and explains their features. I call this the ’real problem’ of consciousness.
In my view, there’s a common mechanism in the brain that underpins both the visual experience of the world around us, and the emotional experience that is part of being a self. This common mechanism is captured by the idea that the brain is a prediction machine, and that everything we experience is a sort of predictive perception – ‘controlled hallucination.’ And then the hard work comes in understanding how and why these shared mechanisms — that the brain is always predicting the causes of its sensory input — shape the conscious contents in the way that they do.
Where we get to at the end of the book is the realization that all of our conscious experiences, whether they’re of being you or of the world around us, are all very, very closely tied to our living bodies — that we perceive the world with, through and because of, our living bodies.
This may or may not count as radical, but I hope it’s at least distinctive.
What experiments and data do you use to underpin your theory that perception is an act of prediction?
It’s a long game, we’re still close to the start. In my group we’ve done some simple experiments over the last few years, where we explicitly manipulate what people expect to see in various situations and ask how that affects what they actually see. These studies are admittedly rather straightforward, and do not constitute anything like a grand proof of any particular theory, but they’re still part of the puzzle.
Where we’re going now, and it’s not just our lab, other labs are doing similar things, is figuring out ways to test the idea that the content of what we perceive is conveyed, not by reading out bottom-up sensory signals that come into the brain from the outside, but built from the top-down or outside-in signals that I think convey the brain’s predictions.
To test hypotheses like this will involve sophisticated combinations of behavioral experiments, where we set up perceptual tasks, with brain imaging, and — critically — with computational models that allow us to build and test explanatory bridges and say, “OK, this is what we would expect to see in the brain-imaging data, if it’s working according to the way we think it’s working.”
What projects are you working on right now?
I’ve recently been very lucky to have been awarded a European Research Council Advanced Grant, which will support my group’s research for the next five years and will allow us to go in a few different directions, including developing quantitative measures of consciousness. This direction builds on work we’ve been doing for quite a while now, some of it in collaboration with CIFAR Fellow Marcello Massimini, on developing measures of ‘complexity’ that we can apply to brain dynamics.
Complexity is rather a vague term. It captures – in some sense – how ‘interesting’ brain dynamics are, and ‘are they completely random, or are they completely ordered? And what’s this interesting space in between?’ We’ve been developing measures of this middle ground for years. And looking ahead, one exciting direction to take this research is in formulating measures of ‘emergence.’
The idea of emergence is that the whole can be more than the sum of the parts. If you look at a flock of birds, the flock seems to have a life of its own that is not just the average of all the positions of the birds – there seems to be an autonomy and independence to the dynamics, the behaviour, of a flock.
People have often applied the concept of emergence to how conscious experiences might ‘emerge’ from brain activity. Sometimes this is done in an unappealing almost spooky sense, as if emergence can magically bridge the explanatory gap between the physical and the mental. But I prefer to see it a different way: conscious experiences are unified and seem to flow with their own autonomy and dynamics, yet they depend on mechanisms that have many parts. In this way, measures of emergence can help explain properties of consciousness in terms of mechanisms – in a ‘real problem’ rather than ‘hard problem’ sense.
I’m also involved with an exciting new art-science project which has the potential to involve millions of people in an adventure about perception and consciousness. The details are still under wraps for now, but I’m really looking forward to being able to say more about that in the months ahead.
The CIFAR Winter School is currently accepting applications until Oct. 7, and you’ll be among the lecturers when the three-day event begins in January. Why should senior PhD students and postdocs apply?
Let me first say that one of the great things about the Winter School is it’s organized by CIFAR Azrieli Global Scholars, rather than by the fellows or by us co-directors. This gives it an energy and freshness that is truly unique. For the 2022 Winter School, above all I’m just really excited that it’s happening. And although it will still be virtual, the organizers and all the CIFAR fellows involved are really keen to make it something special for everyone involved.
There are so many unique features to the Winter School: the opportunities for networking, for collaboration, for one-to-one discussions in small groups with other school attendees, and also with the lecturers. There’s going to be a mixture of lectures, fireside chats, group discussions, and masterclasses on things like science communication.
So, lots of different activities and lots of different ways for participants to meet each other and get to know each other — because that’s really the best thing about the CIFAR Winter School. It builds a community and it’ll build a community that lasts long beyond the days of the Winter School itself.
Is there anything else you’d like to add?
I’d just like to say that being part of the CIFAR community has significantly shaped what I’ve done both in my career over the last few years, and especially in my new book. It’s been a real privilege of mine to be associated with CIFAR and to be able to work with such a brilliant collection of minds and have this unprecedented level of support from the CIFAR organization for what we’re trying to do. It’s been one of the highlights of my career. I’m very happy to be involved with the program, and I’m looking forward to many more years of happy CIFAR Fellowship.