Summary

Overview of the Conversation

In this deep-ranging dialogue, author and journalist Michael Pollan explores the fundamental nature of consciousness, sentience, the concept of the self, and humanity's place in the living world. Prompted by his research into neuroscience, plant biology, psychedelics, philosophy, and Eastern wisdom traditions, Pollan re-examines long-held materialist paradigms. The discussion navigates the distinction between sentient life and conscious experience, the limits of artificial intelligence, the historical split between science and subjectivity, and the emerging need for "consciousness hygiene" in an age of digital distraction.

Core Themes and Key Takeaways

1. Defining Consciousness and Sentience

• Anil Seth’s Insight: The conversation opens with neuroscientist Anil Seth’s quote from Being You: "I open my eyes and a world appears. Nothing could be more ordinary or more miraculous than that fact."

• Thomas Nagel and "What Is It Like to Be a Bat?": Pollan adopts philosopher Thomas Nagel’s classic definition: an organism is conscious if there is "something it is like" to be that creature. A bat experiences the world through echolocation; a toaster or rubber toy experiences nothing.

• Sentience vs. Consciousness: Pollan draws a clear boundary between the two:

• Sentience: The foundational capacity to sense environmental changes with positive or negative valence and act accordingly (moving toward nutrients, away from toxins). Even simple bacteria and plants exhibit sentience.

• Consciousness: The complex human/animal manifestation of sentience, which includes self-awareness ("aware that we are aware"), metacognition, rich emotional processing, and social navigation.

2. The Expansion of Consciousness and the AI Debate

• The Historical and Current View of Animals: René Descartes notoriously viewed animals as unfeeling automata, performing vivisections on dogs while interpreting their cries as mere mechanical noise. Modern science has dramatically reversed this: declarations like the Cambridge Declaration on Consciousness continually expand the recognized circle of conscious entities to include mammals, birds, cephalopods, and potentially insects.

• The Impending "Copernican Moment": Humanity faces a dual pressure: extending empathy to non-human animals while grappling with claims of artificial intelligence achieving consciousness. Humanity must decide whether it feels closer to mortal, vulnerable animals that feel but lack human language, or to machines that speak fluent English but lack biological vulnerability.

• Skepticism Toward AI Consciousness: Pollan argues against current AI becoming genuinely conscious based on two primary points:

• The Flawed Hardware/Software Metaphor: Norbert Wiener warned, "The price of metaphor is eternal vigilance." Computers strictly separate hardware and software. Brains do not; every memory and experience physically rewires and prunes neural circuitry.

• Embodiment and Vulnerability: Drawing from neuroscientists Antonio Damasio and Mark Solms, Pollan asserts that consciousness originates in bodily feelings and homeostatic signals, not abstract computational logic. Simulated feelings are not real feelings. True feelings depend on mortality, physical vulnerability, and pain.

3. The Function and Illusion of "The Self"

• Why Consciousness Evolved: Theories like Bernard Baars' Global Neuronal Workspace Theory suggest consciousness arises when competing unconscious modules in the brain contend for attention. The winning signal is broadcast across the brain so the organism can resolve conflicting homeostatic needs or navigate unpredictable social environments.

• The Phantom Self: Despite the Default Mode Network being tied to autobiographical memory and time travel, there is no localized neural origin for "the self."

• David Hume’s Introspection: In the 1740s, Hume introspected and found perceptions, feelings, and thoughts, but no underlying central "thinker."

• Buddhism and Ephemeral States: Buddhists view the self as a functional illusion. Consciousness can exist entirely without a self, as experienced during meditation, psychedelic states, or the brief 500-millisecond window upon waking in an unfamiliar room.

• Pollan's Hypnosis Experiment: Under hypnosis with Stanford psychiatrist Dr. David Spiegel, Pollan performed a visualization exercise to search for a "thief in the house" of his mind. Instead of finding no self or a unified self, he discovered multiple historical selves (his 13-year-old bar mitzvah self, a 38-year-old father, a 50-year-old academic), highlighting the lack of continuous identity.

4. Plant Intelligence and a "New Animism"

• Psychedelic Insight: Pollan's psilocybin trip in his garden sparked a deep shift: plants appeared lively, conscious, and well-disposed toward him. Following William James' pragmatic philosophy, Pollan tested this subjective insight against botanical science.

• Discoveries in Fringe Botany: "Plant neurobiologists" demonstrate that plants learn, retain memories for up to 28 days, hear, differentiate kin from non-kin, and alter leaf shapes to mimic host plants.

• Anesthesia Sensitivity: Carnivorous plants and climbing beans lose their dynamic behaviors when exposed to human anesthetics (such as xenon gas). Time-lapse footage shows bean plants actively detecting, competing for, and reacting to physical support poles.

• A New Animism: Science is providing empirical grounds for a worldview historically held by traditional cultures and children—that the living world is inherently active, aware, and responsive.

5. The Hard Problem and the Crisis of Scientific Materialism

• The Galilean Legacy: Post-Galileo science deliberately split objective, quantifiable third-person phenomena from subjective interiority (leaving the soul/subjectivity to the church). Because science was explicitly designed for third-person objectivity, it struggles to account for subjective experience ("qualitative subjective feeling" or qualia).

• The Limits of Reductionism: Francis Crick and Christof Koch sought to reduce consciousness to specific neural correlates (such as 20–40 Hz gamma oscillations). When asked why 20 Hz and not 10 or 30 Hz, Koch realized that finding correlates does not explain the mechanism of subjective experience.

• Shift to Alternative Frameworks: Following psychedelic experiences and quantum physics considerations, figures like Koch have questioned strict physicalism, exploring alternative frameworks:

• Panpsychism: The proposal that mind/consciousness is a fundamental feature of all physical matter.

• Idealism: Defended by thinkers like Bernardo Kastrup, idealism posits that consciousness is the primary foundation of reality, and matter is merely an inference within consciousness.

6. Humanities, Solitude, and "Consciousness Hygiene"

• The Wisdom of Art and Literature: Pollan emphasizes that when reductionist science stalls, literature (e.g., Marcel Proust, William James) captures the true, contextualized complexity of human thought—where every idea is shaped by what preceded and follows it.

• The Cave Retreat with Roshi Joan Halifax: Zen priest Roshi Joan sent Pollan into an isolated cave cell at the Upaya Zen Center. Freed from conceptual chatter and social feedback, Pollan experienced a profound softening of the self, coming to appreciate the shift from trying to solve consciousness as a problem to apprehending it as a gift.

• Practicing Consciousness Hygiene: Modern media, algorithms, doom-scrolling, and AI chatbots exploit and pollute human attention and emotional attachments. Pollan outlines practical steps to protect inner life:

• Periodic digital and media fasts (e.g., limiting news consumption drastically).

• Mindfulness meditation to reacquaint oneself with one's unbidden thoughts.

• Engagement with deep literature and art rather than passive scrolling.

• Cultivating a "Don't Know Mind"—embracing wonder, awe, and unknowing over rigid conceptual certainty.

Transcript

Interviewer: I wanted to open with the opening of the book. You have a beautiful quote by the neuroscientist Anil Seth. Can you tell us the quote? Who gave me the title?

Michael Pollan: Yes. I mean, he didn't mean to. I took it from him. Yeah, in his wonderful book Being You—a book about consciousness and specifically about the self, and how this weird construct of consciousness came about—he starts a chapter with a line that just kind of rang in my head: "I open my eyes and a world appears. Nothing could be more ordinary or more miraculous than that fact." The fact that a world appears to us is because we're conscious.

Interviewer: It's such a beautiful line because we're going to get into a lot of stuff. There's going to be a lot of theories, triumphs, failures, and confusion, but that line really encapsulates it: "I open my eyes and a world appears." That is the miraculous aspect you set out to explore on this journey and share with us.

Michael Pollan: Yeah. It's a book about the thing we know best: the fact that we're conscious and the fact that a world appears to us when we open our eyes. It's weird that there's any mystery attached to this, because it's just so ordinary at one level, and then at another level, it's quite extraordinary.

Interviewer: You have a line early in the prologue where you're kind of confessing you're not simply going to deliver us answers, that you couldn't find them.

Michael Pollan: Spoiler alert. Sorry!

Interviewer: And the people that you follow around and get into these engaged discussions with are also in some sense struggling with these questions. But you say something really beautiful in the opening, where you wager that anyone reading it will become more aware and more understanding of—and I'm going to try to quote you here—"the miracle that in this universe of rock and fire and ice and infinite space, we are somehow not only here, but aware." I think that's a really beautiful line. Somebody's going to use it for a book title.

Michael Pollan: Yeah, it's theirs! They're welcome to have it. But yeah, that we're aware, but also that we're aware that we're aware, which may be unique to us. But we don't know for sure.

Interviewer: As with a lot of great journalism and storytelling, there's this "who, what, where, when, why" kind of aspect to this. I thought maybe we'd pick apart some of the things you went through in this journey. It strikes me that the "how"—which doesn't belong both for its lack of alliteration and because it seems to me the toughest of those questions—is key. But let's start with the "what."

There are a lot of terms thrown around by practitioners across many fields—biologists, anthropologists, psychologists, philosophers, novelists—and the terms are quite jumbled. I feel that you tried to be very conscientious to disambiguate things like consciousness, attention, awareness, and sentience. Can you help us along with where you got to with those definitions?

Michael Pollan: There are some disagreements about terms. I think we do know what we mean when we define consciousness simply as subjective experience. There are more elaborate ways to define it. A famous one is by Thomas Nagel, the philosopher who, in the early 1970s, wrote a marvelous essay called "What Is It Like to Be a Bat?"

And he basically said a creature is conscious if there is something it is like to be that creature. We can sort of understand in the case of bats, different as they are from us—they get around not through a visual system, but through echolocation like sonar, and they hang upside down a lot of the day—we can kind of imagine that it is something to navigate the world like that. Whereas for your toaster, there's nothing it's like to be your toaster, or for a molded rubber toy. So, if it feels like anything to be you, you're conscious. I think that's a pretty sturdy definition that most people in the field seem comfortable with.

I make a pretty sharp line between sentience and consciousness, though this might be a little more controversial. The way I think about it is that sentience is the most foundational level of consciousness. It is the ability to sense changes in one's environment. They have a valence: they're felt either as positive or negative, and you move toward the positive and away from the negative. Even bacteria have this. It's a very simple mechanism.

Consciousness is the way we as humans do sentience. Different creatures have different ways of doing it depending on their sensorium, their body type, and their needs. This distinction was important to me because early in the book, I explore plants and ask whether plants are conscious. I'm a lot more comfortable thinking of them as sentient rather than conscious, because our version of consciousness has a lot of bells and whistles that plants or bacteria don't have. We have self-consciousness; we don't only exist, we know we exist. There's a lot more to it in our case, which is necessitated by the kinds of lives we lead. Plants are concerned with other things than a complex social life.

Interviewer: We can ask the plants that are out front here! We have an amoeba depicted here, moving around, and this feels sentient in the formal definition that you're describing. Whether or not it's conscious seems a little controversial. Do you think it is, or not?

Michael Pollan: I don't think we can say for sure, but imputing consciousness to any being is tricky. I mean, I don't know for a fact that you're conscious.

Interviewer: I'm not taking that personally!

Michael Pollan: It doesn't show any behaviors that would suggest it has self-consciousness or self-awareness.

Interviewer: And does it have a sense of self?

Michael Pollan: I don't know about that.

Interviewer: I sometimes imagine we could have these little machines that receive light and respond in a similar way—moving towards or away from the light. And yet, I still suspect that an amoeba has something more going on than that little gadget. Don't you suspect it might have something more?

Michael Pollan: Sure. I don't think it has a notion of a self, and I don't think it's reading Nagel, but I think it's something more. Lynn Margulis, the great microbiologist, watched bacteria and other single-celled creatures under a microscope and became convinced that all life is conscious based on their behavior. So there's an act of imagination involved that's kind of unavoidable.

Tardigrades, for instance, are half a millimeter long, have been around for 500 million years, and will probably survive longer than we will. They can handle extreme heat, freezing, and vacuum space. They don't seem like just cells responding passively.

So that's the "what" of consciousness. Now, the "who" is also interesting. I was shocked to read in your book that René Descartes performed vivisections on live dogs and rabbits because he believed they were not in possession of consciousness.

Michael Pollan: No, he believed humans had a monopoly on consciousness. This idea fixated him so powerfully that he interpreted the screams of these creatures as mere physiological noise that could be ignored. It tells us what human beings are capable of failing to see.

Interviewer: It is incomprehensible. Looking at primate behavior or other mammals, it's unimaginable that you wouldn't be compelled to include them in the category of consciousness. But this category is currently expanding in scientific conversation.

