Consciousness: what it is, where it comes from, do machines can have it and why do we care?

[Samson]

This is a fast moving time in the world of consciousness research. Modern tools are giving us insights impossible before, and the stakes are getting greater as AI smashes through the Turing test and threatens many jobs that previous had been considered the preserve of humans.

I have some snippets that I find interesting, but it seems one thing that is not touched on is why we care. It seems to me this question needs asking before we can start to choose between the different approaches. I do not know the answer, and I couldn't think of good enough answers to make a poll on this question. I will present one answer, feel free to suggest any others: Star Was, certainly TOS and TPS, droids were clearly both conscious and property. We should stop that happening, and if that means we give some taxpayer money to philosophers and ethicists now to stop Elon Musk building that world tomorrow then we should.

Moving on to the actual theories, Daniel C. Dennett does a good job of defining the problem away. My very inexpert summary is that consciousness is actually defined by agency, which requires both a biological basis, which computers do not have, and a understanding of the wider consequences of our actions, which dolphins do not have.

Spoiler I’ve Been Thinking Daniel C. Dennett :

Meaning, consciousness, and free will are entirely natural achievements, robust features of the manifest image that is the product of billions of years of natural selection. I can finally see how to display their main points in a few paragraphs, thanks to recent conversations I’ve been having with Keith Frankish, in Crete, off the coast of Greenland, and on Zoom. Think of a drone, remotely controlled by a (human) drone pilot. The user interface is brilliantly designed to make literally thousands of tiny adjustments to the motors and effectors that are conveniently beneath the notice of the pilot. That design work has to come from somewhere; it is the fruit of hundreds of thousands of hours of intelligent design by specialist engineers. The pilot takes advantage of the user interface provided, but the pilot’s know-how is also brilliantly designed, and that design work doesn’t come for free either; it takes thousands of hours of practice, much of it brute trial and error but also hastened by wise self-manipulation that is itself a product of special training. That training is conveyed to the pilot with the help of language, which is itself brilliantly designed to permit explanations to be shared.

Where does all this design come from? From evolution by natural selection, of several kinds: genetic, intracranial, cultural. First, life evolved and refined itself over several billion years of natural selection. Single-celled organisms—archaea and bacteria—solved the fundamental problem of reproduction, creating and optimizing the genetic code of DNA, the copy-machine ribosomes, the motor proteins, and other elegantly designed mechanisms. Then the eukaryotic revolution gave birth to specialized cells, of greater complexity, the first and most important instance of “technology transfer,” which multiplied the talents of single cells by orders of magnitude, permitting them to come together in multicellular organisms that could discover, through evolution in their own brains, still further good tricks for surviving and thriving in an ever more complicated world, leading eventually to a species, the well-named Homo sapiens, that wasn’t just competent but capable of (imperfect but growing) comprehension of the sources and explanations of its own competences. Our species was able not only to notice its own noticings but also eventually to analyze them and share the analyses with conspecifics, thanks to language. And how did language arise? It too is brilliantly designed and almost all of that design must have come from R & D processes over millions of years by agents who did not yet understand what they were doing and why. Cultural evolution by natural selection of memes added a faster, more bountiful generator of design, which gave human beings more degrees of freedom than were enjoyed by any other living things—along with the problem of how to control those degrees of freedom. Dealing with that problem turns into the problem of what to think about next, and that is what generates the stream of consciousness. There is no Boss, no Traffic Cop, to keep all the ideas in proper order. There is no general, algorithmic solution to this control problem because each innovation creates further degrees of freedom to cope with, and the natural result is a control architecture in human brains that is freewheeling, composed of competitive and collaborative semi-agents—“active symbols,” as Doug Hofstadter puts it generating our streams of consciousness. How do we know about this stream of consciousness? The question betrays the fundamental error that traces back at least to Descartes: what you are is not some separate soul or self, sitting in the Cartesian Theater, but that very control architecture, with the designed competence to tell others—and itself—some of the wonderful things it is doing. As Doug says, I Am a Strange Loop. Put the emphasis on “am”; not “I watch a strange loop” or “I experience a strange loop,” but “I am a strange loop.” Or as I put it in Consciousness Explained:

