What is the best way to understand consciousness? In philosophy, centuries-old debates continue to rage over whether the Universe is divided, following René Descartes, into ‘mind stuff’ and ‘matter stuff’. But the rise of modern neuroscience has seen a more pragmatic approach gain ground: an approach that is guided by philosophy but doesn’t rely on philosophical research to provide the answers. Its key is to recognise that explaining why consciousness exists at all is not necessary in order to make progress in revealing its material basis – to start building explanatory bridges from the subjective and phenomenal to the objective and measurable.

In my work at the Sackler Centre for Consciousness Science at the University of Sussex in Brighton, I collaborate with cognitive scientists, neuroscientists, psychiatrists, brain imagers, virtual reality wizards and mathematicians – and philosophers too – trying to do just this. And together with other laboratories, we are gaining exciting new insights into consciousness – insights that are making real differences in medicine, and that in turn raise new intellectual and ethical challenges. In my own research, a new picture is taking shape in which conscious experience is seen as deeply grounded in how brains and bodies work together to maintain physiological integrity – to stay alive. In this story, we are conscious ‘beast-machines’, and I hope to show you why.

Let’s begin with David Chalmers’s influential distinction, inherited from Descartes, between the ‘easy problem’ and the ‘hard problem’. The ‘easy problem’ is to understand how the brain (and body) gives rise to perception, cognition, learning and behaviour. The ‘hard’ problem is to understand why and how any of this should be associated with consciousness at all: why aren’t we just robots, or philosophical zombies, without any inner universe? It’s tempting to think that solving the easy problem (whatever this might mean) would get us nowhere in solving the hard problem, leaving the brain basis of consciousness a total mystery.

But there is an alternative, which I like to call the real problem: how to account for the various properties of consciousness in terms of biological mechanisms; without pretending it doesn’t exist (easy problem) and without worrying too much about explaining its existence in the first place (hard problem). (People familiar with ‘neurophenomenology’ will see some similarities with this way of putting things – but there are differences too, as we will see.)

There are some historical parallels for this approach, for example in the study of life. Once, biochemists doubted that biological mechanisms could ever explain the property of being alive. Today, although our understanding remains incomplete, this initial sense of mystery has largely dissolved. Biologists have simply gotten on with the business of explaining the various properties of living systems in terms of underlying mechanisms: metabolism, homeostasis, reproduction and so on. An important lesson here is that life is not ‘one thing’ – rather, it has many potentially separable aspects.

In the same way, tackling the real problem of consciousness depends on distinguishing different aspects of consciousness, and mapping their phenomenological properties (subjective first-person descriptions of what conscious experiences are like) onto underlying biological mechanisms (objective third-person descriptions). A good starting point is to distinguish between conscious level, conscious content, and conscious self. Conscious level has to do with being conscious at all – the difference between being in a dreamless sleep (or under general anaesthesia) and being vividly awake and aware. Conscious contents are what populate your conscious experiences when you are conscious – the sights, sounds, smells, emotions, thoughts and beliefs that make up your inner universe. And among these conscious contents is the specific experience of being you. This is conscious self, and is probably the aspect of consciousness that we cling to most tightly.

What are the fundamental brain mechanisms that underlie our ability to be conscious at all? Importantly, conscious level is not the same as wakefulness. When you dream, you have conscious experiences even though you’re asleep. And in some pathological cases, such as the vegetative state (sometimes called ‘wakeful unawareness’), you can be altogether without consciousness, but still go through cycles of sleep and waking.

So what underlies being conscious specifically, as opposed to just being awake? We know it’s not just the number of neurons involved. The cerebellum (the so-called ‘little brain’ hanging off the back of the cortex) has about four times as many neurons as the rest of the brain, but seems barely involved in maintaining conscious level. It’s not even the overall level of neural activity – your brain is almost as active during dreamless sleep as it is during conscious wakefulness. Rather, consciousness seems to depend on how different parts of the brain speak to each other, in specific ways.

A series of studies by the neuroscientist Marcello Massimini at the University of Milan provides powerful evidence for this view. In these studies, the brain is stimulated by brief pulses of energy – using a technique called transcranial magnetic stimulation (TMS) – and its electrical ‘echoes’ are recorded using EEG. In dreamless sleep and general anaesthesia, these echoes are very simple, like the waves generated by throwing a stone into still water. But during conscious states, a typical echo ranges widely over the cortical surface, disappearing and reappearing in complex patterns. Excitingly, we can now quantify the complexity of these echoes by working out how compressible they are, similar to how simple algorithms compress digital photos into JPEG files. The ability to do this represents a first step towards a ‘consciousness-meter’ that is both practically useful and theoretically motivated.