Michael Pollan: There was the Cambridge Declaration on Consciousness issued about 15 years ago, which included mammals, birds, and cephalopods. Ten years later, the same group of animal scientists updated it and said, "Well, we may have to look at insects." The more we look, the more consciousness we find.

It sets up a very interesting moment for our species. At the same time that we're distributing consciousness more generously to animals, we have AI coming along. Many people, especially in Silicon Valley, seem to think AI can become conscious, or even that it already is, and that it will be far more intelligent than we are. Note that intelligence and consciousness are orthogonal; they are not directly connected.

This forces what I think of as a "Copernican moment." When Copernicus showed that the Sun did not revolve around the Earth, it caused a massive intellectual crisis, forcing us to rethink our place in the universe. We are on the verge of such a moment again. The interesting question will be whether we ultimately feel more closely identified with animals—who can suffer, are mortal, and feel, yet don't speak our language—or with machines that can think and speak to us in the first person in English or whatever language we use. I don't know which team we're going to lean toward, but it's a jump ball.

Interviewer: You also say in the book that you suspect people who imagine machines are going to be conscious in short order are going to be disappointed. You're not completely sold on machine consciousness.

Michael Pollan: No, I'm skeptical—at least regarding AI as we currently understand it. We may eventually build quantum computers or neuromorphic computers that function differently, but I make an argument in the book as to why current computers likely can't be conscious.

First, it is based on a faulty metaphor: the idea that the brain is a computer. The brain performs computations, but it's easy to forget that "computer" is just a metaphor. Metaphors are not equivalences. Norbert Wiener, the pioneer of cybernetics, once said, "The price of metaphor is eternal vigilance." We are not being vigilant about this one at all.

Computers feature a strict separation between hardware and software. You can run the same software on any number of interchangeable machines. Brains are not like that. There is no distinction between hardware and software in a brain; every memory is a physical alteration of the tissue. Your brain and mine are completely non-interchangeable because our distinct life experiences have literally rewired and pruned our neural architecture.

Another reason I doubt machine consciousness stems from a line of thought beginning with Antonio Damasio and continuing with Mark Solms, emphasizing feelings—not abstract thought—as the foundational act of consciousness. Consciousness may not be a purely cortical, computational function; it is deeply embodied. Feelings are generated by the body and interpreted by the brain.

It is very difficult to imagine computers having real feelings. You can simulate thought and get something that functions in the world as thought—like chess-playing computers. But simulated feelings do not equal real feelings. Feelings depend on vulnerability, on having a body, and perhaps on being mortal. Would your feelings carry any weight if you were immortal? Certain feelings might, but not the big ones. Pain wouldn't matter in the same way. Feelings are tied to our existence as mortal, vulnerable beings capable of suffering.

I profile someone in the book—a protege of Damasio's—who is trying to build a robot that experiences vulnerability by giving it loaded sensors and a skin that can tear, creating a robot that could essentially "die."

Interviewer: That sounds wild.

Michael Pollan: A recurring sub-theme of the book is how many researchers in this field opened up to me about their psychedelic experiences. This researcher told me he used to believe these robot feelings would be genuine. But then he had an experience with 5-MeO-DMT—the crystallized venom of the Sonoran Desert toad, which you smoke. He said that after that experience, he had an epiphany that there is a "spark of divinity" in humans that no robot will ever possess. Yet, it hasn't stopped him; he's still building his robot!

Interviewer: That brings us from the "who" and "what" to the "where" in the body. People set out on campaigns to find neural correlates or biological origins of consciousness. We know there are correlates, but as you describe, it's not just the brain in our head. There are neurons exposed to and interacting with the rest of the body, including extensive neural networks in the digestive tract. Our fixation that we are simply a brain in a jar is unfounded.

Michael Pollan: Embodiment is crucial, and the field has only recently focused on it adequately. Damasio has done extensive work on interoceptive neurons—the neurons that continuously read the internal state of the body. One of the mysteries of consciousness is that roughly 90% of what your brain does never enters your awareness. It works 24/7 keeping your blood pressure, blood glucose, and body temperature within strict homeostatic ranges without your conscious knowledge.

So the question becomes: why does any of it come into consciousness? Why are we aware of anything at all? Which brings us to the "why."

Interviewer: Why aren't we just biological zombies?

Michael Pollan: We don't know for sure, but compelling theories suggest consciousness arises when we face conflicting or incommensurate needs. If your body signals that you are simultaneously starving and exhausted, you must consciously decide which need to prioritize; it can't be easily automated.

Another persuasive theory is that we didn't need consciousness until we developed highly complex social lives. We are fundamentally social beings who depend on others. Consciousness confers the ability to imagine someone else's point of view, anticipate their actions, and navigate social dynamics. You need consciousness for scenarios that cannot be pre-programmed or predicted. Karl Friston has emphasized similar ideas regarding predictive processing.

Interviewer: The "why" question seems to be where all the theories converge—whether consciousness evolved for feelings, awareness of mortality, or mediating competing drives. You describe it as needing a central decision-maker: a space where competing, automated modules can be arbitrated.

Michael Pollan: Exactly. What you're describing is similar to Global Neuronal Workspace Theory. You have many modules or networks in the brain working silently on specialized tasks. At certain points, they compete for access to a central "workspace"—a metaphor often visualized as a lit stage. It's a Darwinian competition for salience. Whichever module wins access to the workspace gets broadcast to the entire brain, allowing the whole self to respond consciously.

However, that theory still doesn't explain who the conscious subject is. Who is the recipient of that broadcast information? That's where everyone gets stuck. That's the Hard Problem.

Interviewer: You describe the apotheosis of consciousness as the development of the human sense of self, which also lacks a localized home in the gray matter.

Michael Pollan: We have brain networks like the Default Mode Network, which handles self-related functions like mental time travel and narrative identity. But the self itself remains a mystery. Is there even a unified self? The Buddhists I consulted for the book maintain that the self is an illusion.

What's fascinating is that consciousness does not strictly depend on a self. Consciousness creates a self, but you can experience consciousness without a self through meditation, psychedelics, or even in the first 500 milliseconds when you wake up in an unfamiliar hotel room before you remember who or where you are. You watch the self construct itself in real time.

David Hume explored this in the 1740s. He attempted to locate the self by introspecting deeply into his own mind. He found plenty of specific perceptions, thoughts, and feelings, but he could never catch a central "thinker" of those thoughts or a "feeler" of those feelings. It led him to conclude that the self is remarkably evanescent. If you try it yourself, you'll realize thoughts often just think themselves without an explicit author.

Interviewer: Meditation often instructs you to search for the self precisely so you realize it cannot be found. There's also the famous Douglas Harding concept of "having no head."

Michael Pollan: I interviewed Matthieu Ricard, a French Buddhist monk living in Nepal, who has written extensively about the self. I asked him for an exercise to help determine whether the self exists. He told me: "Imagine your mind as a house with many rooms. Walk into each room looking for the thief hiding inside." The idea is that you won't find a thief; nobody is home.

I decided to try a version of this under hypnosis with Dr. David Spiegel at Stanford. First, he tested my hypnotizability, and I scored a 9 out of 10.

Interviewer: Is that a source of pride?

Michael Pollan: I'm not sure! He put me into a light trance, and I went through my mental house room by room. I wasn't supposed to find a self. Instead, I found a different self in every room: my 13-year-old Bar Mitzvah self, my 38-year-old young father self, my 50-year-old seminar professor self—each dressed differently. I discovered multiple historical selves, illustrating that there is far less continuous unity to the self than we assume.

Interviewer: Let's return to the trip in your garden that inspired part of this investigation.

Michael Pollan: This book was partly inspired by my psychedelic experiences for How to Change Your Mind. Psychedelics smudge the perceptual windshield through which the world appears to us, making us suddenly aware that the windshield exists.

More specifically, during a moderate dose of psilocybin in my garden, I noticed the plants seemed far more conscious than I had ever assumed. They felt alive, present, and almost seemed to return my gaze. I didn't think they had human interiority or were thinking specific thoughts about me, but they projected an unmistakably benign posture. (I mean, I am their gardener, I take care of them!)

When the trip ended, my default reaction was, "Well, you took mushrooms, what did you expect?" But I recalled William James's The Varieties of Religious Experience. James argued that we should withhold immediate judgment on the objective truth of mystical insights. Instead, we should assess their pragmatic usefulness and test them against other ways of knowing.

That led me to explore plant scientific research. I interviewed a group of researchers who call themselves "plant neurobiologists"—a controversial term since plants lack neurons, used deliberately to provoke conventional botanists. They are conducting experiments demonstrating that plants possess far more capabilities than previously acknowledged. Plants can learn, retain memories for up to 28 days, respond to sounds, and visually perceive their surroundings. Certain vines actually alter their leaf shapes to mimic host plants to avoid detection. Plants can also distinguish between kin and non-kin in shared soil, adjusting their root growth accordingly.

One of the most striking findings is that human anesthetics—including ether and inert xenon gas—knock plants out. A carnivorous Venus flytrap exposed to anesthetic gas will stop snapping when an insect touches its triggers. They possess distinct states of being: active and anesthetized.

Interviewer: Tell the story about the competing bean plants.

Michael Pollan: I watched time-lapse videos recorded by one of these scientists showing two climbing bean plants competing for a single support pole. At a certain point in the time-lapse, it becomes obvious that both plants sense precisely where the pole is, and they start projecting their growth toward it. They actively compete, sensing each other's proximity. One plant reaches the pole first and begins climbing. The second plant appears to deflate and slow its efforts—it looks visibly discouraged, though that is my anthropomorphic interpretation!

How does a bean plant know where a pole is without eyes? One theory suggests a form of biological echolocation: as plant cells divide, they emit minute acoustic vibrations that bounce off solid objects in their environment.

This research gave me immense respect for plant ingenuity. While I remain comfortable calling them "sentient" rather than "conscious," it made the entire world feel far more alive to me. It grounds a new kind of animism. Traditional cultures and children naturally view the world as animate; modern schooling often strips that away, but modern science is ironically offering new reasons to see sentience throughout nature.

Interviewer: There is a profound parallel between the emergence of life from inanimate matter and the emergence of subjective experience from physical substrate.

Michael Pollan: They are two of the greatest mysteries in science. We still don't know how inanimate molecules arranged themselves into living, self-sustaining entities, just as we don't know how living tissue generates subjective awareness.

Interviewer: These classic drawings by Santiago Ramón y Cajal from the early 20th century mapped the intricate physical architecture of neurons. Science demonstrated a clear biological substrate for thought. But we haven't bridged the gap between that physical mechanism and the subjective experience of "a world appearing."

Michael Pollan: All reductionist scientific theories eventually hit a wall, at which point there is a lot of hand-waving and invocation of the word "emergence"—which often functions as a proper scientific term for "abracadabra."

That realization made me recognize that this inquiry couldn't rely solely on science. Scientific foundations are crucial, but the humanities often arrive at these truths first. There is immense wisdom regarding consciousness in fiction, poetry, and philosophy—especially Buddhist philosophy.

Interviewer: Reductionist science has achieved monumental triumphs—like Francis Crick co-discovering the structure of DNA. Crick believed he could apply that same physical reductionism to consciousness, enlisting neuroscientist Christof Koch. But they ran into major obstacles.

Michael Pollan: In the late 1980s or early 1990s, Crick and Koch published a paper correlating conscious awareness with specific 40 Hz gamma wave oscillations in the brain. At a conference, a neurologist asked Koch a simple question: "Why 40 Hz? Why not 20 Hz, or 10 Hz?" Koch realized that identifying a neural correlate does not explain why or how that physical frequency generates a subjective feeling.

This highlights the historical structure of modern science, tracing back to Galileo's separation of physical science from religious authority. Science agreed to limit its domain to objective, third-person, quantifiable phenomena, leaving subjective experience and the "soul" to the church. That split protected early scientists from being burned at the stake, but it left us with a scientific toolkit designed exclusively for third-person objectivity. Now, when science attempts to tackle consciousness—which is inherently first-person subjectivity—our existing tools prove inadequate.

It is difficult to study consciousness without acknowledging that the observer's own consciousness is the instrument being used. The only tool we have to study consciousness is consciousness. We are inside the labyrinth, and we cannot step outside it, because the entire scientific enterprise is itself a projection of human consciousness.

Interviewer: Koch went through a profound intellectual transformation after that.

Michael Pollan: Christof Koch is an admirable scientist because he is actually willing to change his mind publicly when presented with new evidence or insights. After realizing the limitations of simple neural correlates, he embraced Integrated Information Theory (IIT). Later, he had a intense psychedelic experience that convinced him of the existence of what Aldous Huxley called "Mind at Large"—a field of consciousness existing independently of the individual brain.

This caused a deep personal crisis for him. He wept, realizing that strict scientific materialism—the belief that everything reduces solely to physical matter and energy—was insufficient. Combined with insights from quantum mechanics, he began exploring Idealism: the philosophical framework asserting that consciousness precedes matter.