How do I get to know all about this? How come I can tell you all about what was going on in my head? The answer to the puzzle is simple: Because that is what I am. Because a knower and reporter of such things in such terms is what is me. My existence is explained by the fact that there are these capacities in this body. (p. 410)​

The first step the engineers would take to build a truly autonomous drone with onboard, not remote, control would be to throw away the LED screen with all the colors, since those were designed for an agent with eyes and color vision. There is no movie in your head; there are only the myriad discernments of content you rely on to control your expectations and actions. And then what about free will? It is an achievement, not a metaphysical feature, of normal human beings who have learned how to control their many degrees of freedom—and to keep others from remotely controlling them—by developing such habits of self-stimulation as imagining and reflecting on the outcomes of possible actions. Initiation into adulthood (securing the right to make legally binding promises and move freely around in the world) is itself a wonderful social mechanism and it too was designed by natural selection of memes, refined over thousands of years of civilization. All the glories of life depend on design, and designs can be destroyed or damaged; a brain tumor can destroy your free will, other brain pathologies can render you unconscious or unable to discern the meaning of the words you hear or say or read or write.

It takes vigilant thinking to avoid falling into the trap of imagining a Central Meaner that is the source of meaning, a Self that is the evaluator and enjoyer of consciousness, a Soul that defies physics and makes your decisions. These ideas are familiar and all but irresistible, and many thinkers find the prospect of abandoning them utterly repugnant. If you balk at relinquishing any of them, you are in good company, but just remind yourself that such ideas are all part of the largely benign user illusion that has been cobbled together over the eons by the various kinds of natural selection.

Life is what brings reasons and meaning into existence, and in our neck of the universe only human beings have the thinking tools—based on language—to figure this out. We are the only reasoners—in the strong sense of explainers—on the planet. Reasons predated reasoners by billions of years in the same way numbers predated mathematicians. It took two hydrogen atoms to unite with one oxygen atom to make a water molecule long before anybody could count, and the reason that living things have membranes isolating them (under controlled conditions) from the rest of the world is that living things need to protect themselves from succumbing to the second law of thermodynamics and can’t afford to protect the whole world, a reason that wasn’t appreciated until very, very recently in the history of life on this planet. Every living thing is composed of parts that are the way they are for reasons, but only we human beings represent reasons. Evolution has gifted us, and all living things, with brilliantly designed mechanisms that make life easier than it otherwise would be, but this gives living things competence without comprehension. Trees cope brilliantly under many conditions without a shred of comprehension; so do rabbits and foxes and elephants, who do many wise things without needing to know why these are wise things to do or when to do them. It is our human capacity to frame why-questions and evaluate candidate answers that sets us apart from the rest of nature, not some apparently magical soul that does the understanding and the feeling, the loving and fearing.

Isn’t this terribly anthropocentric? No, it is properly centered on what matters. Control is the key to life and everything that matters. The more things you can control, the more things can matter to you. Chimpanzees and dolphins are at risk from, but oblivious to, the problems of climate change, economic inequality, pollution, and war, but we alone can think that these problems matter enough to devote our energy to solving them. Noblesse oblige. Descartes was right to insist that human beings are profoundly different from other animals; he just made the mistake of trying to isolate that difference in an immaterial and unfathomable lump of mind-stuff. Reverse engineering that fantastic array of competences is the great intellectual adventure we are now embarked on, and it has been my extraordinary good fortune to be in the thick of it for my whole adult life.

Others seem to put more emphasis on the evolutionary aspects. From the nature review as I do not have access to these books:

Spoiler Free Agents by Kevin J. Mitchell and The Four Realms of Existence by Joseph E. LeDoux :

Kevin Mitchell and Joseph LeDoux apply similar evolutionary rationales to explain the emergence of consciousness and agency in their books. Mitchell is a geneticist and neuroscientist at Trinity College Dublin. His Free Agents devotes its first six chapters to an evolutionary account of the development of life and its various faculties. He argues that cognitive traits such as action, perception and choice started from very simple mechanisms that were selected for and honed to maximize fitness, or survival. From reading his book, one gets the strong impression that humans were forced by natural selection to be able to make choices and to become conscious agents.

