I recently stumbled upon a unique theory of consciousness that is inspired by the way computation works when prompt engineering on ChatGPT. I decided to write a piece exploring the theory just for fun. My theory seeks to challenge the age old hard problem of consciousness. I would love for you to read and share your thoughts.
As a preview before any explanation, my theory of consciousness is as follows:
The brain generates conscious experience to offer more time for cognition to promote survival. What is included in consciousness is what the brain determines is most meaningful for survival, and the brain’s own meta-assessment of meaning creates conscious subjective experience.
I know that is a lot, so I hope to provide some context on the question and then explain.
Here are the contents:
Part 1: What is the “Hard Problem of Consciousness”?
Part 2: Philosophical Theories of Consciousness
Part 3: Scientific Theories of Consciousness
Part 4: Experimental Quirks of Consciousness
Part 5: My Theory: Layered Cognition Theory (LCT)
If you’re already familiar with the question and context, please skip to part 5. And if you’re not, feel free to read Parts 1 through 4 to build a solid foundation for the question. However, Part 5 is the focus of this article, and I recommend giving it your most focused attention.
Part 1: What is the “Hard Problem of Consciousness”?
Imagine you’re sitting in a room, and in front of you is a ripe, delicious-looking red apple. You reach out, take a bite, and instantly, there’s a burst of sweetness, a crisp texture, and the bright beautiful red that fills your vision. Have you ever wondered what turns these physical interactions into the vivid experience of sweetness, crunchiness, or “redness”?
This question lies at the heart of what’s known as the “Hard Problem of Consciousness”, coined by philosopher David Chalmers in the 1990s. This isn’t a question of how your eyes detect color, or how your taste buds recognize sweetness. The mechanisms are well understood. The hard problem dives deeper, asking: Why do these processes feel like something? Why and how do physical actions in the brain give rise to the subjective experience of seeing a color, tasting a flavor, or feeling happy, sad, or in love?
Our brain does so much outside of our field of consciousness. It controls our heart rate, respiration, and internal body temperature. There are 500 million neurons in our gut lining that autonomously optimise food digestion. You are presently completing complex calculations on how quickly you should breathe to maintain optimal oxygen and carbon dioxide in your blood, which is crucial for your survival, yet you are unaware of it. So why are we aware of the things we are aware of? And, why does our brain not respond to stimuli without creating this intense conscious experience? Does everyone else have the experience that I have? Or is everyone around me an unconscious zombie? Can I ever truly know? I guess, as Descartes put it, the only thing we can ever truly know is that we are conscious: 'I think, therefore I am.'
This is the hard problem. Understanding the meaning behind what it is to have a subjective experience - the rich, personal tapestry of sensations, emotions, and thoughts that make up our subjective experience.
Part 2: Philosophical Theories of Consciousness
Throughout history, philosophical inquiries into consciousness have generally been categorized into three primary schools of thought:
Dualism: Some suggest consciousness transcends mere physical explanations, advocating for a fundamental distinction between mind and matter. This viewpoint argues that the essence of consciousness encompasses elements beyond the physical realm.
Panpsychism: Others suggest that consciousness is a universal feature of all matter, and the hard problem lies in understanding how complex conscious experiences emerge from simpler forms of consciousness present in smaller entities.
Physicalism/Materialism: Some argue that consciousness can be fully explained by neuroscience and physical laws, suggesting that the hard problem will eventually be solved with more understanding of the brain's workings.
My thoughts on these perspectives:
Dualism holds a long history prominent within major religions. The idea is that our mind is a spirit that is separate from our bodies. Some feel as though consciousness or our inner spirit feels so unique that it cannot be made of the same things as matter, but to me, this lacks imagination. Humans have a long history of attributing things we don’t understand to non-physical elements - whether it be the movement of stars, clouds, or miracles. In my view, reality is often just very complex and hard to understand and Dualism is not the answer.