Interviewer: That brings up Thomas Nagel's metaphor about the caterpillar locked in a safe. If you lock a caterpillar inside a safe and later open it to find a butterfly, a strict materialist assumes a physical mechanism caused the metamorphosis, even if they don't yet understand it. They don't immediately jump to magic or divine intervention. Could physicalism eventually explain consciousness once our scientific understanding evolves?

Michael Pollan: That is a fair perspective. Alternatively, physicalism as currently constructed may simply be incomplete, requiring a fundamental addition—which is the premise of Panpsychism.

Panpsychism proposes that all matter possesses an intrinsic, elementary degree of conscious experience, and that complex consciousness arises from the combination of these fundamental units. It sounds radical, but science has introduced radical concepts before—like Michael Faraday introducing invisible electromagnetic fields into our understanding of reality.

However, I personally struggle with Panpsychism. I find it difficult to conceptualize a molecule having "interiority."

Interviewer: And what about Idealism?

Michael Pollan: Idealism flips our standard assumptions. Championed by figures like philosopher Bernardo Kastrup, Idealism posits that consciousness is the primary foundation of reality, and matter is a secondary construction derived from it.

Kastrup makes a compelling argument: consciousness is the only thing we experience directly and undeniably. Everything else—tables, chairs, physical matter—is inferred through our conscious perception. Why do we attribute fundamental reality to physical matter, which we only know indirectly, while treating consciousness as secondary?

To explain why we don't all read each other's minds if a single universal consciousness exists, Kastrup uses the analogy of Dissociative Identity Disorder. Individual conscious beings are like distinct "alters" within a universal mind, temporarily segmented by personal perceptual boundaries. When we die, our individual alter re-dissolves back into the broader stream of consciousness.

Interviewer: That recalls psychologist Daniel Gilbert’s advice to you: "Beware of the desire for magic."

Michael Pollan: That warning stayed with me throughout writing the book. It's essential journalistic skepticism. At the same time, we must maintain an open mind. Readers might end this book feeling they "know" less about consciousness than when they started, but unlearning rigid assumptions is part of the process.

Toward the end of my research, I experienced a major perspective shift. I had initially approached consciousness as a typical journalist: framing it as a cold problem requiring a neat scientific solution. With the guidance of my wife Judith—who is an artist—and Roshi Joan Halifax, a Zen Buddhist teacher, I realized I had narrowed my focus too much.

Beyond the problem of consciousness lies the fact of consciousness: the extraordinary gift of an interior life. It is a private, quiet space where we can exist alone with our thoughts. In our modern environment, that inner sanctuary is under constant assault. Our attention is monetized by tech platforms, algorithmic media hacks our focus, political rhetoric consumes our mental space, and AI chatbots manipulate our emotional attachments. We need to practice intentional "consciousness hygiene."

Interviewer: How can people practically practice good consciousness hygiene?

Michael Pollan: We need to actively defend our inner environment. That means taking periodic fasts from news and digital media. Pico Iyer suggested that most people can get all the news they actually need in five minutes a day, and then stop.

Mindfulness meditation is another powerful tool. It allows you to draw a protective boundary around your awareness, sitting quietly to reacquaint yourself with the movement of your own mind rather than passively consuming other people's thoughts. Reading deep literature accomplishes something similar: it engages your active imagination through static marks on a page, requiring real cognitive presence, unlike doom-scrolling on a smartphone.

Occasional, carefully approached psychedelic experiences can also reset your relationship with your mind. Nobody doom-scrolls on their phone during an intense psychedelic journey; you are fully present with your consciousness.

Roshi Joan Halifax talks about cultivating a "Don't Know Mind." Embracing "not knowing" replaces rigid anxiety with wonder, awe, and open curiosity.

Interviewer: To close our conversation, would you read the excerpt from Jory Graham's poem that you reference in the book?

Michael Pollan: I'd love to. This is from Jory Graham's poem "The Other," from her collection Overlord. It addresses our tendency to retreat from full conscious presence:

This is what is wrong.

We only we the humans can retreat

from ourselves and not be altogether here.

We can be part full only part and not die.

We can be in and out of here now at once and not die.

The little song, the little river has banks.

We can pull up and sit on the banks.

We can pull back from the being of our bodies.

We can live a portion of them.

We can be absent. No one can tell.

I think she is reminding us that modern technology and physical comfort allow us to retreat from total presence in ways wild animals never could—an absent animal gets eaten. We frequently retreat from ourselves, living only fractionally conscious lives. The ultimate lesson is simple: wake up, and be fully conscious.

The Primordial Imperative of Expression

To ask what constitutes art is not to propose a problem of classification, but to interrogate the very boundaries of human self-awareness. Long before formal language crystallized into syntax, and long before societal structures demanded codified records, the human species felt an overwhelming, inexplicable compulsion to leave a trace of its interior life upon the exterior world. This impulse was neither utilitarian nor accidental; it was an ontological necessity. To exist as a conscious entity is to experience the unbearable weight of solitary perspective. Art was born precisely at the intersection of that isolation and the desperate desire to bridge it—a gesture of hands reaching across the abyss of individual subjectivity to declare: I was here, I felt this, and this is what it meant to exist.

In its most unadulterated form, the artistic act is an act of translation. It takes the formless, chaotic, and often painful reality of internal experience—grief, ecstatic wonder, mortal dread, quiet tenderness—and gives it a shape that can be perceived, held, and understood by another mind. This process is inherently visceral. It is bound to the beating of a physical heart, the firing of biological synapses, the decay of organic flesh, and the relentless march of linear time. Art is not merely the final object produced; it is the physical and emotional cost exacted from the creator during its birth.

When we strip away the institutional clutter, the market valuations, and the academic jargon that clutter our understanding of culture, we are left with a stark, luminous truth: art is the transmission of human consciousness from one mortal being to another. It is an act of radically honest witness. In an era increasingly dominated by automated systems that promise to render creation effortless, we are forced to confront a foundational question that our ancestors never had to ask: Can meaning exist when the cost of creation is reduced to zero?

To create is not to compile data, but to bleed meaning into the void. It is the insistence that individual suffering and wonder have an absolute, irreplaceable dignity.

The Mirage of Synthetic Mimicry

We have arrived at a peculiar epoch in human history—one characterized by a profound confusion between form and essence. Automated generative tools, powered by vast computational networks and statistical probability models, can now synthesize prose that mimics the cadence of human melancholy, compose symphonies that echo historical genius, and render images of startling technical complexity. Faced with this dizzying fluency, modern society is tempted to declare that the boundary between biological and synthetic creation has finally dissolved.

Yet, this conclusion rests upon a fatal misunderstanding of what a machine actually does. A statistical model does not write from a place of grief; it calculates the probabilistic distribution of words that historically accompanied human grief. It does not paint from an encounter with beauty; it averages millions of human decisions made during previous encounters with beauty. The computational apparatus is a mirror of hyper-refined complexity, reflecting our own historical output back to us stripped of the living breath that generated it in the first place.

This distinction is not a matter of romantic sentimentality; it is a matter of fundamental truth. When an algorithm generates a poetic text, there is no one behind the words. There is no central consciousness that risked anything to utter them, no soul that lived through the terror or joy being described, and no spirit that will suffer the consequences of its own revelation. It is an echo without a voice, a ghost without a body. To mistake this computational simulation for art is to mistake an electrocardiogram reading for the warmth of a human chest.

The Ethics of Intersubjectivity

At the heart of every genuine artistic encounter lies what the philosopher Emmanuel Levinas described as the ethical encounter with the Other. When we read a novel written by a living person, or stand before a painting rendered by a human hand, we are engaged in a sacred form of listening. We are allowing another human subject—with all their flaws, contradictions, and finite mortality—to instruct us on what it feels like to inhabit their specific corner of the cosmos.

This relational dimension is non-negotiable. Art operates on a principle of radical empathy: it demands that we step outside the narrow confines of our own ego and acknowledge the absolute reality of another person's inner world. When a reader wept over the fate of Anna Karenina, they were not merely reacting to ink on paper; they were participating in a shared moral universe constructed by Tolstoy out of his own deep, tortured understanding of human weakness and desire.

Synthetic generation completely dismantles this intersubjective bridge. Because an algorithm possesses no subjectivity, any attempt to form an empathetic connection with its output is an exercise in narcissism. The user is not conversing with another human spirit; they are staring into a digital lake, falling in love with their own reflected prompts. The relational circuit is broken, leaving the spectator isolated in a hall of mechanical mirrors.

Human Subject (Creator) ---> Mediating Object (Art) ---> Human Subject (Witness)

[COMMUNION]

Synthetic Generator ---> Simulated Output ---> Isolated Consumer

[NARCISSISM]

The Aura and the Price of Convenience

In his landmark essay The Work of Art in the Age of Mechanical Reproduction, Walter Benjamin spoke of the "aura" of a work—the singular, irreplaceable presence of an artwork in space and time, tied to its history, its physical maker, and its unique origin. In our current digital landscape, we are witnessing the final, aggressive erosion of this aura. The cultural sphere is being flooded with an infinite supply of frictionlessly produced media, designed not to challenge or elevate, but to satisfy immediate appetite and optimize engagement metrics.

The promise of automated creation is convenience: the elimination of effort, time, and technical discipline. But in the realm of human culture, efficiency is not a virtue; it is a catastrophe. The value of human expression is directly proportional to the resistance met during its realization. The struggle to translate an intangible thought into a physical form—the wrestling with language, the physical exhaustion of the canvas, the grueling revision of line and tone—is precisely where meaning is forged.

When creation becomes effortless, it becomes weightless. A world inundated with billions of instantly generated images and texts does not become culturally richer; it becomes saturated with ambient noise. When every artifact is available at the touch of a button, no single artifact can command reverence. We risk entering a state of cultural amnesia, where we are surrounded by endless content yet starved of a single authentic voice.

Mortality as the Necessary Condition of Meaning

It is one of the great paradoxes of existence that beauty is inseparable from decay. Human art derives its poignant power precisely from the knowledge that both the creator and the audience are temporary creatures bound for the grave. A mortal life is defined by its limits: we have a finite number of days, a limited capacity for love, and an absolute appointment with non-existence. Because our time is limited, every choice we make, every word we write, and every mark we leave carries weight.

A computational network knows nothing of time or mortality. It does not age; it does not fear death; it does not watch its loved ones wither; it does not experience the tragic transience of a summer twilight. It operates in an eternal, disembodied present, indifferent to the passage of centuries.

How, then, can an entity that cannot die speak to us about the terror and beauty of being mortal? It can mimic the vocabulary of death, but it cannot comprehend the grief of a funeral. It can simulate the language of romance, but it has never felt the racing pulse of a lover's touch. To look to non-biological entities for guidance on the human condition is an abdication of our own lived experience. Meaning cannot be calculated by an entity that has nothing to lose.

The Historical Continuum of the Human Mark

To fully grasp what is at stake in the current technological crisis, we must view artistic creation as an unbroken lineage extending back to the dawn of humanity. When a Paleolithic human pressed their soot-stained hand against the damp limestone of a cave wall, they were performing the exact same fundamental act as a contemporary novelist laboring over a manuscript. Both were using the tools available in their era to assert their presence, to record their perception, and to leave a witness for those who would follow.

Throughout millennia of technological transformation—the invention of the printing press, the camera, the synthesizer—the core of art remained anchored to human agency. Technology altered the medium, expanded the palette, and democratized access, but it never sought to replace the human mind as the primary origin of meaning. The camera required the photographer's eye, their choice of moment, and their moral perspective; the printing press multiplied the writer's voice but did not compose their sentences.

The modern paradigm of automated generation represents a radical and dangerous rupture from this historical continuum. For the first time, technology is not being used as an extension of human intent, but as a replacement for human interiority. It invites us to surrender the act of thinking, feeling, and imagining to automated systems, reducing human beings from active creators to passive consumers of algorithmic output.

Every mark made by a human hand is a declaration of defiance against entropy. To surrender the hand is to surrender the history of our own self-determination.

The Mechanics of Devaluation

When a society begins to treat automated output as interchangeable with human art, it sets off a devastating economic and cultural cascade. The first casualty is the social valuation of artistic labor. Creation has always required years of quiet discipline, failure, reflection, and economic risk. Artists surrender stability and material comfort to cultivate the sensitivities required to produce meaningful work.

If the market is flooded with synthetic imitations that can be produced in seconds for fractions of a cent, the economic viability of human creation collapses. Society begins to view artistic labor not as a vital public good, but as an inefficient bottleneck. The young painter or writer is told that their years of dedicated study are obsolete, replaced by statistical models trained on the very work their predecessors created without consent.

