computer science

AI and the “generalization faculty”

PNRJ cognitive science, computer science, critique of neoclassical economics, current events, ethics, futurism, inequality, public policy , , , , , , , , , ,

Oct 1 JDN 2460219

The phrase “artificial intelligence” (AI) has now become so diluted by overuse that we needed to invent a new term for its original meaning. That term is now “artificial general intelligence” (AGI). In the 1950s, AI meant the hypothetical possibility of creating artificial minds—machines that could genuinely think and even feel like people. Now it means… pathing algorithms in video games and chatbots? The goalposts seem to have moved a bit.

It seems that AGI has always been 20 years away. It was 20 years away 50 years ago, and it will probably be 20 years away 50 years from now. Someday it will really be 20 years away, and then, 20 years after that, it will actually happen—but I doubt I’ll live to see it. (XKCD also offers some insight here: “It has not been conclusively proven impossible.”)

We make many genuine advances in computer technology and software, which have profound effects—both good and bad—on our lives, but the dream of making a person out of silicon always seems to drift ever further into the distance, like a mirage on the desert sand.

Why is this? Why do so many people—even, perhaps especially,experts in the field—keep thinking that we are on the verge of this seminal, earth-shattering breakthrough, and ending up wrong—over, and over, and over again? How do such obviously smart people keep making the same mistake?

I think it may be because, all along, we have been laboring under the tacit assumption of a generalization faculty.

What do I mean by that? By “generalization faculty”, I mean some hypothetical mental capacity that allows you to generalize your knowledge and skills across different domains, so that once you get good at one thing, it also makes you good at other things.

This certainly seems to be how humans think, at least some of the time: Someone who is very good at chess is likely also pretty good at go, and someone who can drive a motorcycle can probably also drive a car. An artist who is good at portraits is probably not bad at landscapes. Human beings are, in fact, able to generalize, at least sometimes.

But I think the mistake lies in imagining that there is just one thing that makes us good at generalizing: Just one piece of hardware or software that allows you to carry over skills from any domain to any other. This is the “generalization faculty”—the imagined faculty that I think we do not have, indeed I think does not exist.

Computers clearly do not have the capacity to generalize. A program that can beat grandmasters at chess may be useless at go, and self-driving software that works on one type of car may fail on another, let alone a motorcycle. An art program that is good at portraits of women can fail when trying to do portraits of men, and produce horrific Daliesque madness when asked to make a landscape.

But if they did somehow have our generalization capacity, then, once they could compete with us at some things—which they surely can, already—they would be able to compete with us at just about everything. So if it were really just one thing that would let them generalize, let them leap from AI to AGI, then suddenly everything would change, almost overnight.

And so this is how the AI hype cycle goes, time and time again:

  1. A computer program is made that does something impressive, something that other computer programs could not do, perhaps even something that human beings are not very good at doing.
  2. If that same prowess could be generalized to other domains, the result would plainly be something on par with human intelligence.
  3. Therefore, the only thing this computer program needs in order to be sapient is a generalization faculty.
  4. Therefore, there is just one more step to AGI! We are nearly there! It will happen any day now!

And then, of course, despite heroic efforts, we are unable to generalize that program’s capabilities except in some very narrow way—even decades after having good chess programs, getting programs to be good at go was a major achievement. We are unable to find the generalization faculty yet again. And the software becomes yet another “AI tool” that we will use to search websites or make video games.

For there never was a generalization faculty to be found. It always was a mirage in the desert sand.

Humans are in fact spectacularly good at generalizing, compared to, well, literally everything else in the known universe. Computers are terrible at it. Animals aren’t very good at it. Just about everything else is totally incapable of it. So yes, we are the best at it.

Yet we, in fact, are not particularly good at it in any objective sense.

In experiments, people often fail to generalize their reasoning even in very basic ways. There’s a famous one where we try to get people to make an analogy between a military tactic and a radiation treatment, and while very smart, creative people often get it quickly, most people are completely unable to make the connection unless you give them a lot of specific hints. People often struggle to find creative solutions to problems even when those solutions seem utterly obvious once you know them.

I don’t think this is because people are stupid or irrational. (To paraphrase Sydney Harris: Compared to what?) I think it is because generalization is hard.

