The Modular Mind

The other day I was walking through the garden when I looked down, saw one of these, leapt back, screamed loudly enough to notify the entire neighborhood:

(The one in my yard was insect free, however.)

After catching my breath, I wondered, “Is that a wasp nest or a beehive?” and crept back for a closer look. Wasp nest. I mentally paged through my knowledge of wasp nests: wasps abandon nests when they fall on the ground. This one was probably empty and safe to step past. I later tossed it onto the compost pile.

The interesting part of this incident wasn’t the nest, but my reaction. I jumped away from the thing before I had even consciously figured out what the nest was. Only once I was safe did I consciously think about the nest.

So I’ve been reading Gazzaniga’s Who’s in Charge? Free Will and the Science of the Brain. (I’m thinking of making this a Book Club pick; debating between this and Kurzweil’s How to Create a Mind: The Secrets of Human thought Revealed, which I have not read, but comes recommended. Feel free to vote for one, the other, or both.)

Gazzaniga discusses a problem faced by brains trying to evolve to be bigger and smarter: how do you get more neurons working without taking up an absurd amount of space connecting each and every neuron to every other neuron?

Imagine a brain with 5 connected neurons: each neuron requires 4 connections to talk to every other neuron. A 5 neuron brain would thus need space for 10 total connections.

The addition of a 6th neuron would require 5 new connections; a 7th neuron requires 6 new connections, etc. A fully connected brain of 100 neurons would require 99 connections per neuron, for a total of 4,950 connections.

The human brain has about 86 billion neurons.

Connecting all of your neurons might work fine if if you’re a sea squirt, with only 230 or so neurons, but it is going to fail hard if you’re trying to hook up 86 billion. The space required to hook up all of these neurons would be massively larger than the space you can actually maintain by eating.

So how does an organism evolving to be smarter deal with the connectivity demands of increasing brain size?

Human social lives suggest an answer: Up on the human scale, one person can, Dunbar estimates, have functional social relationships with about 150 other people, including an understanding of those people’s relationships with each other. 150 people (the “Dunbar number”) is therefore the amount of people who can reliably cooperate or form groups without requiring any top-down organization.

So how do humans survive in groups of a thousand, a million, or a billion (eg, China)? How do we build large-scale infrastructure projects requiring the work of thousands of people and used by millions, like interstate highways? By organization–that is, specialization.

In a small tribe of 150 people, almost everyone in the tribe can do most of the jobs necessary for the tribe’s survival, within the obvious limits of biology. Men and women are both primarily occupied with collecting food. Both prepare clothing and shelter; both can cook. There is some specialization of labor–obviously men can carry heavier loads; women can nurse children–but most people are generally competent at most jobs.

In a modern industrial economy, most people are completely incompetent at most jobs. I have a nice garden, but I don’t even know how to turn on a tractor, much less how to care for a cow. The average person does not know how to knit or sew, much less build a house, wire up the electricity and lay the plumbing. We attend school from 5 to 18 or 22 or 30 and end up less competent at surviving in our own societies than a cave man with no school was in his, not because school is terrible but because modern industrial society requires so much specialized knowledge to keep everything running that no one person can truly master even a tenth of it.

Specialization, not just of people but of organizations and institutions, like hospitals devoted to treating the sick, Walmarts devoted to selling goods, and Microsoft devoted to writing and selling computer software and hardware, lets society function without requiring that everyone learn to be a doctor, merchant, and computer expert.

Source

Similarly, brains expand their competence via specialization, not denser neural connections.

As UPI reports, Intelligence is correlated with fewer neural connections, not more, study finds:

The smartest people may boast more neurons than those of average intelligence, but their brains have fewer neural connections…

Neuroscientists in Germany recruited 259 participants, both men and women, to take IQ tests and have their brains imaged…

The research revealed a strong correlation between the number of dendrites in a person’s cerebral cortex and their intelligence. The smartest participants had fewer neural connections in their cerebral cortex.

Fewer neural connections overall allows different parts of the brain to specialize, increasing local competence.

All things are produced more plentifully and easily and of a better quality when one man does one thing that is natural to him and does it at the right time, and leaves other things. –Plato, The Republic

The brains of mice, as Gazzinga discusses, do not need to be highly specialized, because mice are not very smart and do not do many specialized activities. Human brains, by contrast, are highly specialized, as anyone who has ever had a stroke has discovered. (Henry Harpending of West Hunter, for example, once had a stroke while visiting Germany that knocked out the area of his brain responsible for reading, but since he couldn’t read German in the first place, he didn’t realize anything was wrong until several hours later.)