Panpsychism is attractive to the free-loving hippy inside us all. If everything that exists in our universe is an emergence of consciousness, our experience is just another drop in the ocean of consciousness, as spoken by Alan Watts. While it is definitely a beautiful idea and still may be sort-of be true, to me, it doesn’t help us make sense of the nature of consciousness as we experience it. It begins to feel slightly meaningless what consciousness is with this perspective. If we believe our kidneys and rocks, are both emergence out of ‘consciousness’, it becomes more just the ‘essence of life’ and becomes quite divorced from understanding human-like consciousness.
Physicalism is the idea that when a baseball hits you on the head, it’s truly lights out. Consciousness is a function specific to the brain, and although it’s hard to find, it’s there. As you will see in what follows, while initially it is difficult to concieve, it is likely the answer.
Part 3: Scientific Theories of Consciousness
If we run with the idea of Physicalism, it leads us to examine where we see consciousness within the brain. We know that the brain is made up of around 100 billion neurons, which operate like a massive biological computer. All cognition arises from complex networks and we have learned that different sections of the brain are responsible for different things. If a neurosurgeon sticks their finger into one section of the brain, they may make a person sing.
If we look for consciousness, it can be tricky to locate using fMRIs because so much of the brain is active during waking awareness. The whole neocortex shown below is where higher-order thinking is present, and this area is commonly associated with consciousness.
The back half of the neocortex (the Occipital lobe) is responsible for much of the signal processing of the brain and is very active during conscious experience as it is responsible for processing vision, so therefore likely has much to do with conscious experience. However, many leading “higher-order theories” today suggest that there is a relationship between the signal-processing part of the brain and the conscious higher-order processing of information.
As we shift specifically into scientific theories of consciousness, there are two primary theories worth discussing:
Integrated Information Theory (IIT): Proposed by Giulio Tononi in 2004, IIT suggests that consciousness arises from the integration of information across various parts of the brain. The theory posits that the more integrated the information, the higher the level of consciousness. This could explain why different states of consciousness (like being awake, asleep, or under anesthesia) have different levels of information integration. This theory implies that consciousness is related to how integrated information processing is, and it may suggest that other forms of consciousness can stem from non-biological information processing, like silicon-based artificial intelligence.
Global Workspace Theory (GWT): Bernard Baars's GWT posits that consciousness is like a bright spotlight, illuminating certain mental processes for cognitive tasks. This theory suggests that when information is brought into the "global workspace" of the brain, it becomes conscious. This could explain why we are only conscious of a fraction of our brain's activities – only the ones that are "on stage" in the global workspace.
A nice way of thinking about GWT is that when something enters your field of consciousness, there is a wave of neurons firing through your brain through the most interconnected section of the brain. When this wave is not ignited, we are not conscious.
In my view, GWT is the best explanation of the mechanism of consciousness but it does not answer the hard problem of why we have a subjective experience.
Part 4: Experimental Quirks of Consciousness
Before we go into my theory, I want to highlight some of the most fascinating experiments that explore the quirks of our conscious experience. Feel free to read these or skim through them lightly. I suspect learning about the following experiments will change the way you view your thoughts.
A) Optical Illusions:
Let’s get started with a few fun optical illusions:
^Illusion 1: These lines are the same length without the arrow heads.
^Illusion 2: The square that both A & B are on is the same colour. You can only tell when you cover everything except the two squares.
^Illusion 3: A static image that appears to be moving because our brain is being tricked into thinking there is motion because of the way our brain processes the image.
Our brain is constantly reprogramming the incoming images to quickly make sense of them. We cannot stop it. As is demonstrated by the following.
B) The McGurk Effect
One of the most incredible experiments relevant to consciousness. A must watch. It’s incredible how quickly the sound is processed differently by the brain.
C) Temporal Binding
When you watch a ball being struck, the visual information reaches your brain quicker than the sound of the impact. Our brains have this incredible ability to sync these sensory inputs, making them appear simultaneous.