People tend to be much better at generalizing within familiar domains where they have a lot of experience or expertise; this shows that there isn’t just one generalization faculty, but many. We may have a plethora of overlapping generalization faculties that apply across different domains, and can learn to improve some over others.

But it isn’t just a matter of gaining more expertise. Highly advanced expertise is in fact usually more specialized—harder to generalize. A good amateur chess player is probably a good amateur go player, but a grandmaster chess player is rarely a grandmaster go player. Someone who does well in high school biology probably also does well in high school physics, but most biologists are not very good physicists. (And lest you say it’s simply because go and physics are harder: The converse is equally true.)

Humans do seem to have a suite of cognitive tools—some innate hardware, some learned software—that allows us to generalize our skills across domains. But even after hundreds of millions of years of evolving that capacity under the highest possible stakes, we still basically suck at it.

To be clear, I do not think it will take hundreds of millions of years to make AGI—or even millions, or even thousands. Technology moves much, much faster than evolution. But I would not be surprised if it took centuries, and I am confident it will at least take decades.

But we don’t need AGI for AI to have powerful effects on our lives. Indeed, even now, AI is already affecting our lives—in mostly bad ways, frankly, as we seem to be hurtling gleefully toward the very same corporatist cyberpunk dystopia we were warned about in the 1980s.

A lot of technologies have done great things for humanity—sanitation and vaccines, for instance—and even automation can be a very good thing, as increased productivity is how we attained our First World standard of living. But AI in particular seems best at automating away the kinds of jobs human beings actually find most fulfilling, and worsening our already staggering inequality. As a civilization, we really need to ask ourselves why we got automated writing and art before we got automated sewage cleaning or corporate management. (We should also ask ourselves why automated stock trading resulted in even more money for stock traders, instead of putting them out of their worthless parasitic jobs.) There are technological reasons for this, yes; but there are also cultural and institutional ones. Automated teaching isn’t far away, and education will be all the worse for it.

To change our lives, AI doesn’t have to be good at everything. It just needs to be good at whatever we were doing to make a living. AGI may be far away, but the impact of AI is already here.

Indeed, I think this quixotic quest for AGI, and all the concern about how to control it and what effects it will have upon our society, may actually be distracting from the real harms that “ordinary” “boring” AI is already having upon our society. I think a Terminator scenario, where the machines rapidly surpass our level of intelligence and rise up to annihilate us, is quite unlikely. But a scenario where AI puts millions of people out of work with insufficient safety net, triggering economic depression and civil unrest? That could be right around the corner.

Frankly, all it may take is getting automated trucks to work, which could be just a few years. There are nearly 4 million truck drivers in the United States—a full percentage point of employment unto itself. And the Governor of California just vetoed a bill that would require all automated trucks to have human drivers. From an economic efficiency standpoint, his veto makes perfect sense: If the trucks don’t need drivers, why require them? But from an ethical and societal standpoint… what do we do with all the truck drivers!?

Reflections on the Chinese Room

PNRJ cognitive science , , , , , , , , , , ,

Jul 12 JDN 2459044

Perhaps the most famous thought experiment in the philosophy of mind, John Searle’s Chinese Room is the sort of argument that basically every expert knows is wrong, yet can’t quite explain what is wrong with it. Here’s a brief summary of the argument; for more detail you can consult Wikipedia or the Stanford Encyclopedia of Philosophy.

I am locked in a room. The only way to communicate with me is via a slot in the door, through which papers can be passed.

Someone on the other side of the door is passing me papers with Chinese writing on them. I do not speak any Chinese. Fortunately, there is a series of file cabinets in the room, containing instruction manuals which explain (in English) what an appropriate response in Chinese would be to any given input of Chinese characters. These instructions are simply conditionals like “After receiving input A B C, output X.”

I can follow these instructions and thereby ‘hold a conversation’ in Chinese with the person outside, despite never understanding Chinese.

This room is like a Turing Test. A computer is fed symbols and has instructions telling it to output symbols; it may ‘hold a conversation’, but it will never really understand language.