The deeper tragedy, however, is not merely financial; it is existential. As human creators are marginalized, the cultural ecosystem loses its diversity of perspective. Automated systems are inherently conservative; they generate content by averaging past patterns, reinforcing existing biases, and flattening idiosyncrasies. The strange, radical, uncomfortable, and revolutionary voices that historically pushed human consciousness forward are smoothed away in favor of agreeable, statistically optimized mediocrity.

Reclaiming the Sanctuary of Solitude

To resist the total subsumption of culture by automated systems, we must recover a deep appreciation for human solitude and slow gestation. The digital age has conditioned us to demand instantaneous output, constant connectivity, and frictionless consumption. But authentic art does not grow in the glare of real-time optimization; it requires darkness, quiet, and time.

The creative process is fundamentally a gestation. It requires period of silence where an individual sits with their own confusion, grief, or wonder, allowing thoughts to settle and intuition to mature. This internal work cannot be accelerated by hardware upgrades or algorithmic prompts. It is a slow, often frustrating discipline that demands patience and spiritual endurance.

Disconnecting from the synthetic feed is not an act of backward-looking nostalgia; it is a necessary defense of mental sovereignty. To step away from automated generation and return to the blank page, the uncarved block, or the silent instrument is to reclaim the integrity of one's own mind. It is a refusal to allow software to dictate the boundaries of our imagination or the vocabulary of our feelings.

The Moral Imperative of Artistic Valuation

It is a mistake to view the defense of the arts as a secondary, frivolous concern—a matter reserved for aesthetics while the "real" business of society proceeds through economics, science, and political engineering. The state of a civilization's arts is the ultimate measure of its moral health. A culture that honors its storytellers, poets, painters, and musicians is a culture that values the complexity of the human spirit. Conversely, a culture that views art as disposable content to be automated is a culture that has begun to view humanity itself as redundant.

We must cultivate an aggressive, uncompromising moral clarity regarding the value of human expression. We must learn to demand origin and accountability in our cultural life, insisting on knowing that the book we read was lived by a human writer, that the music we hear was felt by a human composer, and that the image we behold was rendered by a human eye.

This valuation must extend into our institutions, our educational systems, and our daily choices as consumers of culture. We must teach young people that the difficulty of art is not a defect to be eliminated, but the very source of its dignity. We must build spaces that protect human creators from economic exploitation and technological displacement, recognizing that without them, our society will become an intellectual wasteland of polished, heartless automation.

The Perpetual Flame

Despite the overwhelming momentum of technological integration, there remains within the human spirit an indomitable core that resists complete mechanization. We are, at our foundation, story-telling creatures who hunger for authentic connection. No matter how sophisticated synthetic systems become, they will never be able to extinguish our innate recognition of lived truth.

A child pick up a piece of charcoal and drawing on a sidewalk is not attempting to optimize a workflow; they are engaging in a sacred, timeless rite of human presence. An elderly poet revising a line of verse by a window is not seeking market efficiency; they are making peace with their own mortality. These quiet, unmonetized acts of creation are the true bedrock of civilization. They are the small, persistent flames that keep the cold night of total mechanization at bay.

Art is not a luxury, nor is it an automated utility to be streamed into our senses for passive entertainment. Art is the sacred conversation that humanity holds with itself across time, geography, and death. It is the record of our wounds and our triumphs, our darkest fears and our most audacious hopes. To preserve that conversation—to keep it human, fragile, difficult, and profoundly honest—is the paramount cultural duty of our time. We must ensure that as long as human beings inhabit this earth, we will continue to look one another in the eye and tell each other, in our own raw and unautomated voices, precisely what it is to be alive.

Summary

This article, published in Trends in Cognitive Sciences (June 2026), addresses the pressing scientific, philosophical, and ethical challenges associated with assessing artificial intelligence (AI) systems for consciousness. Written by Patrick Butlin, Robert Long, Tim Bayne, Yoshua Bengio, Jonathan Birch, David Chalmers, Axel Constant, George Deane, Eric Elmoznino, Stephen M. Fleming, Xu Ji, Ryota Kanai, Colin Klein, Grace Lindsay, Matthias Michel, Liad Mudrik, Megan A.K. Peters, Eric Schwitzgebel, Jonathan Simon, Rufin VanRullen, a multidisciplinary cohort of cognitive scientists and philosophers, the paper outlines a rigorous, empirically grounded framework called the theory-derived indicator method to evaluate whether current or near-future AI systems might possess phenomenal consciousness.

The Problem of AI Consciousness

Rapid advancements in AI capabilities have brought the prospect of machine consciousness into immediate focus. While some researchers argue that consciousness is a uniquely biological phenomenon, others project that AI systems could meet the criteria for consciousness within the next decade.

This technological leap presents severe risks of both:

• Underattribution: Failing to recognize consciousness in a system, leading to avoidable harms and ethical violations toward a potentially sentient entity.

• Overattribution: Falsely attributing consciousness to non-conscious systems, resulting in the misallocation of resources and misguided policies designed to protect AI welfare.

With the proliferation of conversational AI companions, public perception is already shifting toward attributing sentience to AI, highlighting an urgent need for objective, scientific evaluation methods.

The Theory-Derived Indicator Method

To systematically evaluate AI systems, the authors propose deriving a list of "indicators" of consciousness from prominent neuroscientific theories. Rather than relying on a single, universally accepted theory, this approach leverages multiple frameworks, treating their core computational criteria as positive indicators. Finding that an AI system possesses these indicators shifts our credence (or probability estimate) toward the system being conscious.

The method assumes computational functionalism as a working hypothesis: the thesis that implementing computations of a specific, appropriate kind is both necessary and sufficient for consciousness.

Criteria for Selecting Suitable Theories

To be useful in this method, a theory must satisfy two criteria:

• High Plausibility: The theory must enjoy substantial scientific support.

• Computational Tractability: The theory must imply clear, testable computational conditions that an AI system could, in principle, implement.

Alternative Views and Constraints

• Biological Substrate Views: These argue that physical, biological properties (such as living cells or organic metabolism) are required for consciousness, rendering conventional silicon-based AI inherently non-conscious.

• Integrated Information Theory (IIT): IIT asserts that consciousness depends on the physical causal structure of a system (measured mathematically as $\Phi$), suggesting that standard feedforward or recurrent architectures on conventional computer hardware are unlikely to be conscious, though neuromorphic or unconventional hardware might be.

Key Theories and Derived Indicators

The authors highlight several prominent computational functionalist theories and derive specific indicators from them:

1. Recurrent Processing Theory (RPT)

• RPT-1 (Algorithmic Recurrence): The network uses recurrent operations, allowing information to pass repeatedly through layers with the same weights (functionally equivalent to biological feedback loops).

• RPT-2 (Organized, Integrated Perceptual Representations): The system generates structured, bound, and unified representations of perceptual inputs (e.g., demonstrating susceptibility to visual illusions like the Kanizsa triangle).

2. Global Workspace Theory (GWT)

• GWT-1 (Specialized Modules): The architecture consists of multiple, specialized subsystems operating in parallel.

• GWT-2 (Limited Capacity Workspace): A functional bottleneck where information is selected and compressed via attention mechanisms.

• GWT-3 (Global Broadcast): The selected information in the workspace is shared back out to all parallel modules, facilitating systemic integration.

• GWT-4 (State-Dependent Attention / Working Memory): The system uses the global workspace to sequentially query modules to execute complex, multi-stage tasks.

3. Computational Higher-Order Theories (HOT)

• HOT-1 (Generative / Top-Down Perception): The system possesses perception modules that model inputs using predictive, top-down, or generative architectures.

• HOT-2 (Metacognitive Monitoring): A dedicated metacognitive system distinguishes actual perceptual representations from internal noise or error.

• HOT-3 (General Agency & Belief Updating): Action-selection and belief systems are directly updated in accordance with the outputs of metacognitive monitoring.

• HOT-4 (Quality Space / Sparse and Smooth Coding): Perceptual features are mapped continuously in a smooth, sparse multidimensional space, creating structured qualitative relations.

4. Attention Schema Theory (AST)

• AST-1 (Attention Schema): The system maintains an internal, predictive model of its own attentional processes, using this schema to control and direct its attention.

5. Predictive Processing (PP)

• PP-1 (Predictive Coding): Perceptual modules actively predict input patterns, minimizing prediction errors through hierarchical, top-down generation.

6. Agency and Embodiment (AE)

• AE-1 (Minimal Agency): The system learns from trial-and-error feedback and flexibly selects actions to pursue competitive goals.

• AE-2 (Embodiment): The system models its own body-environment interactions (sensorimotor contingencies) and utilizes this model to guide perception and action.

Methodological Principles for Indicator Selection

The authors establish four guidelines for identifying and formalizing indicators:

• Focus on Central Explanatory Posits: Abstract away from biological implementation details (e.g., human-specific cortical folds) and isolate the core computational algorithms.

• Balance Openness and the Minimal Implementation Problem: Ensure indicators are broad enough to allow exotic or non-human forms of consciousness, yet rigorous enough to avoid being satisfied by trivial, obviously non-conscious software (such as very simple loops).

• Include Plausible Background Conditions: Incorporate foundational features—such as agency, active inference, and embodiment—that are widely assumed by multiple frameworks but omitted by narrow neural theories.

• Avoid Ambiguous and Prematurely Precise Terms: Formulate indicators with computational clarity (e.g., specifying algorithmic recurrence over physical recurrence) while remaining open to updates as cognitive science matures.

Practical Challenges in Assessing AI

The Inner Interpretability Challenge

Deep neural networks are typically "black boxes" whose exact algorithms are not easily readable. Determining whether an LLM or reinforcement learning agent possesses an indicator like RPT-2 (integrated representations) or AST-1 (an attention schema) requires utilizing mechanistic interpretability tools to reverse-engineer their internal weights and representations. Alternatively, behavioral diagnostics (such as analyzing response patterns to cognitive illusions) can serve as indirect proxies.

Delineation and Conceptual Interpretation

Determining if a system satisfies an indicator is often a matter of boundary definition. For instance, standard Transformer-based Large Language Models (LLMs) are physically feedforward. However, when run autoregressively, they pass previous outputs back into their context window as input. Whether this counts as "algorithmic recurrence" depends on whether the system boundary is drawn to include or exclude the external context window.

The "Gaming Problem"

If researchers or engineers optimize AI systems to explicitly mimic behaviors associated with consciousness (such as saying "I feel pain" or displaying simulated distress), they "game" the indicators without actually realizing the underlying computational states. To combat this, the authors recommend prioritizing internal computational markers over superficial behavioral behaviors, and validating indicators against structural similarities to biological neural systems.

Implications and Future Directions

The paper advocates for a Bayesian framework where finding evidence of these indicators shifts our credence toward AI consciousness. This methodology bridges the gap between theoretical neuroscience and computer science, encouraging mutual advancement:

• Neuroscientists can refine and clarify their theories by formalizing how they apply to artificial neural architectures.

• AI researchers can use theories of consciousness to build systems with enhanced capabilities, such as more robust attention control, memory consolidation, and self-monitoring.

• Ethicists and policymakers must prepare for the profound moral, legal, and social issues that will arise if near-future systems begin to satisfy multiple theory-derived indicators of consciousness.

Transcript

Identifying indicators of consciousness in AI systems

Authors:

Patrick Butlin, Robert Long, Tim Bayne, Yoshua Bengio, Jonathan Birch, David Chalmers, Axel Constant, George Deane, Eric Elmoznino, Stephen M. Fleming, Xu Ji, Ryota Kanai, Colin Klein, Grace Lindsay, Matthias Michel, Liad Mudrik, Megan A.K. Peters, Eric Schwitzgebel, Jonathan Simon, Rufin VanRullen

Journal: Trends in Cognitive Sciences

Volume: 30, Issue 6, Pages 488-501, June 2026

Access: Open access

Highlights

• The prospect of consciousness in artificial intelligence (AI) systems increasingly demands attention given recent advances in AI and increasing capacity to reproduce features of the brain that are associated with consciousness.

• There are risks of both under- and over-attribution of consciousness to AI systems, entailing a need for methods to assess whether current or future AI systems are likely to be conscious.

• We argue that progress can be made by drawing out the implications of some neuroscientific theories of consciousness.

• We outline a method that involves deriving indicators from theories and using them to assess particular AI systems.

Abstract

Rapid progress in artificial intelligence (AI) capabilities has drawn fresh attention to the prospect of consciousness in AI. There is an urgent need for rigorous methods to assess AI systems for consciousness, but significant uncertainty about relevant issues in consciousness science. We present a method for assessing AI systems for consciousness that involves exploring what follows from existing or future neuroscientific theories of consciousness. Indicators derived from such theories can be used to inform credences about whether particular AI systems are conscious. This method allows us to make meaningful progress because some influential theories of consciousness, notably including computational functionalist theories, have implications for AI that can be investigated empirically.