In one study, participants are tasked with a simple action: pressing a button to turn on a light. Initially, this action and the light's response are simultaneous. However, as the experiment progresses, a slight delay is introduced between the button press and the light illumination, eventually reaching up to 100 milliseconds. Throughout this process, the brain adapts to this delay, recalibrating its perception so that the action and the sensory feedback continue to feel instantaneous.
When the delay is suddenly removed, participants report that they perceive the light turning on before they even press the button. This not only highlights the brain's adaptive capabilities in synchronizing sensory information from different modalities but also illustrates how susceptible our perception is to being misled.
D) The Selective Attention Test
This experiment is important for my explanation later. Sight is not all that consciousness is. There is so much visual information that remains outside of consciousness.
Here’s where it starts to get funky…
E) Subliminal Conditioning
In another study, participants are exposed to a blue square flashed briefly before them—so fleetingly that they are not consciously aware of having seen it. Shortly after the blue square appears, they receive a mild electric shock.
After several repetitions of this sequence, the brief presentation of the blue square begins to trigger a physiological stress response in anticipation of the shock, even though the participants are unaware that they saw any blue square.
When subsequently asked if they experienced any physiological changes upon subconsciously viewing the blue square, participants often attribute any sense of stress they feel to unrelated concerns, such as unfinished tasks or personal responsibilities, rather than the subliminal exposure to the square.
In my view, this experiment has profound implications for not only the nature of our consciousness but also the way that we rationalise and interpret our emotions.
F) Split Brain Experiments
During the 1950s, experimenters began slicing the Corpus Callosum in human patients who suffered from untreatable epilepsy. This part of the brain acts as a highway that communicates between the two hemispheres of the brain.
In these experiments, researchers discovered that when different visual information was presented simultaneously to each hemisphere, the participants exhibited fascinating responses. For instance, an image shown only to the right hemisphere eye could not be verbalized by the participant because the speech centers are predominantly located in the left hemisphere. However, the participant could select an object related to the image with the hand that is controlled by the right hemisphere. And if you show the object to the left hemisphere of the brain, the person has no problem articulating what they see.
Additionally, there are many more wild tales such as when a patient’s left hand would physically wrestle against the actions of their right hand, or if you ask them a question and have them write the answer with each hand, you get wildly different answers. For example, when a patient with a split brain is asked to write down what job they want, the hand controlled by the right side of the brain, which is associated with creativity and spatial abilities, would write things like "artist," while the left side of the brain, which is responsible for language and analytical tasks, would write "accountant."
This highlights an incredibly mysterious element to consciousness and our sense of self. It seems as if consciousness is divisible. Could this suggest that there is another conscious being in the back of our head that has different hopes and dreams, but is simply trapped in the subconscious?
Part 5: My Theory: Layered Cognition Theory (LCT)
With all these different experiments and ideas swirling around in my mind, I have often reflected on the Hard Problem and shifted back and forth, concluding at points that it is innate in all information processing, and the more information processing that occurs, the more conscious something is. This resonated with me because I quite liked the idea that we are no different from the information processing of a star being born and that robots may one day be conscious like us.
However, the idea that information processing equals consciousness doesn’t make sense because there are incredibly dense flows of neural activity that live within our subconscious. Those areas are either themselves separately conscious to us, or there is a separation between processing and consciousness. Additionally, this doesn’t explain so many of the experimental quirks of consciousness.
When you consider the quirks discussed above, it feels like consciousness has a purpose… a goal. It feels like it’s trying to help us to interpret the world more clearly to increase our chances of survival. But why? Why not let our body respond blankly to the raw data flowing into the brain without the subjective experience? Hold this thought.
I use AI/ChatGPT a lot. I use it to help me write, think, or learn many times each day. I have begun to optimise the way I use ChatGPT using a concept called “prompt engineering”. Here is a common tip you will hear:
“Sometimes AI gets caught up on simple questions like “What is the fifth word in this sentence?” and often fails. If you ask the AI to write a plan on how it will answer your question before it answers your question it will typically provide a much higher quality result.”
The reason for this is that the AI uses a specific amount of compute power for the prediction of each next word. When you allow it to spread out its computation over more characters, it is offered the opportunity to extend its cognition and improve output. And when I was using AI while reading about the Hard Problem, something clicked.