First, let me note that if this argument were right, it would pretty much doom the entire project of cognitive science. Searle seems to think that calling consciousness a “biological function” as opposed to a “computation” can somehow solve this problem; but this is not how functions work. We don’t say that a crane ‘isn’t really lifting’ because it’s not made of flesh and bone. We don’t say that an airplane ‘isn’t really flying’ because it doesn’t flap its wings like a bird. He often compares to digestion, which is unambiguously a biological function; but if you make a machine that processes food chemically in the same way as digestion, that is basically a digestion machine. (In fact there is a machine called a digester that basically does that.) If Searle is right that no amount of computation could ever get you to consciousness, then we basically have no idea how anything would ever get us to consciousness.

Second, I’m guessing that the argument sounds fairly compelling, especially if you’re not very familiar with the literature. Searle chose his examples very carefully to create a powerfully seductive analogy that tilts our intuitions in a particular direction.

There are various replies that have been made to the Chinese Room. Some have pointed out that the fact that I don’t understand Chinese doesn’t mean that the system doesn’t understand Chinese (the “Systems Reply”). Others have pointed out that in the real world, conscious beings interact with their environment; they don’t just passively respond to inputs (the “Robot Reply”).

Searle has his own counter-reply to these arguments: He insists that if instead of having all those instruction manuals, I memorized all the rules, and then went out in the world and interacted with Chinese speakers, it would still be the case that I didn’t actually understand Chinese. This seems quite dubious to me: For one thing, how is that different from what we would actually observe in someone who does understand Chinese? For another, once you’re interacting with people in the real world, they can do things like point to an object and say the word for it; in such interactions, wouldn’t you eventually learn to genuinely understand the language?

But I’d like to take a somewhat different approach, and instead attack the analogy directly. The argument I’m making here is very much in the spirit of Churchland’s Luminous Room reply, but a little more concrete.

I want you to stop and think about just how big those file cabinets would have to be.

For a proper Turing Test, you can’t have a pre-defined list of allowed topics and canned responses. You’re allowed to talk about anything and everything. There are thousands of symbols in Chinese. There’s no specified limit to how long the test needs to go, or how long each sentence can be.

After each 10-character sequence, the person in the room has to somehow sort through all those file cabinets and find the right set of instructions—not simply to find the correct response to that particular 10-character sequence, but to that sequence in the context of every other sequence that has occurred so far. “What do you think about that?” is a question that one answers very differently depending on what was discussed previously.

The key issue here is combinatoric explosion. Suppose we’re dealing with 100 statements, each 10 characters long, from a vocabulary of 10,000 characters. This means that there are ((10,000)^10)^100 = 10^4000 possible conversations. That’s a ludicrously huge number. It’s bigger than a googol. Even if each atom could store one instruction, there aren’t enough atoms in the known universe. After a few dozen sentences, simply finding the correct file cabinet would be worse than finding a needle in a haystack; it would be finding a hydrogen atom in the whole galaxy.

Even if you assume a shorter memory (which I don’t think is fair; human beings can absolutely remember 100 statements back), say only 10 statements, things aren’t much better: ((10,000)^10)^10 is 10^400, which is still more atoms than there are in the known universe.

In fact, even if I assume no memory at all, just a simple Markov chain that responds only to your previous statement (which can be easily tripped up by asking the same question in a few different contexts), that would still be 10,000^10 = 10^40 sequences, which is at least a quintillion times the total data storage of every computer currently on Earth.

And I’m supposed to imagine that this can be done by hand, in real time, in order to carry out a conversation?

Note that I am not simply saying that a person in a room is too slow for the Chinese Room to work. You can use an exaflop quantum supercomputer if you like; it’s still utterly impossible to store and sort through all possible conversations.

This means that, whatever is actually going on inside the head of a real human being, it is nothing like a series of instructions that say “After receiving input A B C, output X.” A human mind cannot even fathom the total set of possible conversations, much less have a cached response to every possible sequence. This means that rules that simple cannot possibly mimic consciousness. This doesn’t mean consciousness isn’t computational; it means you’re doing the wrong kind of computations.

I’m sure Searle’s response would be to say that this is a difference only of degree, not of kind. But is it, really? Sometimes a sufficiently large difference of degree might as well be a difference of kind. (Indeed, perhaps all differences of kind are really very large differences of degree. Remember, there is a continuous series of common ancestors that links you and I to bananas.)

Moreover, Searle has claimed that his point was about semantics rather than consciousness: In an exchange with Daniel Dennett he wrote “Rather he [Dennett] misstates my position as being about consciousness rather than about semantics.” Yet semantics is exactly how we would solve this problem of combinatoric explosion.