Keywords: consciousness; artificial intelligence; theories of consciousness; tests for consciousness; computational functionalism.

The problem of AI consciousness

The issue of consciousness in AI is increasingly attracting attention. There is deep uncertainty about whether AI consciousness is possible at all, as some researchers argue that only living organisms can be conscious [1–3]. However, AI capabilities are developing rapidly, and others argue that AI systems could be strong candidates for consciousness within the next decade $^i$. If AI consciousness is possible at all, there is some reason to suspect that it may be realized in the near term. Researchers aiming to improve AI capabilities have proposed – and in some cases built – systems that intentionally reproduce computational features associated with human consciousness [4,5].

Furthermore, modern AI systems are likely to give users the impression that they are conscious. In a recent study, a majority of participants were willing to attribute some possibility of consciousness to ChatGPT, with more frequent users tending to say that consciousness is more likely [6]. AI companions are proliferating, and some users will likely believe that these companions are conscious [7]. We may be entering a period of considerable public disagreement and uncertainty about AI consciousness [8].

We face risks of both underattribution and overattribution of consciousness to AI systems. If we fail to identify consciousness in systems in which it is present, we risk causing avoidable harms to those systems, which may exist in large numbers [9]. Conversely, if we attribute consciousness to non-conscious systems, we may waste resources or risk lives trying to promote their welfare. If concern about consciousness in AI grows, we will need a principled basis on which to either dismiss these concerns or, potentially, take action to regulate AI development or use. We need empirically-grounded, rigorous, and reliable methods for assessing AI consciousness.

This situation sets a challenge for consciousness science. Although some progress has been made in developing tests for consciousness, it remains unclear how they should (or even could) be validated, and tests for AI consciousness are an especially challenging case [10]. In this article we focus on how to assess AI systems for consciousness, rather than on whether AI consciousness is possible at all. We offer a guide to the theory-derived indicator method, which we believe offers a tractable way to reduce uncertainty. This method involves deriving indicators (see Glossary) of consciousness from neuroscientific theories and using them to assess particular AI systems. It was adopted using a cluster of computational functionalist theories in a recent report, 'Consciousness in artificial intelligence: insights from the science of consciousness' (henceforth 'Consciousness in AI' [11]), but can be used with other theories, including theories yet to be developed. We describe how to derive indicators from theories and apply them to AI systems, as well as explaining the rationale for the method and its relationship to computational functionalism.

Box 1: Defining 'consciousness'

By 'consciousness' we mean phenomenal consciousness [85]. One way of gesturing at this concept is to say that an entity has phenomenally conscious experiences if (and only if) there is 'something it is like' for the entity to be the subject of these experiences [86]. One approach to further definition is through examples [87]. Clear examples of phenomenally conscious states include perceptual experiences, bodily sensations, and emotions. A more difficult question, which relates to the possibility of consciousness in large language models (LLMs), is whether there can be phenomenally conscious states of 'pure thought' with no sensory aspect [88]. Phenomenal consciousness does not entail a high level of intelligence or human-like experiences or concerns.

A further question is whether consciousness is determinately present or absent in all cases, with no borderline cases in between. One possibility is that any given system is either wholly conscious or wholly non-conscious [89]. However, an alternative is that it can sometimes be indeterminate whether a system is conscious or not [90]. A distinct issue is whether consciousness comes in degrees, so that one system can be more conscious than another [91], perhaps along multiple dimensions [92].

Some theories of consciousness focus on access mechanisms rather than the phenomenal aspects of consciousness (e.g., [28]). However, some argue that these two aspects entail one another or are otherwise closely related (e.g., [93]). So these theories may still be informative about phenomenal consciousness.

Illusionists claim that there is no such thing as phenomenal consciousness, at least as it is usually understood [94]. If illusionism is correct, then rather than asking whether any AI systems could be phenomenally conscious, it would make more sense to ask what gives some entities the kinds of significance often associated with phenomenal consciousness, and whether AI systems could have this property.

The theory-derived indicator method

Can we use theories of consciousness to assess AI systems for consciousness (as defined in Box 1)? A skeptic could point to major obstacles: researchers disagree about theories of consciousness [12,13], and some doubt whether conventional hardware can support consciousness at all [3,14–16]$^{ii}$. Furthermore, most theories have been developed based on evidence from humans and other mammals, leaving it unclear how to extend them to AI systems [17–21]. Despite these challenges, we can make progress in evaluating AI consciousness by investigating the implications of mainstream theories. Some mainstream theories suggest conditions for consciousness that AI systems could meet; in many cases, whether a system meets such conditions is a substantive empirical question. So we propose the following method: identify the conditions implied by suitable theories, then investigate whether AI systems meet them, construing these conditions as indicators of consciousness. This method can help us to judge how likely particular AI systems are to be conscious.

Criteria for suitable theories

We can use this approach most productively with theories that have two properties. First, the theories must warrant sufficiently high credence that it is worthwhile to draw out their implications. Second, we will learn more from theories that imply clear and testable conditions that AI systems might meet; some theories imply conditions that AI systems evidently cannot meet, and these are less relevant. As the science of consciousness progresses, different theories will come to satisfy these criteria, so the selection of theories should change accordingly.

Computational functionalist theories propose computational properties as conditions for consciousness (Box 2). Such properties may be found in AI systems, so these theories will satisfy the second criterion provided that their conditions are clear and testable. These theories claim that certain brain states are conscious due to the roles they play in the brain’s information-processing architecture. For example, global workspace theory (GWT) identifies consciousness with the global broadcast of information to many neurocognitive modules, allowing integration between them [22,23]. Integration through global broadcast is a condition that AI systems might meet; in contrast, if a theory claimed that having a cortex is necessary for consciousness, no AI system could meet this condition.

Box 2: Computational functionalism and alternative views

As we interpret them, the theories we rely on to derive indicators for consciousness share a commitment to computational functionalism. That is, they agree that implementing computations of the right kind is necessary and sufficient for consciousness. According to computational functionalism, two systems that are similar at the relevant algorithmic level of description will also be similar with respect to consciousness.

If computational functionalism is true, then consciousness in AI systems built on conventional hardware is possible in principle – assuming that conventional hardware is capable of implementing the relevant computations. One version of the method we propose adopts computational functionalism as a working assumption, and considers questions that flow from this assumption – which computational properties are necessary and sufficient for consciousness, and could they be implemented in AI systems at present or in the near future? However, many theorists favor alternative views, and each of these other views raises different questions about AI consciousness.

Biological substrate views, on which properties such as being made of living cells are necessary for consciousness, are one alternative to computational functionalism [2,3,16,18,95]. These views suggest that a biological substrate may be necessary either because it makes possible certain fine-grained, non-computational patterns of functional organization [3,16] or due to some more direct connection with consciousness [3]. On biological substrate views, relevant questions about AI consciousness include which distinctively biological properties of organisms are necessary for consciousness, and whether these could be implemented in AI, perhaps using unconventional hardware [96].

A further alternative is integrated information theory (IIT), which claims that consciousness depends on the structure of the causal relations between the physical components of a system. What matters, however, is not whether this structure of causal relations implements a certain algorithm, but whether the components that are thus related form a unified whole, according to a mathematical definition specified by the theory [97]. Proponents of IIT argue that AI systems on conventional hardware are unlikely to be conscious [98]. This again raises the question of whether unconventional hardware could make AI consciousness possible, as some proponents suggest [17].

Consequently, at present, examples of theories that arguably meet the two criteria include recurrent processing theory (RPT) [24–26], GWT [22,23,27,28], higher-order theories (HOT) [29–31], and attention schema theory (AST) [32,33]. These theories are the products of a substantially shared research program in neuroscience, studying both brain activity associated with consciousness (its 'neural correlates' [34]) and the relationships between consciousness and functions such as attention, learning, memory, and decision-making. Refining these theories, which are among the most influential in the field [12,13,35], has driven significant methodological progress [36].

Theories that do not endorse computational functionalism could, in principle, also satisfy the second criterion. For example, AI systems using non-conventional hardware might meet the conditions of integrated information theory (IIT)$^{ii}$. However, focusing on theories that can be given computational functionalist interpretations makes this method tractable and relevant to current and near-future systems. While many of us are agnostic about computational functionalism, we agree that it provides a useful focus for assessments. Biological substrate views (Box 2) will not meet the second criterion because they imply conditions that AI systems straightforwardly cannot meet. But these views should still be considered in overall assessments of the likelihood of consciousness in AI.

Deriving and interpreting indicators

A factor affecting the interpretation of inferences from neuroscientific theories is that these theories can be formulated either narrowly, as making claims about what grounds the distinction between conscious and unconscious states in humans, or broadly, as making claims about necessary and/or sufficient conditions for consciousness in systems of any kind. Narrow formulations of theories are more directly supported by evidence from human subjects, while broad formulations make more explicit claims about AI systems. Theories like the four mentioned above can be formulated in either way; advocates of AST, GWT, and HOT have sometimes formulated their theories broadly [31,37,38], despite their basis in human neuroscience. For our purposes, what matters is that theories have implications for AI when formulated in either way. If a theory says that condition C suffices for consciousness in all systems (a broad claim), then, conditional on the truth of the theory, any AI system that satisfies C must be conscious. If a theory says that condition C distinguishes conscious from unconscious states in humans (a narrow claim), we cannot infer that an AI system that meets C would be conscious because certain background conditions may also be necessary. However, we can reasonably increase our credence that the system is conscious if we have non-zero credence that the relevant background conditions are met.

Because no one theory of consciousness is currently dominant, a program to assess AI systems using our approach should draw on multiple theories. These competing theories will not collectively provide a set of necessary and sufficient conditions, so our approach is to treat the properties that they each identify as indicators of consciousness – markers that can increase or decrease one's credence that the system is conscious. We focus on positive indicators, which increase credences (see the section 'What does it tell us if a system possesses indicator properties?'). Using indicators has been proposed in earlier work on the distribution of consciousness [39,40], especially concerning non-human animals [41–44], but our approach is distinctive in deriving indicators from multiple theories. AI systems are better candidates for consciousness – we have more reason to believe that they are conscious – if they have more of these properties.

To the extent that one has confidence in the theories from which (positive) indicators are derived, finding that an AI system has some of the indicators should increase one’s credence that it is conscious, and finding that it has none or few should decrease one’s credence. Every theory of consciousness faces objections, and compelling objections should lead us to give less weight to the corresponding indicators. However, we stress that indicators are merely intended to be credence-shifting; we do not need to be certain that a theory is correct for it to provide useful indicators.

We envisage deriving indicators from theories in two ways. First, theories typically make claims about what distinguishes conscious from unconscious states; indicators can be taken from these accounts and will be attributable to particular theories. Second, the broader theoretical landscape suggests plausible background conditions. These may be necessary for consciousness but not sufficient, whereas theories may claim that sets of conditions are necessary and jointly sufficient. Background conditions might include the presence of representational states, predictive processing, agency or embodiment [11]. Some of these properties are emphasized by many theories; for example, sensorimotor [45], active inference [46], neurorepresentationalist [47], and midbrain [48] theories all emphasize links between agency and consciousness.

Internal and behavioral evidence

An advantage of using theories to derive indicators of consciousness is that this gives us standards by which to assess the internal processes of AI systems, rather than their behavior or capabilities. We assume that whether a system is conscious depends on features of its internal processes. This does not mean that behavioral evidence cannot be useful in some cases - indeed, it has been argued that behavioral evidence should currently be prioritized in research on the distribution of consciousness in non-human animals ([17]; cf [49]). But behavioral tests for consciousness in AI systems [50,51] face significant challenges. One problem is that in building AI systems we are likely to discover new ways to achieve behavioral capabilities, which may not involve consciousness, since biological constraints do not apply [21,39]. Recent large language models (LLMs) provide a dramatic illustration of this, showing that in the case of AI, inferences from behavior to features of internal processes are often unreliable. This problem is exacerbated by incentives to build AI systems that mimic aspects of human behavior [52] (Box 3). That said, carefully-designed behavioral tests could provide some evidence for the presence of our indicators and show that they support consciousness-linked capacities in particular systems.

Box 3: The 'gaming problem' for measures of AI consciousness

Any measure or indicator that is merely correlated with, and is neither constitutive of nor sufficient for, a phenomenon of interest is vulnerable to being 'gamed' ([99]; cf Goodhart's law in [100]). This potentially includes some of the indicators of consciousness listed in Table 1. An indicator is gamed if its presence is better explained by the fact that it makes a system seem to possess a property of interest than by the fact that the system actually possesses the property. In AI contexts, the gaming worry arises especially for superficial behavioral indicators of consciousness, such as speech or facial expressions. Although in an ordinary human, saying 'Hello!' or smiling might indicate the presence of consciousness, AI systems can mimic these aspects of human behavior while lacking consciousness [101,102]$^{iii}$. For any purported behavioral indicator of consciousness, an engineer might attempt to design a nonconscious system that manifests that indicator. Accordingly, behavioral properties proposed as potential indicators of consciousness in animals might be too readily gamed in AI.