At some point, he throws indeterminism into the mix. The Universe is not deterministic, he argues: it involves some degree of randomness, with events sometimes seemingly governed by the flip of a coin. The same is true of the brain, in his view. This indeterminism is adaptive, making humans less predictable and hence more able to survive and fight opponents.

Does such indeterminism by itself endow people with free will? No, says Mitchell: there’s nothing free in being governed by a coin flip. But indeterminism in an organism’s responses does allow it to have some influence on its future. The ability to create and express meaning is crucial here: it endows our reasons for doing things, and our reasoning about reasons (which Dennett also emphasizes), with causal power. Dennett wants to dispel the ‘illusion’ of self, but for Mitchell, the self, with all its goals, desires and beliefs, is real, and key to our free will. Together with the meaningfulness of the patterns of our neural activity, it allows us to exert top-down control, to plan ahead and to continuously shape ourselves as we interact with the world. For Mitchell, such conscious, rational control of our actions is nothing other than our free will. It is a biological, evolved function — as Dennett argues too.

LeDoux agrees. In The Four Realms of Existence LeDoux, a neuroscientist at New York University, suggests that there are four basic varieties of life on Earth: biological, neurobiological, cognitive and conscious. The book provides an in-depth description of these realms (I found the cognitive one especially thought-provoking) and describes how they evolved, in a way that is reminiscent of Mitchell’s approach. In this scheme, most living things occupy only the biological realm. Organisms with nervous systems are also neurobiological. Of these, some animals show model-based behaviour — using past experience to predict the future effects of their actions, and in doing so optimizing outcomes. These count as cognitive creatures.

The fourth and least common realm is the conscious one. LeDoux values the ability to verbally report the content of experiences as the prime indicator of consciousness, a position that is not shared by all. He emphasizes the importance of activity in the prefrontal cortex in allowing the creation of higher-order states that re-present the content of experience (although the special role of this brain area again is debated).

LeDoux further differentiates between types of consciousness, ranging from simpler forms to the explicit, content-rich type that humans have. He argues that we should aim to connect each type of consciousness with a different prefrontal brain architecture, and judge claims of animal consciousness on that basis. For example, because all mammals share the same mesocortical prefrontal areas, they might have “whatever kind of consciousness these areas enable in humans”. However, some prefrontal brain structures are unique to humans, arguably endowing us (and possibly some other great apes) with some unusual aspects of consciousness, such as the ability for mental time travel, that are not shared with other animals.

These are all interesting arguments, but they seem to share the same problem that I identified at the start, if you do not know what you are trying to do then you do not know what features are important. If we are trying to ask "what artificial consciousness deserves rights" then some are going to explicitly exclude it. LeDoux defines away the possibility, arguing that consciousness can exist only in biological beings. Dennett says an AI it is not an agent with beliefs and desires, or, in Dennett’s words, an intentional system (yet).

However Mitchell's Free Agents provides a ‘recipe’ for creating artificial systems that resemble humans, that have general intelligence and agency. It is to follow the evolutionary trajectory that got us here: embodiment, sensing, acting, with some motivation and learning abilities, and a drop of indeterminacy. The nature articles author has done a review of the current literatures opinion on the potential indicators of consciousness and applied artificial systems, and found that while current AI systems fail to meet these criteria there are no technical barriers for building a system that will satisfy them.

This at least gives us something to work with, and I can see some sort of implicit "we are conscious so we should protect things like us" logic (though I have not seen anything like that put) I am still not really sure why. If some other entity come to the same functionality by some other means, so not involving biology, evolution or global workspace theory or whatever neural structures gives us consciousness, why would that be any less worth of recognition or protection?