Suppose that instead of simply having a list of symbol sequences, the file cabinets contained detailed English-to-Chinese dictionaries and grammars. After reading and memorizing those, then conversing for awhile with the Chinese speaker outside the room, who would deny that the person in the room understands Chinese? Indeed what other way is there to understand Chinese, if not reading dictionaries and talking to Chinese speakers?

Now imagine somehow converting those dictionaries and grammars into a form that a computer could directly apply. I don’t simply mean digitizing the dictionary; of course that’s easy, and it’s been done. I don’t even mean writing a program that translates automatically between English and Chinese; people are currently working on this sort of thing, and while still pretty poor, it’s getting better all the time.

No, I mean somehow coding the software so that the computer can respond to sentences in Chinese with appropriate responses in Chinese. I mean having some kind of mapping within the software of how different concepts relate to one another, with categorizations and associations built in.

I mean something like a searchable cross-referenced database, so that when asked the question, “What’s your favorite farm animal?” despite never having encountered this sentence before, the computer can go through a list of farm animals and choose one to designate as its ‘favorite’, and then store that somewhere so that later on when it is again asked it will give the same answer. And then why asked “Why do you like goats?” the computer can go through the properties of goats, choose some to be the ‘reason’ why it ‘likes’ them, and then adjust its future responses accordingly. If it decides that the reason is “horns are cute”, then when you mention some other horned animal, it updates to increase its probability of considering that animal “cute”.

I mean something like a program that is programmed to follow conversational conventions, so when you ask it its name, will not only tell you something; it will ask you your name in return, and stores that information for later. And then it will map the sound of your name to known patterns of ethnic naming conventions, and so when you say your name is “Ling-Ling Xu” it asks “Is your family Chinese?” And then when you say “yes” it asks “What part of China are they from?” and then when you say “Shanghai” it asks “Did you grow up there?” and so on. It’s not that it has some kind of rule that says “Respond to ‘Shanghai’ with ‘Did you grow up there?’”; on the contrary, later in the conversation you may say “Shanghai” and get a different response because it was in a different context. In fact, if you were to keep spamming “Shanghai” over and over again, it would sound confused: “Why do you keep saying ‘Shanghai’? I don’t understand.”

In other words, I mean semantics. I mean something approaching how human beings actually seem to organize the meanings of words in their brains. Words map to other words and contexts, and some very fundamental words (like “pain” or “red”) map directly to sensory experiences. If you are asked to define what a word means, you generally either use a lot of other words, or you point to a thing and say “It means that.” Why can’t a robot do the same thing?

I really cannot emphasize enough how radically different that process would be from simply having rules like “After receiving input A B C, output X.” I think part of why Searle’s argument is so seductive is that most people don’t have a keen grasp of computer science, and so the difference between a task that is O(N^2) like what I just outlined above doesn’t sound that different to them compared to a task that is O(10^(10^N)) like the simple input-output rules Searle describes. With a fast enough computer it wouldn’t matter, right? Well, if by “fast enough” you mean “faster than could possibly be built in our known universe”, I guess so. But O(N^2) tasks with N in the thousands are done by your computer all the time; no O(10^(10^N)) task will ever be accomplished for such an N within the Milky Way in the next ten billion years.

I suppose you could still insist that this robot, despite having the same conceptual mappings between words as we do, and acquiring new knowledge in the same way we do, and interacting in the world in the same way we do, and carrying on conversations of arbitrary length on arbitrary topics in ways indistinguishable from the way we do, still nevertheless “is not really conscious”. I don’t know how I would conclusively prove you wrong.

But I have two things to say about that: One, how do I know you aren’t such a machine? This is the problem of zombies. Two, is that really how you would react, if you met such a machine? When you see Lieutenant Commander Data on Star Trek: The Next Generation, is your thought “Oh, he’s just a calculating engine that makes a very convincing simulation of human behavior”? I don’t think it is. I think the natural, intuitive response is actually to assume that anything behaving that much like us is in fact a conscious being.

And that’s all the Chinese Room was anyway: Intuition. Searle never actually proved that the person in the room, or the person-room system, or the person-room-environment system, doesn’t actually understand Chinese. He just feels that way, and expects us to feel that way as well. But I contend that if you ever did actually meet a machine that really, truly passed the strictest form of a Turing Test, your intuition would say something quite different: You would assume that machine was as conscious as you and I.