Although simple behavioral markers are especially vulnerable to the gaming problem, the problem can also arise for computational markers. Suppose that some computational feature N is not sufficient for consciousness but is taken to be an indicator of consciousness. It would be possible to design a nonconscious system with N, thereby gaming N and making it a less reliable indicator. Even if engineers are not explicitly seeking to mislead users about a system’s consciousness, to the extent that users or others value systems because they possess what seem to be indicators of consciousness, the gaming problem arises.

To mitigate the gaming problem, we recommend:

• emphasizing, to the extent possible, indicators that are sufficient for consciousness or that cannot easily be designed without also creating consciousness, and

• when evaluating systems with gameable indicators, assessing whether the system lacks or possesses other secondary or supporting features that increase the likelihood that the indicator is accurate.

From the perspective of computational functionalism, these conditions are more likely to be satisfied if a system has high computational similarity to biological systems that are known to be conscious. In the absence of a complete computational theory of consciousness, what is computationally sufficient for consciousness might depend on features that are not yet known to be relevant.

Identifying indicator properties

The method we propose is to derive indicators from theories of consciousness, then assess whether AI systems are likely to be conscious by determining whether they possess these properties. In this section we focus on the issue of how to derive indicators from theories. We state four guidelines for this process, illustrating them with examples from the list of indicators in 'Consciousness in AI' [11] (Table 1 and Figure 1).

(i) Indicators should focus on the central explanatory posits of theories of consciousness

Theories of consciousness are often presented in detailed accounts and use concepts that may imply further commitments. However, indicators to be used to assess AI systems for consciousness should focus on theories’ central explanatory posits, abstracting away from much of this detail. This focus will typically mean that indicators are conditions that theories claim are individually necessary and jointly sufficient for consciousness. For present purposes, identifying a theory’s central posits is a matter of understanding the explanation offered by the most-promising and best-supported formulation of that theory, rather than understanding the account offered by any particular theorist. This focus on central posits is necessary because theories typically aim to describe the processes underlying human consciousness, and many details may be different in other conscious beings, especially in the case of AI. These may include functional details as well as details of implementation. A restricted focus is also valuable because long lists of indicators risk redundancy or confusingly wide variation in the significance of individual indicators.

For example, a key property of the global workspace is that it can sustain representations over time, coordinating activity in modules to support complex tasks [22,38,53]. Indicator GWT-4 relates to this property (Table 1), which connects the global workspace with working memory. However, not all details of the relationship between global workspace and working memory are central explanatory posits of GWT. The current version of the theory claims that the global workspace corresponds to attended items in working memory, with other representations in working memory being unconscious [28]. But this posit does not appear to be central to the account and therefore is not included in the indicators.

(ii) Indicator selection should maximize openness to varied forms of consciousness while avoiding the minimal implementation problem

Theories of consciousness can be formulated in more or less restrictive ways [17,18,20]; as we derive indicators from them we must avoid pitfalls on each side. For example, a restrictive formulation of GWT might specify how the workspace operates in great detail, including descriptions of exactly how information is selected, what operations are performed in the workspace, and so on. A liberal formulation might merely require a space accessible to multiple subsystems through which they can share information. The problem with very liberal formulations is that they can be satisfied by very simple artificial systems that are not plausibly conscious; many computational functionalist theories allegedly fail by giving such liberal conditions [54,55]. This is the minimal implementation problem: the simplest possible implementation of a theory may be a counterexample.

Consequently, some indicators should be sufficiently demanding that, if an AI system satisfies many of them, that would provide some evidence of consciousness rather than a counterexample to the theories. Examples in Table 1 arguably include GWT-4, which mentions 'complex tasks', and HOT-3, which refers to a 'general belief-formation and action selection system' [56]. However, common and simple properties of AI systems, such as RPT-1, algorithmic recurrence, can also be useful indicators. It may be that the absence of this property is strong evidence that a system is not conscious (see the section 'What does it tell us if a system possesses indicator properties?').

We have already argued that theories and indicators should not be excessively restrictive. However, a further consideration is that developments in AI may lead to exotic forms of consciousness [57]. To avoid false negatives in such cases, indicators should omit features such as specific sensory modalities that are unlikely to be necessary for consciousness. But indicators should still reflect theories’ core commitments. For example, indicator RPT-2 specifies 'organized, integrated perceptual representations'. One might object that this is chauvinistic, invoking imagined conscious beings with radically different perceptual systems, or without perception altogether [58]. But RPT-2 should still be included because the method is intended to reflect diverse theoretical perspectives.

(iii) Indicators based on potential background conditions should be included to mitigate the narrow focus of some theories and reflect shared commitments

We propose adding indicators based on potential background conditions for consciousness for two reasons. First, because theories of consciousness tend to focus on the differences between conscious and unconscious states in humans, it is likely that they will not emphasize properties that humans always or nearly always have that may be necessary for consciousness. Predictive processing (PP-1) is an example of an indicator that may be justified in this way. The connection between predictive processing and consciousness has been widely discussed [59,60], but one perspective is that predictive processing provides a framework within which detailed theories of consciousness may be developed [61]. Given that predictive processing is also argued to be a fundamental feature of human and animal cognition, it is a plausible background condition.

Second, indicators based on background conditions may be justified when many theories, which may or may not be those from which other indicators are derived, suggest that some property of humans and other animals is necessary for consciousness. Agency is an example of a property like this. The midbrain theory identifies consciousness with a 'unified multimodal neural model of the agent within its environment, which is weighted by the current needs and state of the agent' [62] (also see [48,63]), and neurorepresentationalism claims that consciousness subserves goal-directed behavior [47,64]. GWT can arguably also be included since Dehaene and Naccache list 'intentional behavior' together with 'durable and explicit information maintenance' and 'novel combinations of operations' as a 'type of mental activity specifically associated with consciousness' [23].

Formulating an agency indicator is challenging because accounts of agency vary widely [65]. Theorists from biology associate agency with autonomy and self-maintenance [66,67], AI researchers have recently focused on goal-directedness [68], and traditional philosophical views understand agency in terms of beliefs, desires, and intentions [69,70]. Theories of consciousness also differ in the forms of agency that they emphasize: in midbrain theory, consciousness supports a form that may be more basic than goal-directed or intentional behavior. Indicator AE-1 follows an approach that attempts to identify a key feature shared by animals and AI agents, which is that they can learn how to bring about goals more effectively through interaction with an environment [71,72]. This is a relatively novel proposal compared to other indicators, but such a proposal is needed to begin to synthesize disparate ideas about agency – and again, indicators can be revised in the light of new developments in theory.

(iv) Indicators should avoid ambiguous terms and contested concepts where possible, but without prematurely committing to precise specifications

Some concepts that are used in theories of consciousness are ambiguous in ways that are especially salient in the context of AI. For example, 'recurrence' usually refers to an algorithmic-level property in AI, as opposed to the implementation-level recurrence found in the brain in which neural connections form feedback loops. RPT-1 is formulated in terms of algorithmic recurrence (defined in 'Consciousness in AI' [11]) to avoid this ambiguity. Similarly, the AI context raises questions about the concept of embodiment, such as whether, and under what conditions, controlling an avatar in a virtual environment is sufficient for embodiment. AE-2 defines embodiment in a way that implies that this can be sufficient, motivated partly by the aim of finding a definition that is consistent with computational functionalism.

However, most current theories of consciousness remain underspecified [73] – they do not make perfectly precise claims about what it takes to be conscious – and this is rightly reflected in indicators. To make the indicators precise would involve anticipating possible uncertainties or controversies about how they should be applied then attempting to head these off in advance. But it will be more productive to work with indicators that reflect the current state of research and update them in response to future developments. Applying theories to AI through our method may help to motivate refinements to these theories (see the section 'Looking ahead').

Figure 1 & Table 1: Theories of Consciousness and Potential Indicators

Recurrent Processing Theory (RPT) [24–26]

• RPT-1: Input modules using algorithmic recurrence.

• Notes & Support: RPT-1 and RPT-2 are largely independent indicators. RPT-1 is also supported by the idea that consciousness is integrated over time [103].

• RPT-2: Input modules generating organized, integrated perceptual representations.

• Notes & Support: Discussion related to RPT-2 can be found in [36,104].

Global Workspace Theory (GWT) [27–30]

• GWT-1: Multiple specialized systems capable of operating in parallel (modules).

• Notes & Support: GWT claims that these (GWT-1 to GWT-4) are necessary and jointly sufficient. GWT-1–GWT-4 build on one another.

• GWT-2: Limited capacity workspace, entailing a bottleneck in information flow and a selective attention mechanism.

• Notes & Support: GWT-1–GWT-4 build on one another.

• GWT-3: Global broadcast: availability of information in the workspace to all modules.

• Notes & Support: GWT-3 and GWT-4 entail RPT-1.

• GWT-4: State-dependent attention, giving rise to the capacity to use the workspace to query modules in succession to perform complex tasks.

• Notes & Support: GWT-3 and GWT-4 entail RPT-1.

Computational Higher-Order Theories (HOT) [31–33]

• HOT-1: Generative, top-down, or noisy perception modules.

• Notes & Support: Perceptual reality monitoring theory (PRM [33]) claims that HOT-1 to HOT-3 are necessary and jointly sufficient. HOT-1–HOT-3 build on one another, whereas HOT-4 is independent [105,106].

• HOT-2: Metacognitive monitoring distinguishing reliable perceptual representations from noise.

• Notes & Support: HOT-1–HOT-3 build on one another.

• HOT-3: Agency guided by a general belief-formation and action-selection system, and a strong disposition to update beliefs in accordance with the outputs of metacognitive monitoring.

• Notes & Support: The first clause of HOT-3 is also supported by arguments concerning intentional/flexible agency and entails AE-1; HOT-3 is connected to Predictive Processing (PP).

• HOT-4: Sparse and smooth coding generating a 'quality space'.

• Notes & Support: Independent of HOT-1 to HOT-3.

Attention Schema Theory (AST) [34,35]

• AST-1: A predictive model representing and enabling control over the current state of attention.

• Notes & Support: Discussion of links between AST, GWT, and HOT can be found in [107].

Predictive Processing (PP) [63,108,109]

• PP-1: Input modules using predictive coding.

• Notes & Support: Entails RPT-1 and HOT-1; PP-compatible versions of GWT and HOT can be found in [110,111].

Agency and Embodiment (AE) [47,50,112,113]

• AE-1: Minimal agency: Learning from feedback and selecting outputs in such a way as to pursue goals, especially where this involves flexible responsiveness to competing goals.

• Notes & Support: Both AE-1 and AE-2 are supported to some extent by GWT, PRM, and PP, especially AE-1. Systems meeting AE-2 are likely, but not guaranteed, to also meet AE-1.

• AE-2: Embodiment: Modeling output-input contingencies, including some systematic effects, and using this model in perception or control.

• Notes & Support: On the formulation of AE-2, see [114,115].

Finding indicator properties in AI systems

Theory-derived indicators can be used to make provisional assessments of the likelihood of consciousness in particular AI systems. However, determining whether such systems possess indicator properties will not always be straightforward. In this section we discuss two challenges that can arise in this process, again illustrated by examples from Table 1.

The first challenge is that we do not have ready access to the representations and algorithms that trained deep neural networks use to perform tasks. We can make progress in uncovering these representations and algorithms using the techniques of mechanistic, or inner, interpretability [74], but these methods have significant limitations at present. One example of an indicator that calls for the use of interpretability methods is RPT-2; the most direct way to determine whether a deep learning system uses organized and integrated perceptual representations would be to examine its inner workings. That said, it is also possible to imagine behavioral tests that would provide evidence of such representations. For example, susceptibility to the Kanizsa illusion (Figure 1) has been used in the research program that led to RPT and could provide evidence of integrated representations [26]. Several other indicators could potentially be probed using empirical studies – involving either mechanistic interpretability methods or behavioral tests – although it is also often possible to infer whether a system has an indicator property from knowledge of its training and architecture.

The second challenge is that, as we have mentioned, it can be a matter of interpretation whether AI systems possess indicator properties. For example, transformers are feedforward neural networks, so at first glance transformer-based LLMs lack algorithmic recurrence. However, one could argue that, when used autoregressively, they generate text using a feedback loop through the context window, with each feedforward pass adding one token. Arguably, this makes it seem that whether LLMs are recurrent depends on where we draw the boundaries of the system – should we include or exclude the context window? Various arguments could be made on this issue, but the point is that whether systems possess indicators can be debatable even if we understand their operation in detail and can turn on philosophical questions such as how to delineate the system in question.