Complexity measures of consciousness have already been used to track changing levels of awareness across states of sleep and anaesthesia. They can even be used to check for any persistence of consciousness following brain injury, where diagnoses based on a patient’s behaviour are sometimes misleading. At the Sackler Centre, we are working to improve the practicality of these measures by computing ‘brain complexity’ on the basis of spontaneous neural activity – the brain’s ongoing ‘echo’ – without the need for brain stimulation. The promise is that the ability to measure consciousness, to quantify its comings and goings, will transform our scientific understanding in the same way that our physical understanding of heat (as average molecular kinetic energy) depended on the development, in the 18th century, of the first reliable thermometers. Lord Kelvin put it this way: ‘In physical science the first essential step in the direction of learning any subject is to find principles of numerical reckoning and practicable methods for measuring some quality connected with it.’ More simply: ‘To measure is to know.’

But what is the ‘quality’ that brain-complexity measures are measuring? This is where new theoretical ideas about consciousness come into play. These start in the late 1990s, when Gerald Edelman (my former mentor at the Neurosciences Institute in San Diego) and Giulio Tononi – now at the University of Wisconsin in Madison – argued that conscious experiences were unique in being simultaneously highly ‘informative’ and highly ‘integrated’.

the maths that captures the co-existence of information and integration maps onto the emerging measures of brain complexity

Consciousness is informative in the sense that every experience is different from every other experience you have ever had, or ever could have. Looking past the desk in front of me through the window beyond, I have never before experienced precisely this configuration of coffee cups, computers and clouds – an experience that is even more distinctive when combined with all the other perceptions, emotions and thoughts simultaneously present. Every conscious experience involves a very large reduction of uncertainty – at any time, we have one experience out of vastly many possible experiences – and reduction of uncertainty is what mathematically we mean by ‘information’.

Consciousness is integrated in the sense that every conscious experience appears as a unified scene. We do not experience colours separately from their shapes, nor objects independently of their background. The many different elements of my conscious experience right now – computers and coffee cups, as well as the gentle sounds of Bach and my worries about what to write next – seem tied together in a deep way, as aspects of a single encompassing state of consciousness.

It turns out that the maths that captures this co-existence of information and integration maps onto the emerging measures of brain complexity I described above. This is no accident – it is an application of the ‘real problem’ strategy. We’re taking a description of consciousness at the level of subjective experience, and mapping it to objective descriptions of brain mechanisms.

Some researchers take these ideas much further, to grapple with the hard problem itself. Tononi, who pioneered this approach, argues that consciousness simply is integrated information. This is an intriguing and powerful proposal, but it comes at the cost of admitting that consciousness could be present everywhere and in everything, a philosophical view known as panpsychism. The additional mathematical contortions needed also mean that, in practice, integrated information becomes impossible to measure for any real complex system. This is an instructive example of how targeting the hard problem, rather than the real problem, can slow down or even stop experimental progress.

When we are conscious, we are conscious of something. What in the brain determines the contents of consciousness? The standard approach to this question has been to look for so-called ‘neural correlates of consciousness’ (NCCs). In the 1990s, Francis Crick and Christof Koch defined an NCC as ‘the minimal set of neuronal events and mechanisms jointly sufficient for a specific conscious percept’. This definition has served very well over the past quarter century because it leads directly to experiments. We can compare conscious perception with unconscious perception and look for the difference in brain activity, using (for example) EEG and functional MRI. There are many ways of doing this. One of the most popular is binocular rivalry, in which different images are presented to each eye so that conscious perception flips from one to the other (while sensory input remains constant). Another is masking, in which a briefly flashed image is rapidly followed by a meaningless pattern. Here, whether the first image is consciously perceived depends on the delay between the image and the mask.

Experiments such as these have identified brain regions that are consistently associated with conscious perception, independently of whether that perception is visual, auditory or in some other sensory modality. The most recent chapter in this story involves experiments that try to distinguish between those brain regions involved in reporting about a conscious percept (eg, saying: ‘I see a face!’) from those involved in generating the conscious percept itself. But as powerful as these experiments are, they do not really address the ‘real’ problem of consciousness. To say that a posterior cortical ‘hot-spot’ (for instance) is reliably activated during conscious perception does not explain why activity in that region should be associated with consciousness. For this, we need a general theory of perception that describes what brains do, not just where they do it.