Mental illness is different from physical illness.

PNRJ cognitive science, personal , , , , , , , , , , , ,

Post 311 Oct 13 JDN 2458770

There’s something I have heard a lot of people say about mental illness that is obviously well-intentioned, but ultimately misguided: “Mental illness is just like physical illness.”

Sometimes they say it explicitly in those terms. Other times they make analogies, like “If you wouldn’t shame someone with diabetes for using insulin, why shame someone with depression for using SSRIs?”

Yet I don’t think this line of argument will ever meaningfully reduce the stigma surrounding mental illness, because, well, it’s obviously not true.

There are some characteristics of mental illness that are analogous to physical illness—but there are some that really are quite different. And these are not just superficial differences, the way that pancreatic disease is different from liver disease. No one would say that liver cancer is exactly the same as pancreatic cancer; but they’re both obviously of the same basic category. There are differences between physical and mental illness which are both obvious, and fundamental.

Here’s the biggest one: Talk therapy works on mental illness.

You can’t talk yourself out of diabetes. You can’t talk yourself out of myocardial infarct. You can’t even talk yourself out of migraine (though I’ll get back to that one in a little bit). But you can, in a very important sense, talk yourself out of depression.

In fact, talk therapy is one of the most effective treatments for most mental disorders. Cognitive behavioral therapy for depression is on its own as effective as most antidepressants (with far fewer harmful side effects), and the two combined are clearly more effective than either alone. Talk therapy is as effective as medication on bipolar disorder, and considerably better on social anxiety disorder.

To be clear: Talk therapy is not just people telling you to cheer up, or saying it’s “all in your head”, or suggesting that you get more exercise or eat some chocolate. Nor does it consist of you ruminating by yourself and trying to talk yourself out of your disorder. Cognitive behavioral therapy is a very complex, sophisticated series of techniques that require years of expert training to master. Yet, at its core, cognitive therapy really is just a very sophisticated form of talking.

The fact that mental disorders can be so strongly affected by talk therapy shows that there really is an important sense in which mental disorders are “all in your head”, and not just the trivial way that an axe wound or even a migraine is all in your head. It isn’t just the fact that it is physically located in your brain that makes a mental disorder different; it’s something deeper than that.

Here’s the best analogy I can come up with: Physical illness is hardware. Mental illness is software.

If a computer breaks after being dropped on the floor, that’s like an axe wound: An obvious, traumatic source of physical damage that is an unambiguous cause of the failure.

If a computer’s CPU starts overheating, that’s like a physical illness, like diabetes: There may be no particular traumatic cause, or even any clear cause at all, but there is obviously something physically wrong that needs physical intervention to correct.

But if a computer is suffering glitches and showing error messages when it tries to run particular programs, that is like mental illness: Something is wrong not on the low-level hardware, but on the high-level software.

These different types of problem require different types of solutions. If your CPU is overheating, you might want to see about replacing your cooling fan or your heat sink. But if your software is glitching while your CPU is otherwise running fine, there’s no point in replacing your fan or heat sink. You need to get a programmer in there to look at the code and find out where it’s going wrong. A talk therapist is like a programmer: The words they say to you are like code scripts they’re trying to get your processor to run correctly.

Of course, our understanding of computers is vastly better than our understanding of human brains, and as a result, programmers tend to get a lot better results than psychotherapists. (Interestingly they do actually get paid about the same, though! Programmers make about 10% more on average than psychotherapists, and both are solidly within the realm of average upper-middle-class service jobs.) But the basic process is the same: Using your expert knowledge of the system, find the right set of inputs that will fix the underlying code and solve the problem. At no point do you physically intervene on the system; you could do it remotely without ever touching it—and indeed, remote talk therapy is a thing.

What about other neurological illnesses, like migraine or fibromyalgia? Well, I think these are somewhere in between. They’re definitely more physical in some sense than a mental disorder like depression. There isn’t any cognitive content to a migraine the way there is to a depressive episode. When I feel depressed or anxious, I feel depressed or anxious about something. But there’s nothing a migraine is about. To use the technical term in cognitive science, neurological disorders lack the intentionality that mental disorders generally have. “What are you depressed about?” is a question you usually can answer. “What are you migrained about?” generally isn’t.