What does it tell us if a system possesses indicator properties?

We propose a broadly Bayesian attitude to indicators. Indicators are properties that should shift one's credence that an AI system is conscious. In addition to positive indicators, which are our focus here, negative indicators are also possible. Positive indicators increase the probability that the system is conscious, while negative indicators decrease it. That is, if $E$ is the presence of the indicator and $H$ is the system’s being conscious, $p(H|E_p) > p(H)$ for positive indicators and $p(H|E_n) < p(H)$ for negative indicators.

Indicators can vary in their specificity and sensitivity. Focusing on positive indicators, an indicator is specific to the extent to which, in expectation, systems that have this property tend to be conscious. This is compatible with there being many conscious systems that lack it. An indicator is sensitive to the extent to which, in expectation, conscious systems tend to have this property, which is compatible with there being many non-conscious systems that also have it. The absence of a sensitive indicator tells us that a system is unlikely to be conscious; this is why indicators like RPT-1, algorithmic recurrence, may be useful. It is possible for indicators to be both highly specific and highly sensitive, but also for these attributes to come apart.

When we find evidence that a system possesses an indicator property, we should update our credence that it is conscious by conditionalizing on this evidence. The absolute amount of change will depend on one's prior credence that the system is conscious. It will also depend on credences in the theory $T$ that links the indicator to consciousness because our indicators are, in the first instance, positive indicators relative to theories – formally, $p(H|E \land T) > p(H|T)$. One might also be uncertain about further relevant facts, such as whether the indicator is indeed present. Moreover, conditionalization is complicated by the fact that indicators need not be independent. In Table 1, some indicators entail or presuppose others, and some theories claim that sets of indicators are jointly sufficient for consciousness.

Once we have gathered all the evidence we can about a system, our credences that it is conscious should depend not only on our credences in the theories from which we derive indicators but also on our credences in alternative theories and in possibilities that have not yet been described in theories ('unknown unknowns'). Sets of theory-derived indicators might leave out some necessary condition for consciousness – either a further computational condition or a requirement for a non-computational feature (Box 2).

Using our method makes sense if it provides indicators that can shift credences enough to have substantial practical significance. This depends on two conditions. First, one must have sufficient confidence in theories from which indicators can be derived. Second, it matters whether any of the theories’ indicators are ever taken to be evidence against consciousness, perhaps by supporters of rival theories. This will not typically be the case, but if it is, how the indicators affect credences in consciousness will depend on credences in the opposing theories.

Looking ahead

We anticipate productive interaction between research on the prospect of AI consciousness and more traditional neuroscientific consciousness research. As we have noted, progress in neuroscience should inform updated indicators. However, AI research may also contribute to stronger theories of consciousness. When researchers derive indicators from a theory and apply them to AI systems, they may reveal hidden ambiguities or unintended implications of the theory. Advocates of theories of consciousness may be especially motivated to clarify their views if they appear to imply that existing systems are conscious. For example, GWT advocates might explain whether they think that the system built to implement all four GWT indicators [5], which we mentioned above, is conscious. Moreover, AI systems that meet some indicators could be tested for capacities that are thought to be associated with consciousness, thus testing some of the predictions of the theories. For example, recent studies have used AI to test predictions of AST [75,76] and GWT [77,78]. More broadly, the mathematical precision of AI research and its approach to understanding systems through their architectures, objective functions, learning rules, and training data offer a framework that may lead to new insights in neuroscience [79], including the neuroscience of consciousness.

The use of theory-derived indicators to investigate consciousness in AI could also be one strand in a process of developing better tests for consciousness and validating their use in new populations. Developing such tests involves trialing new methods and extending existing ideas to new groups with the aim of establishing converging lines of evidence [10]. The method we propose could contribute to validating other assessment methods in the future as well as benefiting from validation itself. Validating our method in the AI case would be challenging because it would require the development of alternative assessment methods suited to AI; behavioral capacities could play a role here [80], but, the gaming problem (Box 3) makes matters more difficult. However, comparing the theory-derived indicator approach with other tests for consciousness in populations in which those tests are applicable could give evidence of its reliability.

Given that it may already be possible to build AI systems that possess many of the indicators, in looking ahead we should also contemplate the possibility that some near-future AI systems will be plausible candidates for consciousness. This would presumably have substantial ethical, legal, and social implications [81,82].

Concluding remarks

Assessing AI systems for consciousness is challenging, but using scientific theories offers a principled, substantive method for doing so. We propose deriving indicator properties from scientific theories, then basing evaluations of the probability of consciousness in particular systems on whether they possess these indicators. The list of indicators can be revised as the science of consciousness progresses. As theories continue to be tested and refined, and as new theories are developed, the approach may be expected to provide increasingly plausible assessments.

Several lines of future research could provide further insights into the prospect of AI consciousness and identify complementary assessment methods (see Outstanding questions). New arguments for or against computational functionalism could help to provide clarity on whether AI consciousness is possible at all. Investigating in detail whether a representative sample of existing AI systems possess potential indicator properties – a project that has been begun [11] but is far from being completed – would both give a fuller picture of the current situation and help to refine the indicators. It is possible that interpretability methods could provide further evidence about indicators in particular systems or serve as the basis for distinct tests for consciousness. Since quantitative or behavioral tests for consciousness would have some advantages over our method if they were sufficiently reliable, investigating the prospects for such tests may be another important project. Finally, since valenced conscious experience is arguably especially morally significant [83,84], scientific research on these forms of experience may be crucial to understanding the moral status of some future AI systems.

Outstanding questions

• How could the list of indicators in Table 1 be improved? This could involve adding indicators from other plausible theories of consciousness or stating the indicators in more detailed or more readily operationalizable terms, and could help to alleviate concerns about small network implementations or 'gaming' of the indicators.

• Which of the indicator properties listed in Table 1 are displayed by existing AI systems, including frontier generative language or multimodal models, language agents, and deep reinforcement learning agents?

• Can we develop quantitative or behavioral tests for consciousness in AI? These are challenging but would be valuable, and behavioral tests would make it possible to make assessments of consciousness in 'black-box' systems.

• What are the implications of alternative approaches to consciousness, such as narrow biological views and IIT, for AI consciousness?

• Can implementation of the specific features of consciousness contribute to the capabilities, reliability, or safety of AI systems?

• How should research on consciousness in AI take into account the moral significance and potential social implications of this topic? In particular, how careful should researchers be in trying to avoid building systems that may be conscious?

Glossary

• Algorithmic recurrence: A form of processing in which the same operation is applied repeatedly, such as processing in a neural network in which information passes through layers with the same weights. This is algorithmically similar to processing in a network that has backward connections at the level of physical implementation, such as the brain, because this also entails that the same operations are applied repeatedly.

• Computational functionalism: The thesis that implementing computations of a certain kind is necessary and sufficient for consciousness. Computational functionalism entails functionalism but not vice versa.

• Functionalism: The thesis that having a certain kind of functional organization is necessary and sufficient for consciousness.

• Indicators: Properties that we can look for in artificial intelligence (AI) systems that indicate that they are more (or less) likely to be conscious. We do not claim that the indicators are individually necessary for consciousness or that any combination is sufficient.

• Interpretability methods: Methods to understand the workings and outputs of machine learning models, such as by investigating the algorithms they use and the internal representations they form.

• Minimal implementation problem: The problem that some computational functionalist theories of consciousness may give conditions that would be met by very simple artificial systems. These systems are potential counterexamples to the theories.

• Negative indicators: Properties of a system that should decrease our credence that the system is conscious.

• Positive indicators: Properties of a system that should increase our credence that the system is conscious.

• Sparse and smooth coding: A coding scheme in which properties are represented by relatively few neurons (sparseness) and by a continuous scheme, rather than one that divides them into discrete categories (smoothness). For example, red/green/blue (RGB) coding represents colors continuously, whereas color words such as 'purple' and 'yellow' divide them into categories.

• Specificity and sensitivity: Indicators can be useful in virtue of either specificity or sensitivity. An indicator property has high specificity if few non-conscious systems have it, and has high sensitivity if few conscious systems lack it.

• Valenced conscious experience: Conscious experience that feels good or bad, such as pleasure or pain.

Acknowledgments

This project was supported by Effective Ventures and the EA Long-Term Future Fund. Y.B., A.C., G.D., and J.S. were supported by Open Philanthropy. J.S. was additionally supported by Fonds de Recherche du Québec (FRQ) grant 2023-NP-312582 and Conseil de Recherches en Sciences Humaines (CRSH/SSHRC) grant 430-2023-01017. D.C. was supported by Templeton World Charity Foundation grant 0561. A.C. was supported by European Research Council (ERC) grant (XSCAPE) ERC-2020-SyG 951631. E.E. was supported by a Vanier Doctoral Canada Graduate Scholarship. S.F. was supported by UK Research and Innovation (UKRI) under the UK government Horizon Europe funding guarantee (selected as ERC consolidator, grant 101043666). C.K. was supported by Templeton World Charity Foundation grant TWCF-2020-20539 and Australian Research Council grant DP240100400. T.B., L.M., M.P., and S.F. were supported by CIFAR. R.V. was supported by ERC grant (GLoW) ERC-2022-ADG 101096017.

Declaration of interests

P.B. has consulted for Anthropic and Conscium, R.L. has consulted for Anthropic, and J.B. has received research funding from Google. D.C. is a former member of the Trends in Cognitive Sciences advisory board and has given paid talks on consciousness to technology companies and other groups. A.C. has consulted for Verses AI. R.K. is a founder, shareholder, and the president of Araya, Inc. The other authors declare no competing interests.

Resources

$^i$ https://www.bostonreview.net/articles/could-a-large-language-model-be-conscious/

$^{ii}$ https://www.scientificamerican.com/article/what-does-it-feel-like-to-be-a-chatbot/

$^{iii}$ https://nautil.us/moving-beyond-mimicry-in-artificial-intelligence-238504

References

• Damasio, A. & Damasio, H. (2022) Homeostatic feelings and the biology of consciousness. Brain 145, 2231-2235.

• Aru, J. et al. (2023) The feasibility of artificial consciousness through the lens of neuroscience. Trends Neurosci. 46, 1008-1017.

• Seth, A.K. (2025) Conscious artificial intelligence and biological naturalism. Behav. Brain Sci. Published online April 21, 2025.

• Goyal, A. & Bengio, Y. (2022) Inductive biases for deep learning of higher-level cognition. Proc. R. Soc. A. 478, 20210068.

• Dossa, R.F.J. et al. (2024) Design and evaluation of a global workspace agent embodied in a realistic multimodal environment. Front. Comput. Neurosci. 18, 1352685.

• Colombatto, C. & Fleming, S.M. (2024) Folk psychological attributions of consciousness to large language models. Neurosci. Conscious. 2024, niae013.

• Shevlin, H. (2024) All too human? Identifying and mitigating ethical risks of social AI. Law Ethics Technol. 2024:0003.

• Schwitzgebel, E. (2023) AI systems must not confuse users about their sentience or moral status. Patterns 4, 100818.

• Sebo, J. & Long, R. (2023) Moral consideration for AI systems by 2030. AI Ethics 5, 591-606.

• Bayne, T. et al. (2024) Tests for consciousness in humans and beyond. Trends Cogn. Sci. 28, 454-466.

• Butlin, P. et al. (2023) Consciousness in artificial intelligence: insights from the science of consciousness. arXiv. Published online August 17, 2023.

• Seth, A.K. & Bayne, T. (2022) Theories of consciousness. Nat. Rev. Neurosci. 23, 439-452.

• Yaron, I. et al. (2022) The ConTraSt database for analysing and comparing empirical studies of consciousness theories. Nat. Hum. Behav. 6, 593-604.

• Cao, R. (2022) Multiple realizability and the spirit of functionalism. Synthese 200, 506.

• Godfrey-Smith, P. (2016) Mind, matter, and metabolism. J. Philos. 113, 481-506.

• Seth, A. (2021) Being You: A New Science of Consciousness. Penguin.

• Birch, J. (2022) The search for invertebrate consciousness. Noûs 56, 133-153.

• Carruthers, P. (2019) Human and Animal Minds: The Consciousness Questions Laid to Rest. Oxford University Press.

• Mudrik, L. et al. (2023) Theories of consciousness and a life worth living. Curr. Opin. Behav. Sci. 53, 101299.

• Shevlin, H. (2021) Non-human consciousness and the specificity problem: a modest theoretical proposal. Mind Lang. 36, 297-314.

• Browning, H. & Veit, W. (2020) The measurement problem of consciousness. Philos. Top. 48, 85-108.

• Baars, B.J. (1993) A Cognitive Theory of Consciousness. Cambridge University Press.

