As yet, science can offer no standard, widely agreed definition of consciousness, partly because researchers use the same word to describe a variety of phenomena such as wakefulness, awareness, attention, responsiveness to stimuli and inner life. It is thus a field of broad interdisciplinary study.
There is also significant linguistic and cultural variance in what people understand by the term “consciousness”.1 This is also true of “disorders” of consciousness: the experience of what might be termed delusions or hallucinations by western medicine can be seen as normal and even valuable in many cultures. Responses to the idea of non-human consciousness are also culturally differentiated.
Currently, there is no agreed theory of how consciousness arises.2 However, progress is being made here, particularly by the Cogitate consortium’s innovative and rigorous experimental approach to testing the predictions of leading theories, within a framework of “adversarial collaboration” that uses preregistered experimental tests.3 While no single theory has emerged as a definitive winner, the study established a new standard for theory-driven testing.
The lack of deep theoretical understanding has not prevented the adoption of technologies and conceptual advances that help make decisions in a clinical setting. We have devised methods to quantify the degree of presence or absence of consciousness (as inner life, awareness, mental response to stimuli), allowing us to define whether a patient is in a vegetative state, for instance, and to evaluate whether their consciousness can be “augmented”; that is, shifted to a more elevated state of consciousness — whether by intervention or through the body’s natural healing processes.4
As with many medical applications, technologies that spring from a clinical setting will eventually benefit the broader population. This is because the same technologies that diagnose or modulate consciousness when there is a deficit or disorder can be pressed into service to enhance or augment healthy, functioning consciousness. They will also be helpful in more philosophical areas, improving our understanding of free will, the self, and the ethics of cases at the edges of sentience.
Meanwhile, work is underway to determine the developmental emergence of consciousness in humans,5 and the extent to which machines and animals can have or develop consciousness,6,7 Researchers also aim to understand whether organoids and other synthetic biological organisms are capable of developing a kind of consciousness.8 This is likely to spark debate on the ethics of the field itself, and of assignment of moral status and rights.
Key takeaways
Although central to the experience of being human, consciousness has proved difficult to investigate scientifically, with researchers struggling to reach consensus on many aspects of the field. While it is clear that human consciousness is a multi-faceted phenomenon, its relation to brain states — including those induced by injury, degradation and trauma — is dizzyingly complex. Understanding where, when (and whether) non-human consciousness arises, whether in other animals or in machines, together with the developmental origins of consciousness in humans, is a useful path towards further understanding of the brain and of the wider phenomenon of consciousness. Drugs, electrical stimulation and in some cases cell replacement, genetic and other biotech interventions all show promise for modulation of human consciousness for therapeutic, recreational and knowledge-gathering purposes. Eventually, we may be able to extend the phenomenon to create beyond-human consciousness that augments our evolved abilities in ways that might facilitate an existence lived alongside intelligent machines.
Understanding the difference between all of these various types of human consciousness will be a vital part of the effort to treat injury and disease, and to understand various aspects of human rights issues associated with consciousness. Research efforts are aiming to do this, with some success in the application of diagnostic tools that can differentiate between a vegetative state and a minimally conscious state, for instance.10 Functional MRI and EEG can be used to demonstrate performance of cognitive tasks in some comatose patients who do not follow verbal commands.11 Intervention in coma patients, aiming at recovery, is becoming a realistic, if long-term, prospect.12,13 There is hope that better understanding will help detect and assess early memory loss in ageing and in Alzheimer’s, disease, and guide treatment and care.14At the other end of life the need to assess, consciousness “in the cradle” in the prenatal and infant stages , is becoming an issue with medical, educational , legal and ethical implications.15
Neurosurgical research demonstrates that specific neuronal activity underlies the emergence of concepts, memories, and intentions in human consciousness. Recent work involving direct recordings from the human brain has highlighted the role of specific regions of the thalamus in mediating conscious perception, showing flow of involved information from these thalamic nuclei to the prefrontal cortex.16 Further exploration here, with emerging brain recording and stimulation tools, partnered with AI-based analysis, could help decode conscious and preconscious states.17,18 Interaction with sleeping humans can also help to explore the space between the conscious and unconscious brain.19
The brain remains surprisingly active even when consciousness is absent. Research using intracranial recordings in sleeping patients demonstrated that neurons can continue to modulate their firing rates in response to complex auditory stimuli.20 This suggests that substantial sensory processing survives the loss of conscious awareness. Research on sleep-dependent replay — the reactivation of neural patterns associated with prior waking experiences — suggests that sleep is not merely a passive state but an active period of memory consolidation, critical for our ability to consciously recall past events.21 An important challenge for future work is determining whether targeted stimulation during sleep can enhance recollection of events experienced before sleep and perhaps even influence subsequent learning and cognition.