But like mental disorders, neurological disorders are directly linked to the functioning of the brain, and often seem to operate at a higher level of functional abstraction. The brain doesn’t have pain receptors on itself the way most of your body does; getting a migraine behind your left eye doesn’t actually mean that that specific lobe of your brain is what’s malfunctioning. It’s more like a general alert your brain is sending out that something is wrong, somewhere. And fibromyalgia often feels like it’s taking place in your entire body at once. Moreover, most neurological disorders are strongly correlated with mental disorders—indeed, the comorbidity of depression with migraine and fibromyalgia in particular is extremely high.

Which disorder causes the other? That’s a surprisingly difficult question. Intuitively we might expect the “more physical” disorder to be the primary cause, but that’s not always clear. Successful treatment for depression often improves symptoms of migraine and fibromyalgia as well (though the converse is also true). They seem to be mutually reinforcing one another, and it’s not at all clear which came first. I suppose if I had to venture a guess, I’d say the pain disorders probably have causal precedence over the mood disorders, but I don’t actually know that for a fact.

To stretch my analogy a little, it may be like a software problem that ends up causing a hardware problem, or a hardware problem that ends up causing a software problem. There actually have been a few examples of this, like games with graphics so demanding that they caused GPUs to overheat.

The human brain is a lot more complicated than a computer, and the distinction between software and hardware is fuzzier; we don’t actually have “code” that runs on a “processor”. We have synapses that continually fire on and off and rewire each other. The closest thing we have to code that gets processed in sequence would be our genome, and that is several orders of magnitude less complex than the structure of our brains. Aside from simply physically copying the entire brain down to every synapse, it’s not clear that you could ever “download” a mind, science fiction notwithstanding.

Indeed, anything that changes your mind necessarily also changes your brain; the effects of talking are generally subtler than the effects of a drug (and certainly subtler than the effects of an axe wound!), but they are nevertheless real, physical changes. (This is why it is so idiotic whenever the popular science press comes out with: “New study finds that X actually changes your brain!” where X might be anything from drinking coffee to reading romance novels. Of course it does! If it has an effect on your mind, it did so by having an effect on your brain. That’s the Basic Fact of Cognitive Science.) This is not so different from computers, however: Any change in software is also a physical change, in the form of some sequence of electrical charges that were moved from one place to another. Actual physical electrons are a few microns away from where they otherwise would have been because of what was typed into that code.

Of course I want to reduce the stigma surrounding mental illness. (For both selfish and altruistic reasons, really.) But blatantly false assertions don’t seem terribly productive toward that goal. Mental illness is different from physical illness; we can’t treat it the same.

How personality makes cognitive science hard

PNRJ cognitive science, core principles, heuristics and biases, Uncategorized , , , , , , , , , , , , , ,

August 13, JDN 2457614

Why is cognitive science so difficult? First of all, let’s acknowledge that it is difficult—that even those of us who understand it better than most are still quite baffled by it in quite fundamental ways. The Hard Problem still looms large over us all, and while I know that the Chinese Room Argument is wrong, I cannot precisely pin down why.

The recursive, reflexive character of cognitive science is part of the problem; can a thing understand itself without understanding understanding itself, understanding understanding understanding itself, and on in an infinite regress? But this recursiveness applies just as much to economics and sociology, and honestly to physics and biology as well. We are physical biological systems in an economic and social system, yet most people at least understand these sciences at the most basic level—which is simply not true of cognitive science.

One of the most basic facts of cognitive science (indeed I am fond of calling it The Basic Fact of Cognitive Science) is that we are our brains, that everything human consciousness does is done by and within the brain. Yet the majority of humans believe in souls (including the majority of Americans and even the majority of Brits), and just yesterday I saw a news anchor say “Based on a new study, that feeling may originate in your brain!” He seriously said “may”. “may”? Why, next you’ll tell me that when my arms lift things, maybe they do it with muscles! Other scientists are often annoyed by how many misconceptions the general public has about science, but this is roughly the equivalent of a news anchor saying, “Based on a new study, human bodies may be made of cells!” or “Based on a new study, diamonds may be made of carbon atoms!” The misunderstanding of many sciences is widespread, but the misunderstanding of cognitive science is fundamental.