Now let’s take a thought experiment step-by-step (almost through the eons of evolution):
Imagine that we are a simpler life form where our brain causes the body to act without a unified cognitive response. One small section of the brain processes sensory information and then immediately fires a response, just like a reflex, without needing to be processed by the whole brain.
A clear enhancement that evolution may evoke would be for the response to stimuli to require input from all sections of the brain, assuming that a quick response was not required. Instead of a small section of the brain initiating a response, a superior response would arise from the brain as a unified whole. Under fMRI, this would likely appear as some kind of ‘neural wave’ as the signal passes through all areas of the brain. This is how all animal brains operate today: a signal moves through neural networks within the brain. Imagine, instead of one senator commenting on a bill, every senator is allowed to comment, resulting in far higher quality output.
Now if we’re to step up another evolutionary notch, imagine that you have a neural wave that passes through every section of the brain (as above), but you also have a custom-designed area of the brain for the neural wave to pass through, almost like a ‘neural highway’. In humans, this would likely represent the most interconnected section of the brain’s neocortex (as GWT suggests above). This highway connects your neural wave to every important section of the brain, and each of those sections of the brain has been given the right sensory information to allow them to contribute efficiently. This means the brain can provide a high-quality response quickly, with input from every area of the brain without a slow wave that travels around the whole brain.
Now, if evolution wanted to take it a step further, it might connect this highway with an idea. This idea is that the brain sits atop something called the ‘self’. That is, the brain creates an idea that all the biological mechanisms connected to it are unified. They are all ‘one body’. It processes all relevant information but places it in the context that a ‘self’ exists in space and time, and this helps cognition to be optimised for the survival of the ‘self’. This ‘self’ is an idea and it represents all biological body parts that are connected to the brain and are relevant for survival.
One step further: let’s take this neural highway and upgrade it a bit with something I will call a ‘replay system’. As all the incoming sensory information flows along the neural highway, there is an instant replay system built in to review all incoming information. The replay allows the brain to have extra time to process and respond to the input (like ChatGPT above). It allows the information to become more integrated and to more accurately represent information that is important for survival.
By quickly interpreting the external world, which is relevant for survival, the replay system more accurately interprets input and extends the time of cognition to respond to external stimuli, thereby increasing the chances of survival of the body attached to the brain. However, it doesn’t just replay all visual sensory input. It integrates different sensory inputs into a single flow of information and cleans the ‘data’ as it replays it. This cleverly improves our perception of reality (like Temporal Binding above), and sometimes results in bizarre hallucination-like effects where the input is slightly distorted (like the McGurk Effect, and optical illusions).
Take a moment to note what is included in consciousness. While your eyes are constantly processing incredible amounts of visual information, not all of it is conscious (as per the Selective Attention Test above). You only become conscious of the parts of your visual field that your brain concludes are most relevant for your survival, and it centers your conscious experience around that information. Generally, the present moment is very relevant for the brain’s continued survival, so it is constantly replaying our life moment-by-moment to extend cognition on what is immediately before us. However, not always: sometimes at night, when your eyes are closed, visual information is no longer as relevant for your survival so you are not conscious of the present moment. Instead, you are replaying awkward social interactions or stressing about all the work you need to do, which at that moment feels more important to reflect on for survival.
This is something the brain is working out in every moment. It is sifting through sensory input and selecting that which is most meaningful for survival in that moment. The brain is continually calculating what should be included in your conscious experience based on its judgment of what is most important for survival, and only when you’re in a place of safety will your brain diverge from anything other than your present moment. If you’re skiing quickly down a mountain, you become engulfed in the present moment as your brain completely focuses cognition on your immediate risks, because it determines focusing on avoiding trees is most meaningful for survival. And in this moment right now, do you feel the sensation of your teeth in your mouth or your feet inside your shoes? These were excluded until I mentioned them because they were not considered essential for survival at this moment according to your brain.