More recently, recordings obtained with high-density Neuropixels probes during anesthesia revealed that neurons in the hippocampus can respond selectively even to semantic and syntactic features of speech.22 Such work could also help with fundamental science, such as developing theoretical understanding of consciousness by identifying areas of interest for probing the neural correlates of consciousness, and contribute to quantification and classification of consciousness in humans.
Interesting advances are being made in the study of machine consciousness.23 Researchers are making progress in developing methods to assess AI systems for consciousness.24 and investigating machine implementations of neuroscientific theories of consciousness.25 Efforts to understand whether machine consciousness is possible, and what might constitute it, provide an important anchor for our understanding of consciousness.26 This can in turn help anchor our understanding of artificial consciousness, as demonstrated by a published framework for identifying indicators of consciousness.27 The authors propose systematic criteria inspired by neuroscience theories of consciousness that could be used to evaluate artificial systems. Their approach is part of a general effort over recent few years that exploits ways in which neuroscience theories can benefit from study of AI architectures and vice versa.
In animals, the issue of consciousness is more immediate, pertaining to issues of moral status, and with ramifications for agriculture and food economics. A cohort of researchers have suggested there is “strong scientific support for attributions of conscious experience to other mammals and to birds”, and that “empirical evidence indicates at least a realistic possibility of conscious experience in all vertebrates (including reptiles, amphibians, and fishes) and many invertebrates (including, at minimum, cephalopod mollusks, decapod crustaceans, and insects)”.28 This is a controversial claim, but one that highlights the lack of agreement about the nature and extent of consciousness in the natural world. Further exploration of the issue can only improve our understanding of the phenomenon, although funding for this work is limited. There is also a growing awareness of the possibility that some rudimentary form of consciousness could arise in brain organoids, which are proto-organs grown in vitro from human brain tissue.29 However, despite the potential ubiquity of non-human conscious experience, caution is needed due to the possibility that very complex cognitive abilities can unfold without conscious experience.30
A major international analysis found that psychedelics consistently increased communication between transmodal networks—including the default mode, frontoparietal, and limbic systems—and unimodal
sensory and motor networks.32 The analysis also revealed significant changes in subcortical structures, including the thalamus, caudate, putamen, and cerebellum. These findings suggest that psychedelic states are characterised not simply by cortical disinhibition but by widespread reconfiguration of interactions between cortical and subcortical circuits, providing new insights into the neural mechanisms underlying conscious experience.33
Increasingly, experience with chemical modulations of consciousness are proving useful in medical contexts. Psilocybin and MDMA are being used to treat post-traumatic stress disorder and depression, for example.34,35
Electrical modulation of consciousness is also seeing success in invasive and non-invasive forms. Brain-computer interfaces are developing at pace.36 More precisely targeted non-invasive technology like tDCS has successfully brought minimally conscious patients back towards higher consciousness.37 Deep brain stimulation is showing therapeutic promise for treating disorders of consciousness associated with impaired arousal or awareness.38 “Closed-loop” neuromodulation is able to act on, and sense, brain state based on feedback.39,40 There is also growing scope for editing of memory and/or urges and wishes through electrical implants, as the technological tools required are already available.41
Stem cells provide another approach to neuromodulation. Rat stem cells, for instance, have restored sensory abilities within a mouse brain that had its own olfactory neurons rendered inert.42 It seems that stem cell injection can modify cellular composition of neural circuits, potentially altering conscious experience.
Newly developed brain-to-brain interfaces — currently possible invasively in animals46 and noninvasively in humans47 — may allow for direct communication between two brains without involving the peripheral nervous system, instantiating intelligence and perhaps consciousness that has a distributed or collective aspect.48 In addition, two-person fMRI scanning and hyperscanning49 is opening up new possibilities to investigate possibilities of shared experience leading to some form of shared consciousness.50
Experiments in robotics are indicating that robots might learn the ability to think in abstract terms, planning and estimating the outcomes of prospective actions without the need to physically execute them. This may have application in averting workplace accidents and recovering from damage.51 Via brain-computer interfaces, robot self-simulation may also open up pathways for humans to augment their experience. There is scope for human-machine hybrid “teams” to improve industrial decision-making and other tasks.52