In the 19th century, the German polymath Hermann von Helmholtz proposed that the brain is a prediction machine, and that what we see, hear and feel are nothing more than the brain’s best guesses about the causes of its sensory inputs. Think of it like this. The brain is locked inside a bony skull. All it receives are ambiguous and noisy sensory signals that are only indirectly related to objects in the world. Perception must therefore be a process of inference, in which indeterminate sensory signals are combined with prior expectations or ‘beliefs’ about the way the world is, to form the brain’s optimal hypotheses of the causes of these sensory signals – of coffee cups, computers and clouds. What we see is the brain’s ‘best guess’ of what’s out there.

It’s easy to find examples of predictive perception both in the lab and in everyday life. Walking out on a foggy morning, if we expect to meet a friend at a bus stop, we might perceive her to be there, until closer inspection reveals a stranger. We can also hear words in nonsensical streams of noise, if we are expecting these words (play ‘Stairway to Heaven’ backwards and you can hear satanic poetry). Even very basic elements of perception are shaped by unconscious beliefs encoded in our visual systems. Our brains have evolved to assume (believe) that light comes from above, which influences the way we perceive shapes in shadow.

people consciously see what they expect, rather than what violates their expectations

The classical view of perception is that the brain processes sensory information in a bottom-up or ‘outside-in’ direction: sensory signals enter through receptors (for example, the retina) and then progress deeper into the brain, with each stage recruiting increasingly sophisticated and abstract processing. In this view, the perceptual ‘heavy-lifting’ is done by these bottom-up connections. The Helmholtzian view inverts this framework, proposing that signals flowing into the brain from the outside world convey only prediction errors – the differences between what the brain expects and what it receives. Perceptual content is carried by perceptual predictions flowing in the opposite (top-down) direction, from deep inside the brain out towards the sensory surfaces. Perception involves the minimisation of prediction error simultaneously across many levels of processing within the brain’s sensory systems, by continuously updating the brain’s predictions. In this view, which is often called ‘predictive coding’ or ‘predictive processing’, perception is a controlled hallucination, in which the brain’s hypotheses are continually reined in by sensory signals arriving from the world and the body. ‘A fantasy that coincides with reality,’ as the psychologist Chris Frith eloquently put it in Making Up the Mind (2007).

Armed with this theory of perception, we can return to consciousness. Now, instead of asking which brain regions correlate with conscious (versus unconscious) perception, we can ask: which aspects of predictive perception go along with consciousness? A number of experiments are now indicating that consciousness depends more on perceptual predictions, than on prediction errors. In 2001, Alvaro Pascual-Leone and Vincent Walsh at Harvard Medical School asked people to report the perceived direction of movement of clouds of drifting dots (so-called ‘random dot kinematograms’). They used TMS to specifically interrupt top-down signalling across the visual cortex, and they found that this abolished conscious perception of the motion, even though bottom-up signals were left intact.

More recently, in my lab, we’ve been probing the predictive mechanisms of conscious perception in more detail. In several experiments – using variants of the binocular rivalry method mentioned earlier – we’ve found that people consciously see what they expect, rather than what violates their expectations. We’ve also discovered that the brain imposes its perceptual predictions at preferred points (or phases) within the so-called ‘alpha rhythm’, which is an oscillation in the EEG signal at about 10 Hz that is especially prominent over the visual areas of the brain. This is exciting because it gives us a glimpse of how the brain might actually implement something like predictive perception, and because it sheds new light on a well-known phenomenon of brain activity, the alpha rhythm, whose function so far has remained elusive.

Predictive processing can also help us understand unusual forms of visual experience, such as the hallucinations that can accompany psychosis or psychedelic trips. The basic idea is that hallucinations occur when the brain pays too little attention to incoming sensory signals, so that perception becomes unusually dominated by the brain’s prior expectations. Different sorts of hallucination – from simple geometric experiences of lines, patterns and textures to rich hallucinatory narratives full of objects and people – can be explained by the brain’s over-eagerness to confirm its predictions at different levels in the cortical hierarchy. This research has significant clinical promise since it gets at the mechanisms that underlie the symptoms of psychiatric conditions, in much the same way that antibiotics tackle the causes of infection while painkillers do not.