References:

This thread takes much of the title and content from this nature review.
Nature review on end of bet between philosopher and neuroscientist
New Scientist on bet between philosopher and neuroscientist
Nature review on the Turing test and the next steps
Arxiv review on AI consciousness

 

[Gorbles]

The core problem in the definition in my opinion will always come back to internal bias r.e. justification. We've essentially anthropomorphised it as a concept, whereby (similar to what you're saying about not knowing what features are important) we self-insert based on how we understand consciousness. And the main reason we struggle with this is because in a practical way it ties back to ethical considerations (similar to "sentience", with various arguments having some overlap between the two, or at least a relationship between the two states). So it's kind of a perverse incentive from the offset? Even if started with good intentions, the research inevitably bends to "and therefore at what level does this have consequences".

 

[Farm Boy]

I can see some sort of implicit "we are conscious so we should protect things like us" logic (though I have not seen anything like that put) I am still not really sure why.

Because we don't protect things merely because they are conscious. In fact, our greatest dangers are generally perceived as other conscious beings, and that perception isn't a crazy threat-assessment.

But I think you're trying to loop in a concept of sacred. And while an artificial consciousness isn't likely to have much similarity to us, overall, you don't need uniformity to be a sacred thing.

 

[Birdjaguar]

Consciousness is on a continuum of complexity. Certainly all living things have some degree of consciousness (awareness of self and surroundings: short version). I would personally extend that continuum back into the world of the non living things.

 

[Klaus Hergersheimer]

No, I don't care whether something has a consciousness or not. Rather I respect it because that's what mature people do.

Like, I could go kick and break rocks around when I go on hikes, but I don't because 1) that makes me look stupid and 2) it's wasteful of my time and energy. I respect rocks for what they are.

This is the whole problem I have with talks about "animal rights", too. As if some social movement has to be generated around something people used to call being an adult, being conscientious, being a good steward of the earth, being a lady or gentleman, knowing your actions have consequences, etc.

 

[EvaDK]

Consciousness manifests itself when certain biological parameters in the development of the brain are met - as far as we can tell. It then evolves into something entirely unique without the interference of an outside source.

That's why a similar proces will likely never take place in some form of artificial being, machine or technology of our creation. Everything I've seen or read about, is basically just a programmed set of instructions that are designed to mimic intelligence/consciousness/self awareness/human behavior. Thousands of years of philosophical debate hasn't really contributed much to our understanding of ourselves either.

 

[bernie14]

Sure. AI just needs to develop some narcissistic traits and there you go....

 

[Broken_Erika]

Sure. AI just needs to develop some narcissistic traits and there you go....

The Future of AI

 

[Birdjaguar]

Consciousness manifests itself when certain biological parameters in the development of the brain are met - as far as we can tell. It then evolves into something entirely unique without the interference of an outside source.

That's why a similar proces will likely never take place in some form of artificial being, machine or technology of our creation. Everything I've seen or read about, is basically just a programmed set of instructions that are designed to mimic intelligence/consciousness/self awareness/human behavior. Thousands of years of philosophical debate hasn't really contributed much to our understanding of ourselves either.

So unmanifested consciousness is a state? could the difference between manifested and unmanifested consciousness be connected to our limited ability notice it?

 

[Birdjaguar]

This at least gives us something to work with, and I can see some sort of implicit "we are conscious so we should protect things like us" logic (though I have not seen anything like that put) I am still not really sure why. If some other entity come to the same functionality by some other means, so not involving biology, evolution or global workspace theory or whatever neural structures gives us consciousness, why would that be any less worth of recognition or protection?

Defining consciousness around ourselves (humans) is kinda like saying "white males like me are superior to all others." What happens if you treat consciousness as a feature of the universe?

 

[Thorgalaeg]

I think it is all a matter of complexity, at enough complexity the things we use to identify with consciousness begin to manifest. Computers are still very, very, very far away of reaching the required complexity level. Read somewhere a fruit fly brain with only about 200,000 neurons has more computational power and is more complex than any CPU.

