Imaging can be used to measure the degree of variation and complexity across brain regions that correspond to conscious states.Credit: AJ Photo/Hop Americain/SPLAt first glance, human consciousness seems like the one thing we should never have to doubt. We know we are conscious because we experience the world in vivid detail: the small luxury of a piece of chocolate; a song imbued with sadness; or a childhood memory that takes us back to another place and time.But for such a fundamental aspect of reality, what underlies consciousness — the experience of being aware of one’s own existence — has proved deceptively hard to pin down. Researchers and philosophers have spent centuries debating how and why physical processes in the brain result in the subjective, first-person experiences we all seem to have. This issue, known as the ‘hard problem’ of consciousness, continues to drive research today.Nature Spotlight: NeuroscienceScientists are still far from explaining how brain activity begets an entire inner world, but the field is making progress. The advances that researchers have made include the ability to reliably measure brain activity and connectedness and probe for signs of hidden awareness.The emerging picture — based on insights from not just neuroscience, but also far-flung disciplines such as plant science, artificial intelligence and astrobiology — is not of consciousness as a single light that can be switched on, but as a coordinated condition that is potentially more widespread than previously thought. The rapid rise of AI, for example, has fuelled discussions over whether only humans can be conscious — and of whether a brain is even required for consciousness.“People are often troubled by the fact that our understanding remains somewhat vague,” says Matthias Michel, a philosopher of cognitive science at the Massachusetts Institute of Technology in Cambridge. “However, it’s very important to note that definitions in the history of science come at the end of the scientific process, not at the beginning. I’d argue we’re now in the thick of it, which is an exciting place to be.”The seed of consciousnessThere is no direct test for evaluating human consciousness, but researchers have made advances in establishing what are known as neural correlates of consciousness (NCCs). These criteria, which can be a single metric or a combination of events, identify the minimum neuronal mechanism required to kickstart a conscious experience. Specific neurons might need to fire, for example, and particular brain activity patterns could signify an internal experience. Below these thresholds, a person might move through life not fully conscious of their behaviour — zoning out while driving, for instance, only to snap back to attention moments later.Some NCCs have been determined by studying people in states we don’t typically associate with consciousness, such as those who are asleep, anaesthetized or in a coma. According to Melanie Boly, a neurologist at the University of Wisconsin–Madison, this type of research is crucial to help separate true NCCs from activities that merely signify being awake and responsive to our surroundings.“That’s been a big breakthrough, isolating the true features of consciousness, rather than simply measuring something that is associated with or a consequence of the conscious experience,” Boly says.Is social media changing children’s brains? Here’s what the science saysThrough that work, scientists know that about one-quarter of individuals who present as medically unconscious are, in fact, experiencing a type of covert consciousness in which they continue to perceive the world around them. In a landmark 2024 study, a subset of 241 people who had been left seemingly unresponsive by brain injuries were asked to imagine themselves engaging in activities such as playing tennis or opening and closing their hand. Dozens of patients’ brains became active on imaging scans during these challenges — calling into question whether consciousness persists in otherwise-inert states1.In recognition of this finding, researchers have moved away from thinking of consciousness as a binary experience towards an understanding that consciousness is a dynamic process that exists on a continuum. From these principles, it has been possible to develop metrics that test the underlying scientific assumptions about consciousness. One of the most promising of these is called the perturbational complexity index (PCI)2.PCI evaluates a person’s level of consciousness by assessing how their brain responds to a brief magnetic pulse, as measured by electroencephalography (EEG), which traces the brain’s electrical activity using electrodes. In a conscious brain, the neural response usually spreads while remaining complex and distinct from one brain region to another — features referred to as integration and differentiation, respectively. During sleep or under anaesthesia, the response becomes simpler and more repetitive.