Of the many distinctive experiences within our inner universes, one is very special. This is the experience of being you. It’s tempting to take experiences of selfhood for granted, since they always seem to be present, and we usually feel a sense of continuity in our subjective existence (except, of course, when emerging from general anaesthesia). But just as consciousness is not just one thing, conscious selfhood is also best understood as a complex construction generated by the brain.

There is the bodily self, which is the experience of being a body and of having a particular body. There is the perspectival self, which is the experience of perceiving the world from a particular first-person point of view. The volitional self involves experiences of intention and of agency – of urges to do this or that, and of being the causes of things that happen. At higher levels, we encounter narrative and social selves. The narrative self is where the ‘I’ comes in, as the experience of being a continuous and distinctive person over time, built from a rich set of autobiographical memories. And the social self is that aspect of self-experience that is refracted through the perceived minds of others, shaped by our unique social milieu.

In daily life, it can be hard to differentiate these dimensions of selfhood. We move through the world as seemingly unified wholes, our experience of bodily self seamlessly integrated with our memories from the past, and with our experiences of volition and agency. But introspection can be a poor guide. Many experiments and neuropsychological case studies tell a different story, one in which the brain actively and continuously generates and coordinates these diverse aspects of self-experience.

Our experiences of being and having a body are ‘controlled hallucinations’ of a very distinctive kind

Let’s take the example of bodily selfhood. In the famous ‘rubber-hand illusion’, I ask you to focus your attention on a fake hand while your real hand is kept out of sight. If I then simultaneously stroke your real hand and the fake hand with a soft paintbrush, you may develop the uncanny feeling that the fake hand is now, somehow, part of your body. This reveals a surprising flexibility in how we experience ‘owning’ our bodies and raises a question: how does the brain decide which parts of the world are its body, and which aren’t?

To answer this, we can appeal to the same process that underlies other forms of perception. The brain makes its ‘best guess’, based on its prior beliefs or expectations, and the available sensory data. In this case, the relevant sensory data include signals specific to the body, as well as the classic senses such as vision and touch. These bodily senses include proprioception, which signals the body’s configuration in space, and interoception, which involves a raft of inputs that convey information from inside the body, such as blood pressure, gastric tension, heartbeat and so on. The experience of embodied selfhood depends on predictions about body-related causes of sensory signals across interoceptive and proprioceptive channels, as well as across the classic senses. Our experiences of being and having a body are ‘controlled hallucinations’ of a very distinctive kind.

Research in our lab is supporting this idea. In one experiment, we used so-called augmented reality to develop a new version of the rubber-hand illusion, designed to examine the effects of interoceptive signals on body ownership. Participants viewed their surroundings through a head-mounted display, focusing on a virtual reality version of their hand, which appeared in front of them. This virtual hand was programmed to flash gently red, either in time or out of time with their heartbeat. We predicted that people would experience a greater sense of identity with the virtual hand when it was pulsing synchronously with their heartbeat, and this is just what we found. Other laboratories are finding that similar principles apply to other aspects of conscious self. For example, we experience agency over events when incoming sensory data match the predicted consequences of actions – and breakdowns in experienced agency, which can happen in conditions such as schizophrenia – can be traced to abnormalities in this predictive process.

These findings take us all the way back to Descartes. Instead of ‘I think therefore I am’ we can say: ‘I predict (myself) therefore I am.’ The specific experience of being you (or me) is nothing more than the brain’s best guess of the causes of self-related sensory signals.

There is a final twist to this story. Predictive models are good not only for figuring out the causes of sensory signals, they also allow the brain to control or regulate these causes, by changing sensory data to conform to existing predictions (this is sometimes called ‘active inference’). When it comes to the self, especially its deeply embodied aspects, effective regulation is arguably more important than accurate perception. As long as our heartbeat, blood pressure and other physiological quantities remain within viable bounds, it might not matter if we lack detailed perceptual representations. This might have something to do with the distinctive character of experiences of ‘being a body’, in comparison with experiences of objects in the world – or of the body as an object.

And this returns us one last time to Descartes. In dissociating mind from body, he argued that non-human animals were nothing more than ‘beast machines’ without any inner universe. In his view, basic processes of physiological regulation had little or nothing to do with mind or consciousness. I’ve come to think the opposite. It now seems to me that fundamental aspects of our experiences of conscious selfhood might depend on control-oriented predictive perception of our messy physiology, of our animal blood and guts. We are conscious selves because we too are beast machines – self-sustaining flesh-bags that care about their own persistence.