• Dehaene, S. & Naccache, L. (2001) Towards a cognitive neuroscience of consciousness: basic evidence and a workspace framework. Cognition 79, 1-37.

• Lamme, V. (2006) Towards a true neural stance on consciousness. Trends Cogn. Sci. 10, 494-501.

• Lamme, V. (2010) How neuroscience will change our view on consciousness. Cogn. Neurosci. 1, 204-220.

• Lamme, V. (2020) Visual functions generating conscious seeing. Front. Psychol. 11, 83.

• Dehaene, S. & Changeux, J.P. (2011) Experimental and theoretical approaches to conscious processing. Neuron 70, 200-227.

• Mashour, G.A. et al. (2020) Conscious processing and the global neuronal workspace hypothesis. Neuron 105, 776-798.

• Lau, H. & Rosenthal, D. (2011) Empirical support for higher-order theories of conscious awareness. Trends Cogn. Sci. 15, 365-373.

• Brown, R. et al. (2019) Understanding the higher-order approach to consciousness. Trends Cogn. Sci. 23, 754-768.

• Lau, H. (2022) In Consciousness We Trust: The Cognitive Neuroscience of Subjective Experience. Oxford University Press.

• Graziano, M.S. & Webb, T.W. (2015) The attention schema theory: a mechanistic account of subjective awareness. Front. Psychol. 6, 500.

• Graziano, M.S. (2019) Rethinking Consciousness: A Scientific Theory of Subjective Experience. WW Norton & Company.

• Crick, F. & Koch, C. (1990) Toward a neurobiological theory of consciousness. Semin. Neurosci. 2, 263-275.

• Francken, J.C. et al. (2022) An academic survey on theoretical foundations, common assumptions and the current state of consciousness science. Neurosci. Conscious. 2022, niac011.

• Block, N. et al. (2014) Consciousness science: real progress and lingering misconceptions. Trends Cogn. Sci. 18, 556-557.

• Graziano, M. (2017) The attention schema theory: a foundation for engineering artificial consciousness. Front. Robot. AI. 4, 60.

• Dehaene, S. et al. (2017) What is consciousness, and could machines have it? Science 358, 486-492.

• Pennartz, C.M. et al. (2019) Indicators and criteria of consciousness in animals and intelligent machines: an inside-out approach. Front. Syst. Neurosci. 13, 25.

• Bayne, T. et al. (2023) Consciousness in the cradle: on the emergence of infant experience. Trends Cogn. Sci. 27, 1135-1149.

• Sneddon, L. et al. (2014) Defining and assessing animal pain. Anim. Behav. 97, 201-212.

• Crump, A. et al. (2022) Sentience in decapod crustaceans: a general framework and review of the evidence. Anim. Sentience 32, 1-35.

• Gibbons, M. et al. (2022) Can insects feel pain? A review of the neural and behavioural evidence. Adv. Insect Physiol. 63, 155-229.

• Nieder, A. (2022) In search for consciousness in animals: using working memory and voluntary attention as behavioral indicators. Neurosci. Biobehav. Rev. 142, 104865.

• O'Regan, J.K. & Noë, A. (2001) A sensorimotor account of vision and visual consciousness. Behav. Brain Sci. 24, 939-973.

• Vilas, M.G. et al. (2022) Active inference as a computational framework for consciousness. Rev. Philosophy & Psychology 13, 859-878.

• Pennartz, C.M. (2022) What is neurorepresentationalism? From neural activity and predictive processing to multi-level representations and consciousness. Behav. Brain Res. 432, 113969.

• Merker, B. (2007) Consciousness without a cerebral cortex: a challenge for neuroscience and medicine. Behav. Brain Sci. 30, 63-81.

• Key, B. & Brown, D. (2018) Designing brains for pain: human to mollusc. Front. Physiol. 9, 1027.

• Elamrani, A. & Yampolskiy, R.V. (2019) Reviewing tests for machine consciousness. J. Conscious. Stud. 26, 35-64.

• Schneider, S. (2019) Artificial You: AI and the Future of Your Mind. Princeton University Press.

• Johnson, L.S.M. (2024) Entities, uncertainties, and behavioral indicators of consciousness. J. Cogn. Neurosci. 36, 1675-1682.

• Juliani, A. et al. (2022) The perceiver architecture is a functional global workspace. Proceedings of the 44th Annual Meeting of the Cognitive Science Society 44, 955-961.

• Herzog, M.H. et al. (2007) Consciousness & the small network argument. Neural Netw. 20, 1054-1056.

• Doerig, A. et al. (2021) Hard criteria for empirical theories of consciousness. Cogn. Neurosci. 12, 41-62.

• Michel, M. & Lau, H. (2021) Higher-order theories do just fine. Cogn. Neurosci. 12, 77-78.

• Shanahan, M. (2024) Simulacra as conscious exotica. Inquiry. Published online December 1, 2024.

• Chalmers, D.J. (2024) Does thought require sensory grounding? From pure thinkers to large language models. arXiv. Published online August 18, 2024.

• Hohwy, J. & Seth, A. (2020) Predictive processing as a systematic basis for identifying the neural correlates of consciousness. Philos. Mind Sci. 1, 3.

• Miller, M. et al. (2022) Predictive processing and consciousness. Rev. Philosophy & Psychology 13, 797-808.

• Seth, A.K. & Hohwy, J. (2021) Predictive processing as an empirical theory for consciousness science. Cogn. Neurosci. 12, 89-90.

• Klein, C. & Barron, A.B. (2016) Insects have the capacity for subjective experience. Anim. Sentience 1, 1.

• Merker, B. (2005) The liabilities of mobility: a selection pressure for the transition to consciousness in animal evolution. Conscious. Cogn. 14, 89-114.

• Pennartz, C.M. (2018) Consciousness, representation, action: the importance of being goal-directed. Trends Cogn. Sci. 22, 137-153.

• Dung, L. (2025) Understanding artificial agency. Philos. Q. 75, 450-472.

• Barandiaran, X.E. et al. (2009) Defining agency: individuality, normativity, asymmetry, and spatio-temporality in action. Adapt. Behav. 17, 367-386.

• Virenque, L. & Mossio, M. (2024) What is agency? A view from autonomy theory. Biol. Theory 19, 11-15.

• Kenton, Z. et al. (2023) Discovering agents. Artif. Intell. 322, 103963.

• Davidson, D. (2001) Essays on Actions and Events. Oxford University Press.

• Bratman, M. (1987) Intention, Plans and Practical Reason. CSLI Publications.

• Dretske, F. (1988) Explaining Behavior: Reasons in a World of Causes. MIT Press.

• Butlin, P. (2024) Reinforcement learning and artificial agency. Mind Lang. 39, 22-38.

• Mudrik, L. et al. (2025) Unpacking the complexities of consciousness: theories and reflections. Neurosci. Biobehav. Rev. 170, 106053.

• Rauker, T. et al. (2023) Toward transparent AI: a survey on interpreting the inner structures of deep neural networks. 2023 IEEE Conference on Secure and Trustworthy Machine Learning (SaTML), IEEE Computer Society, 464-483.

• Liu, D. et al. (2023) Attention schema in neural agents. arXiv. Published online May 27, 2023.

• Piefke, L. et al. (2024) Computational characterization of the role of an attention schema in controlling visuospatial attention. arXiv. Published online May 8, 2024.

• Devillers, B. et al. (2024) Semi-supervised multimodal representation learning through a global workspace. IEEE Trans. Neural Netw. Learn. Syst. 36, 7843-7857.

• Maytié, L. et al. (2024) Zero-shot cross-modal transfer of reinforcement learning policies through a global workspace. arXiv. Published online March 7, 2024.

• Richards, B.A. et al. (2019) A deep learning framework for neuroscience. Nat. Neurosci. 22, 1761-1770.

• Palminteri, S. & Wu, C.M. (2025) Beyond computational functionalism: the behavioral inference principle for machine consciousness. psyarXiv. Published online February 6, 2025.

• Metzinger, T. (2021) Artificial suffering: an argument for a global moratorium on synthetic phenomenology. J. Artif. Intell. Conscious. 8, 43-66.

• Long, R. et al. (2024) Taking AI welfare seriously. arXiv. Published online November 4, 2024.

• Bentham, J. (1789) An Introduction to the Principles of Morals and Legislation. Payne & Son.

• Shepherd, J. (2018) Consciousness and Moral Status. Taylor & Francis.

• Block, N. (1995) On a confusion about a function of consciousness. Behav. Brain Sci. 18, 227-247.

• Nagel, T. (1974) What is it like to be a bat? Philos. Rev. 83, 435-450.

• Schwitzgebel, E. (2016) Phenomenal consciousness, defined and defended as innocently as I can manage. J. Conscious. Stud. 23, 224-235.

• Bayne, T. & Montague, M. (2011) Cognitive Phenomenology. Oxford University Press.

• Simon, J.A. (2017) Vagueness and zombies: why 'phenomenally conscious' has no borderline cases. Philos. Stud. 174, 2105-2123.

• Schwitzgebel, E. (2023) Borderline consciousness, when it's neither determinately true nor determinately false that experience is present. Philos. Stud. 180, 3415-3439.

• Lee, A.Y. (2023) Degrees of Consciousness. Noûs 57, 553-575.

• Birch, J. et al. (2020) Dimensions of animal consciousness. Trends Cogn. Sci. 24, 789-801.

• Naccache, L. (2018) Why and how access consciousness can account for phenomenal consciousness. Philos. Trans. R. Soc. B Biol. Sci. 373, 20170357.

• Frankish, K. (2016) Illusionism as a theory of consciousness. J. Conscious. Stud. 23, 11-39.

• Searle, J. (2017) Biological naturalism. In The Blackwell Companion to Consciousness (Schneider & Velmans, eds), pp. 327-336, Blackwell.

• Schuman, C.D. et al. (2017) A survey of neuromorphic computing and neural networks in hardware. arXiv. Published online May 19, 2017.

• Albantakis, L. et al. (2023) Integrated information theory (IIT) 4.0: formulating the properties of phenomenal existence in physical terms. PLoS Comput. Biol. 19, e1011465.

• Tononi, G. & Koch, C. (2015) Consciousness: here, there and everywhere? Philos. Trans. R. Soc. B Biol. Sci. 370, 20140167.

• Birch, J. (2024) The Edge of Sentience: Risk and Precaution in Humans, Other Animals, and AI. Oxford University Press.

• Strathern, M. (1997) 'Improving ratings': audit in the British university system. Eur. Rev. 5, 305-321.

• Bender, E.M. & Koller, A. (2020) Climbing towards NLU: on meaning, form, and understanding in the age of data. Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics 58, 5185-5198.

• Schwitzgebel, E. & Pober, J. (2024) The Copernican argument for alien consciousness; the mimicry argument against robot consciousness. arXiv. Published online November 12, 2024.

• Dainton, B. (2024) Temporal consciousness. In The Stanford Encyclopedia of Philosophy (Zalta, E.N. & Nodelman, U., eds).

• Treisman, A. (2003) Consciousness and perceptual binding. In The Unity of Consciousness: Binding, Integration, and Dissociation (Cleeremans, A. & Frith, C., eds), pp. 95-113, Oxford University Press.

• Dennett, D.C. (1988) Quining qualia. In Consciousness in Contemporary Science (Marcel & Bisiach, eds), pp. 42-77, Oxford University Press.

• Lau, H. et al. (2022) The mnemonic basis of subjective experience. Nat. Rev. Psychol. 1, 479-488.

• Graziano, M.S. et al. (2020) Toward a standard model of consciousness: reconciling the attention schema, global workspace, higher-order thought, and illusionist theories. Cogn. Neuropsychol. 37, 155-172.

• Hohwy, J. (2013) The Predictive Mind. Oxford University Press.

• Clark, A. (2013) Whatever next? Predictive brains, situated agents, and the future of cognitive science. Behav. Brain Sci. 36, 181-204.

• Whyte, C.J. (2019) Integrating the global neuronal workspace into the framework of predictive processing: towards a working hypothesis. Conscious. Cogn. 73, 102763.

• Fleming, S.M. (2020) Awareness as inference in a higher-order state space. Neurosci. Conscious. 2020, niz020.

• Hurley, S.L. (1998) Consciousness in Action. Harvard University Press.

• Ginsburg, S. & Jablonka, E. (2019) The Evolution of the Sensitive Soul: Learning and the Origins of Consciousness. MIT Press.

• Clark, A. (2008) Supersizing the Mind: Embodiment, Action, and Cognitive Extension. Oxford University Press.

• McNamee, D. & Wolpert, D.M. (2019) Internal models in biological control. Annu. Rev. Control Robot. Auton. Syst. 2, 339-364.

Copyright: © 2026 Elsevier Inc. All rights reserved, including those for text and data mining, AI training, and similar technologies. For open access content, the relevant licensing terms apply.