“PCI is considered to be a very accurate correlate of consciousness that can be useful in many research settings,” says Johan Frederik Storm, a neurophysiologist at the University of Oslo. Widespread clinical adoption has unfortunately been hampered by the fact that it’s a laborious metric to gather, he adds, “but even still, it is gaining traction”.Stress-testing theoriesThis growing understanding has led to the development of hypotheses that move beyond simply measuring aspects of consciousness towards a more foundational understanding of what it is. At present, there are more than 450 such theories, many of which share overlapping features even as they distinguish themselves by scale or focus. The two most popular are the global neuronal workspace theory (GNWT) and the integrated information theory (IIT).The GNWT posits that most brain activity is restricted to specialized areas — visual regions process shapes and colours, for example, and language areas process words. Consciousness emerges when information becomes important enough to be broadcast across regions. A face whose representation remains inside the visual system might trigger a simple, unconscious reaction, but a face that triggers regions across the brain, collectively known as the global workspace, can be used by many systems to build a conscious experience.The IIT tries to answer a different question: what is experience like? When we view the world or recall a memory, we experience a cohesive scene, not separate elements of colour, sound and touch. The IIT argues that consciousness depends on how strongly the parts of a system affect one another while still remaining distinct — there must be both integration and differentiation to give rise to diverse states and experiences. A conscious system must be more than a collection of independent parts, and it must form a whole that has a state that cannot be fully explained by looking at each part separately.A major issue with proving such hypotheses is that researchers have often tested them in ways that lack true neutrality and rigour. A 2022 analysis of hundreds of experiments found that research into consciousness has often been done in ways that pre-emptively bias the results in support of a particular theory3. This has made it a challenge to directly compare studies of consciousness, according to Christof Koch, a consciousness researcher at the Allen Institute for Brain Science in Seattle, Washington, who has studied the topic for decades.“People become very attached to their theories, and that confirmation bias can sometimes get in the way of an honest, collaborative effort,” he says.Lucia Melloni, a cognitive neuroscientist at the Ruhr University Bochum in Germany, recently led a first-of-its-kind effort to force a direct comparison of the two schools of thought. She brought together proponents of both theories, as well as neutral participants, and asked them to agree in advance to a set of experiments and predictions. The group probed foundational assumptions of both theories4 using three measurement techniques — EEG; functional magnetic resonance imaging (fMRI), which tracks brain activity by detecting tiny changes in blood flow; and magnetoencephalography (MEG), which measures magnetic fields produced by the brain’s electrical currents.Plants lack brains, but are capable of decision-making and problem-solving.Credit: Jason Milligan/Courtesy of UnderstoryOne of the main differences between the two ideas is which part of the brain is considered most essential to the act of being. The GNWT predicts a sudden ‘ignition’ of widespread brain activity when a stimulus prompts a conscious experience, and so the prefrontal cortex is expected to play an important part because it uses information to form complex behaviours. ITT focuses instead on a ‘hot zone’ located towards the back of the brain that includes parts of the systems responsible for vision, hearing and touch, because it follows that this is where the relevant stimuli converge to generate a full experience.The team found conscious information represented in both regions. However, the results also challenged important predictions of both theories. The long-lasting activity predicted by the IIT to occur in the back of the cortex was not clearly represented, and the ignition expected by the GNWT was limited. As such, the study did not settle the debate, instead suggesting that both theories might offer parts of an answer.Melloni doesn’t see this as a failure, and considers such results the most exciting outcome. “I’m happy with what we found because neither theory was left untouched,” she says. “Having your predictions confirmed doesn’t leave you anywhere to grow, whereas having your thinking challenged makes you do better science.”How far can consciousness spread?One way that scientists are attempting to expand their thinking is through interdisciplinary collaborations, drawing evidence from fields as distant as plant and non-human animal biology, robotics and even the search for extraterrestrial life to flesh out the conditions under which consciousness can arise. It’s not certain, for instance, that brains — or even neurons — are required. Astrobiologists have long been searching for sentient life, yet given the expanse of the known Universe, they cannot reasonably assume it will look anything like us.