 

[Tee Kay]

Thinking about consciousness makes me not want to be conscious so I try not to think about it.

 

[Farm Boy]

Thousands of years of philosophical debate hasn't really contributed much to our understanding of ourselves either.

Oh, I don't know. But it does reset with every new iteration and the more finished products are constantly lost.

 

[Birdjaguar]

Oh, I don't know. But it does reset with every new iteration and the more finished products are constantly lost.

What does that mean? You lost me.

 

[Farm Boy]

We constantly lose the type of understanding that I think we're talking about. About 1.93 people every second.

 

[Birdjaguar]

We constantly lose the type of understanding that I think we're talking about. About 1.93 people every second.

So our consciousness is collective as well as individual? I can go with that. But if we are losing 1.93 people every second, how many are we gaining? Does quantity matter at the individual level or is such collective consciousness fixed and independent of head count? Do critters count in your view?

 

[Farm Boy]

I thought Eva was talking about what understanding philosophical debate yields.

But yes, knowledge leaks between people, stuff can be taught. One of our greatest strengths, that. People die for each other, willingly, and they die anyways. We're a big multicellular bioworm writhing through the ages. But we lose irreplaceable units with unique understandings that cannot be replicated accurately and cannot be taught with reliable accuracy. That can only really be done by showing, through sharing a life, and still the results are deliciously variable. The replacements each need to build their own houses of understanding from scratch. Hopefully, with some help.

 

[EvaDK]

So unmanifested consciousness is a state? could the difference between manifested and unmanifested consciousness be connected to our limited ability notice it?

Well, that's not what I meant to say ; I guess it depends on your definition of the word 'manifest'. I meant to say that consciousness likely 'sparks into existence' in the early stage of brain development. Does consciousness exist prior to this incident/spark in some other state (if that's what your asking about) - I don't think so, but obviously I can't prove either way. No one can.

As for our perception in detecting consciousness in places other than in ourselves; that's also a philosophical debate that has been ongoing since ancient times. A tree doesn't exhibit any behavior that suggests it has consciousness; yet no one can scientifically prove that it doesn't. You can construct the most compelling 10,000 word argument ever formulated, that trees indeed have consciousness - and in response, I can put on my well worn bathrobe and slippers and quote The Dude 'Yeah, well, that's just, like, your opinion, man." Where to go from there? Nowhere.

I thought Eva was talking about what understanding philosophical debate yields.

The Achilles heel of philosophy and philosophical debate, is that it is - almost - entirely based on subjective opinions, experiences and thought. Natural sciences are occupied with explaining the natural world and describe how it works. Philosophy is more occupied with studying the human experience and ask questions about existence, meaning, reason, morality, knowledge and so on. Natural science is about understanding the Universe, of which we are just a tiny speck; Philosophy is about understanding ourselves.

I'm not bashing philosophy here; the questions asked are perfectly valid. But the Universe can never answer them for us; we are stuck in our subjective experience every single one of us and we will likely formulate our own answers to those questions, answers that are only valid to you.

 

[Birdjaguar]

I guess it depends on your definition of the word 'manifest'. I meant to say that consciousness likely 'sparks into existence' in the early stage of brain development. Does consciousness exist prior to this incident/spark in some other state (if that's what your asking about) - I don't think so, but obviously I can't prove either way. No one can.

I see. Some trigger point gets reached. Do you think any animals have consciousness?

 

[Farm Boy]

I'm not bashing philosophy here; the questions asked are perfectly valid. But the Universe can never answer them for us; we are stuck in our subjective experience every single one of us and we will likely formulate our own answers to those questions, answers that are only valid to you.

Why is it an Achilles heel, then?

The answers will not be identical like people are not identical, but given there at fields with shared understandings and expertise to be had, I would not say the answers being entirely subjective and bespoke to the individual is fully true. There are harmonies to be had. Just that there is a hell of a lot of qualitative to be had in this universe, which is beyond the scope of natural sciences.