So what makes cognitive science so much harder? I have come to realize that there is a deep feature of human personality that makes cognitive science inherently difficult in a way other sciences are not.

Decades of research have uncovered a number of consistent patterns in human personality, where people’s traits tend to lie along a continuum from one extreme to another, and usually cluster near either end. Most people are familiar with a few of these, such as introversion/extraversion and optimism/pessimism; but the one that turns out to be important here is empathizing/systematizing.

Empathizers view the world as composed of sentient beings, living agents with thoughts, feelings, and desires. They are good at understanding other people and providing social support. Poets are typically empathizers.

Systematizers view the world as composed of interacting parts, interlocking components that have complex inner workings which can be analyzed and understood. They are good at solving math problems and tinkering with machines. Engineers are typically systematizers.

Most people cluster near one end of the continuum or the other; they are either strong empathizers or strong systematizers. (If you’re curious, there’s an online test you can take to find out which you are.)

But a rare few of us, perhaps as little as 2% and no more than 10%, are both; we are empathizer-systematizers, strong on both traits (showing that it’s not really a continuum between two extremes after all, and only seemed to be because the two traits are negatively correlated). A comparable number are also low on both traits, which must quite frankly make the world a baffling place in general.

Empathizer-systematizers understand the world as it truly is: Composed of sentient beings that are made of interacting parts.

The very title of this blog shows I am among this group: “human” for the empathizer, “economics” for the systematizer!

We empathizer-systematizers can intuitively grasp that there is no contradiction in saying that a person is sad because he lost his job and he is sad because serotonin levels in his cingulate gyrus are low—because it was losing his job that triggered other thoughts and memories that lowered serotonin levels in his cingulate gyrus and thereby made him sad. No one fully understands the details of how low serotonin feels like sadness—hence, the Hard Problem—but most people can’t even seem to grasp the connection at all. How can something as complex and beautiful as a human mind be made of… sparking gelatin?

Well, what would you prefer it to be made of? Silicon chips? We’re working on that. Something else? Magical fairy dust, perhaps? Pray tell, what material could the human mind be constructed from that wouldn’t bother you on a deep level?

No, what really seems to bother people is the very idea that a human mind can be constructed from material, that thoughts and feelings can be divisible into their constituent parts.

This leads people to adopt one of two extreme positions on cognitive science, both of which are quite absurd—frankly I’m not sure they are even coherent.

Pure empathizers often become dualists, saying that the mind cannot be divisible, cannot be made of material, but must be… something else, somehow, outside the material universe—whatever that means.

Pure systematizers instead often become eliminativists, acknowledging the functioning of the brain and then declaring proudly that the mind does not exist—that consciousness, emotion, and experience are all simply illusions that advanced science will one day dispense with—again, whatever that means.

I can at least imagine what a universe would be like if eliminativism were true and there were no such thing as consciousness—just a vast expanse of stars and rocks and dust, lifeless and empty. Of course, I know that I’m not in such a universe, because I am experiencing consciousness right now, and the illusion of consciousness is… consciousness. (You are not experiencing what you are experiencing right now, I say!) But I can at least visualize what such a universe would be like, and indeed it probably was our universe (or at least our solar system) up until about a billion years ago when the first sentient animals began to evolve.

Dualists, on the other hand, are speaking words, structured into grammatical sentences, but I’m not even sure they are forming coherent assertions. Sure, you can sort of imagine our souls being floating wisps of light and energy (ala the “ascended beings”, my least-favorite part of the Stargate series, which I otherwise love), but ultimately those have to be made of something, because nothing can be both fundamental and complex. Moreover, the fact that they interact with ordinary matter strongly suggests that they are made of ordinary matter (and to be fair to Stargate, at one point in the series Rodney with his already-great intelligence vastly increased declares confidently that ascended beings are indeed nothing more than “protons and electrons, protons and electrons”). Even if they were made of some different kind of matter like dark matter, they would need to obey a common system of physical laws, and ultimately we would come to think of them as matter. Otherwise, how do the two interact? If we are made of soul-stuff which is fundamentally different from other stuff, then how do we even know that other stuff exists? If we are not our bodies, then how do we experience pain when they are damaged and control them with our volition? The most coherent theory of dualism is probably Malebranche’s, which is quite literally “God did it”. Epiphenomenalism, which says that thoughts are just sort of an extra thing that also happens but has no effect (an “epiphenomenon”) on the physical brain, is also quite popular for some reason. People don’t quite seem to understand that the Law of Conservation of Energy directly forbids an “epiphenomenon” in this sense, because anything that happens involves energy, and that energy (unlike, say, money) can’t be created out of nothing; it has to come from somewhere. Analogies are often used: The whistle of a train, the smoke of a flame. But the whistle of a train is a pressure wave that vibrates the train; the smoke from a flame is made of particulates that could be used to smother the flame. At best, there are some phenomena that don’t affect each other very much—but any causal interaction at all makes dualism break down.