We are not quite finished…
Let’s now imagine that this brain has all the features we described above: a custom neural highway, with an in-built replay system for extending cognition, and a system continually calculating what is most meaningful for the survival of the ‘self’. And then evolution creates a version of this replay system but now attaches a sense of meaning to the replay itself. It builds in another layer on top of the replay mechanism that is focused on assessing the meaning of the contents of the replay, which is also included in the replay. Not only does it run a replay of the present moment, but it runs a continual replay of the brain’s own meta-assessment of what within that replay has meaning. The brain is building in a meta-assessment of the meaningfulness of our present moment as part of our replay function in order to extend cognition of that meta-assessment, with the hopes of increasing our chances of the survival of the self. This meta-assessment is combined with the concept of the ‘self’ to give us this sense that every moment of human awareness holds some degree of value. This perception is ‘consciousness’ or ‘subjective experience’.
The fact that our brain assigns a level of meaning to every moment of our conscious awareness gives us the sense that it feels like something to be a human. What does it mean to feel? To feel is to see an apple as red, and not only see the redness but to attribute meaning to the seeing of the redness itself (all while maintaining a sense of ‘self’). We don’t just see redness, we feel the sight of it has meaning. And every moment of our life has a sense of feeling to it. This is what the hard problem cannot explain: why it feels like something to be human. I would argue that evolution has created this sense of “meaning” or “feeling” or “subjective experience” or “consciousness” as part of the layers of cognition in the neocortex of the brain developed to promote survival.
To hopefully paint the overall picture more clearly, here are the layers of cognition:
Level 1: The brain processes sensory information that the apple is red in the occipital lobe (rear of the brain) and spreads this information across the brain.
Level 2: The brain selects the sight of the red apple to be brought into consciousness (into the replay mechanism of the neural highway in the neocortex) to give more time for cognition (while maintaining a sense of self) because it concludes the redness is meaningful for survival.
Level 3: Within the replay mechanism of the neocortex, the brain then assigns a level of meaning to the sight of the redness itself as part of its mechanism to assess its level of meaning. This meta-assessment is included in the experience of consciousness itself to give more time for cognition (which of course, is also all done while maintaining a sense of self), and it gives us the experience of feeling.
To very clearly articulate a key point: I argue that to feel something is the same as the brain perceiving a moment as holding meaning. It’s not always perceived as meaningful by the ‘self’ observing, but the moment in time carries weight because that experience is relevant for survival, and gives us the sense of ‘feeling’. The moment of seeing the redness of the apple itself carries weight. If that moment wasn’t relevant or important for your survival, it would be as if you weren’t really conscious in that moment, and you didn’t really feel anything.
(You may also argue that there is a level 4, for all the moments in our lives that we spend consciously assessing the meaning of moments in our lives (this is getting inception-level trippy), but consciousness is already achieved on level 3.)
The brain’s attribution of meaning to the act of seeing itself (still while maintaining a sense of ‘self’) is consciousness. It is seeing with the eyes of a person and feeling the experience of being a person seeing a piece of fruit, and believing that experience is meaningful. Sounds like life, right?
This theory helps us understand every moment of our waking consciousness, and why every thought in our awareness is there, but it’s also important to remember that our brain is a clunky biological machine, and it does this all imperfectly. Little tidbits of our dreams slip through into our waking life for a brief moment after we wake up. There is some residue in our consciousness after being knocked out, which is sometimes one of darkness or a mysterious dreamland. Our near-death experiences are the brain backflipping through trippy visuals as it tries to keep you alive. Or we see something that has an eerie feel like something previously, and we experience a sense of déjà vu. Our conscious awareness is not cleanly cut from other biological mechanisms. There is a chaotic firing of signals as it tries to interpret the world accurately. And our consciousness sometimes makes mistakes as it collects information on our reality, but it is flexible, agile, and everchanging, forever focused on becoming more optimised for our survival. Sometimes it feels as if our consciousness changes as our perception of meaning slowly shifts over the years, or that other people have drastically different experiences, and with this theory, all of this makes sense.
Sometimes the replay system itself even breaks and the replay is of things separate from the information input, leading people into psychosis. Sometimes the replay becomes oversensitive to details holding meaning, like with obsessive-compulsive disorder (OCD). Or sometimes, it fails to optimise which input holds meaning to replay, like with attention deficit disorder (ADD). Leveraging this theory, we have an opportunity to experimentally validate the cause of neurological issues as our understanding of neural circuitry deepens. Additionally, this has implications for our concept of meaning itself. What if any sense of meaning is nothing but a mechanism for optimising cognition? The implications of this question are for another time.
In conclusion, we have beautiful evolution-optimised minds and it feels like we can choose what we believe to be meaningful. And that which we find meaningful is what our consciousness largely becomes, so we should choose wisely.
Bonus: Does this answer the hard problem?
Well, many people think no. However, I feel satisfied. Let me explain why by engaging with the main criticisms.
As a recap, the hard problem is essentially seeking to answer the why and how of conscious experience. Fundamentally, people argue the experience of being a human is so unbelievably magical that there must be a supernatural explanation, however, there is another cohort of individuals that points to all the evidence stacking up of consciousness arising within the brain. The common response is: “We can understand how sensory information is processed physically by the brain, but we cannot understand why it “feels” like something to be human”.
In my theory, I argue that the experience of “feeling” is the attribution of meaning to experiences. I have sought to explain why the brain attributes meaning to every moment, and I have sought to dispel this common criticism of the physicalist interpretation of consciousness.
Some will still say, “Well, you have explained why evolution has generated a conscious experience, but you haven’t answered how ‘qualia’ can arise from biological tissue.” “Qualia” is a word I have avoided, which is commonly raised in this discussion and philosophers use to represent ‘units’ of conscious experience. I think it drives much more confusion than clarity. It holds implicit that conscious experience is more than meets the eye.
Let me share an analogy. One time my grandma walked in and observed my brother and I playing FIFA on Xbox (a soccer video game). She could not comprehend that the game was being controlled by us. She thought we were lying. The graphics were too real. She could not believe that they were being generated in a game, it was too much for her to bear.
Trying to explain how this game is possible is difficult. It requires you to go into wildly complex electronics and theories of computation that honestly are too complex for me to process, let alone my grandma. I also struggle to comprehend how humans invented electronic machines that allowed me to control my favourite soccer player on pixels on a screen and score epic goals from outside the 18-yard box.
I believe that conscious experience is like this. It is difficult for us to comprehend how a complex biological computer sitting upon our neck could create something so magical. We give it labels like ‘qualia’ to quantify the magic, but it just smothers the underlying complexity. While taking a class in neuro-engineering at MIT with Ed Boyden, he explained to us the endless complexity of a single neuron, the hundreds of different types, and the way computation in a single neuron can loop back on itself, sometimes send signals in reverse, and even shift around its connections in just a moment. And we have 100 billion of these within our heads forming endlessly complex networks that are different in every human.
I don’t believe that consciousness requires an explanation outside of physicality. I think it requires understanding the mystifying complexity of the brain.
This leads to the next and final criticism: “the zombie problem”. Some will argue the only way to solve the hard problem of consciousness is to be able to answer with certainty whether other humans around us are truly conscious. I believe that answering this question is similar to answering the question of what is at the core of the Earth. This is a question that would have seemed completely impossible a thousand years ago, but with time, our grasp of geology deepened and we now feel confident of our answer even though we can’t see it firsthand. With solid theoretical models and as we more deeply understand the brain, we will build up empirical evidence for how the mind creates consciousness and eventually be able to confidently conclude whether our peers or pets are conscious, but this will all rely on inference and evidence-based theories.
Bonus Part 2: Can AI ever be conscious?
I have thoughts. Coming soon.
I would love for others to sincerely engage with my thoughts and connect experimental results to support or challenge my suggestions as I have alas left much research unread. I would love to hear your thoughts, comments, or questions. Here is my email. Thanks for reading.