How can highly intelligent, highly educated philosophers and scientists make such basic errors? I think it has to be personality. They have deep, built-in (quite likely genetic) intuitions about the structure of the universe, and they just can’t shake them.

And I confess, it’s very hard for me to figure out what to say in order to break those intuitions, because my deep intuitions are so different. Just as it seems obvious to them that the world cannot be this way, it seems obvious to me that it is. It’s a bit like living in a world where 45% of people can see red but not blue and insist the American Flag is red and white, another 45% of people can see blue but not red and insist the flag is blue and white, and I’m here in the 10% who can see all colors and I’m trying to explain that the flag is red, white, and blue.

The best I can come up with is to use analogies, and computers make for quite good analogies, not least because their functioning is modeled on our thinking.

Is this word processor program (LibreOffice Writer, as it turns out) really here, or is it merely an illusion? Clearly it’s really here, right? I’m using it. It’s doing things right now. Parts of it are sort of illusions—it looks like a blank page, but it’s actually an LCD screen lit up all the way; it looks like ink, but it’s actually where the LCD turns off. But there is clearly something here, an actual entity worth talking about which has properties that are usefully described without trying to reduce them to the constituent interactions of subatomic particles.

On the other hand, can it be reduced to the interactions of subatomic particles? Absolutely. A brief sketch is something like this: It’s a software program, running on an operating system, and these in turn are represented in the physical hardware as long binary sequences, stored by ever-so-slightly higher or lower voltages in particular hardware components, which in turn are due to electrons being moved from one valence to another. Those electrons move in precise accordance with the laws of quantum mechanics, I assure you; yet this in no way changes the fact that I’m typing a blog post on a word processor.

Indeed, it’s not even particularly useful to know that the electrons are obeying the laws of quantum mechanics, and quite literally no possible computer that could be constructed in our universe could ever be large enough to fully simulate all these quantum interactions within the amount of time since the dawn of the universe. If we are to understand it at all, it must be at a much higher level—and the “software program” level really seems to be the best one for most circumstances. The vast majority of problems I’m likely to encounter are either at the software level or the macro hardware level; it’s conceivable that a race condition could emerge in the processor cache or the voltage could suddenly spike or even that a cosmic ray could randomly ionize a single vital electron, but these scenarios are far less likely to affect my life than, say, I accidentally deleted the wrong file or the battery ran out of charge because I forgot to plug it in.

Likewise, when dealing with a relationship problem, or mediating a conflict between two friends, it’s rarely relevant that some particular neuron is firing in someone’s nucleus accumbens, or that one of my friends is very low on dopamine in his mesolimbic system today. It could be, particularly if some sort of mental or neurological illness in involved, but in most cases the real issues are better understood as higher level phenomena—people being angry, or tired, or sad. These emotions are ultimately constructed of axon potentials and neurotransmitters, but that doesn’t make them any less real, nor does it change the fact that it is at the emotional level that most human matters are best understood.

Perhaps part of the problem is that human emotions take on moral significance, which other higher-level entities generally do not? But they sort of do, really, in a more indirect way. It matters a great deal morally whether or not climate change is a real phenomenon caused by carbon emissions (it is). Ultimately this moral significance can be tied to human experiences, so everything rests upon human experiences being real; but they are real, in much the same way that rocks and trees and carbon emissions are real. No amount of neuroscience will ever change that, just as no amount of biological science would disprove the existence of trees.

Indeed, some of the world’s greatest moral problems could be better solved if people were better empathizer-systematizers, and thus more willing to do cost-benefit analysis.