I read, about a decade ago, that male and female brains have different levels, and patterns, of internal connectivity. (Here and here are articles on the subject.) These differences in connectivity may allow men and women to excel at different skills, and since we humans are a social species that can communicate by talking, this allows us to take cognitive modality beyond the level of a single brain.

So modularity lets us learn (and do) more things, with the downside that sometimes knowledge is highly localized–that is, we have a lot of knowledge that we seem able to access only under specific circumstances, rather than use generally.

For example, I have long wondered at the phenomenon of people who can definitely do complicated math when asked to, but show no practical number sense in everyday life, like the folks from the Yale Philosophy department who are confused about why African Americans are under-represented in their major, even though Yale has an African American Studies department which attracts a disproportionate % of Yale’s African American students. The mathematical certainty that if any major in the whole school that attracts more African American students, then other majors will end up with fewer, has been lost on these otherwise bright minds.

Yalies are not the only folks who struggle to use the things they know. When asked to name a book–any book–ordinary people failed. Surely these people have heard of a book at some point in their lives–the Bible is pretty famous, as is Harry Potter. Even if you don’t like books, they were assigned in school, and your parents probably read The Cat in the Hat and Green Eggs and Ham to you when you were a kid. It is not that they do not have the knowledge as they cannot access it.

Teachers complain all the time that students–even very good ones–can memorize all of the information they need for a test, regurgitate it all perfectly, and then turn around and show no practical understanding of the information at all.

Richard Feynman wrote eloquently of his time teaching future science teachers in Brazil:

In regard to education in Brazil, I had a very interesting experience. I was teaching a group of students who would ultimately become teachers, since at that time there were not many opportunities in Brazil for a highly trained person in science. These students had already had many courses, and this was to be their most advanced course in electricity and magnetism – Maxwell’s equations, and so on. …

I discovered a very strange phenomenon: I could ask a question, which the students would answer immediately. But the next time I would ask the question – the same subject, and the same question, as far as I could tell – they couldn’t answer it at all! For instance, one time I was talking about polarized light, and I gave them all some strips of polaroid.

Polaroid passes only light whose electric vector is in a certain direction, so I explained how you could tell which way the light is polarized from whether the polaroid is dark or light.

We first took two strips of polaroid and rotated them until they let the most light through. From doing that we could tell that the two strips were now admitting light polarized in the same direction – what passed through one piece of polaroid could also pass through the other. But then I asked them how one could tell the absolute direction of polarization, for a single piece of polaroid.

They hadn’t any idea.

I knew this took a certain amount of ingenuity, so I gave them a hint: “Look at the light reflected from the bay outside.”

Nobody said anything.

Then I said, “Have you ever heard of Brewster’s Angle?”

“Yes, sir! Brewster’s Angle is the angle at which light reflected from a medium with an index of refraction is completely polarized.”

“And which way is the light polarized when it’s reflected?”

“The light is polarized perpendicular to the plane of reflection, sir.” Even now, I have to think about it; they knew it cold! They even knew the tangent of the angle equals the index!

I said, “Well?”

Still nothing. They had just told me that light reflected from a medium with an index, such as the bay outside, was polarized; they had even told me which way it was polarized.

I said, “Look at the bay outside, through the polaroid. Now turn the polaroid.”

“Ooh, it’s polarized!” they said.

After a lot of investigation, I finally figured out that the students had memorized everything, but they didn’t know what anything meant. When they heard “light that is reflected from a medium with an index,” they didn’t know that it meant a material such as water. They didn’t know that the “direction of the light” is the direction in which you see something when you’re looking at it, and so on. Everything was entirely memorized, yet nothing had been translated into meaningful words. So if I asked, “What is Brewster’s Angle?” I’m going into the computer with the right keywords. But if I say, “Look at the water,” nothing happens – they don’t have anything under “Look at the water”!

The students here are not dumb, and memorizing things is not bad–memorizing your times tables is very useful–but they have everything lodged in their “memorization module” and nothing in their “practical experience module.” (Note: I am not necessarily suggesting that thee exists a literal, physical spot in the brain where memorized and experienced knowledge reside, but that certain brain structures and networks lodge information in ways that make it easier or harder to access.)

People frequently make arguments that don’t make logical sense when you think them all the way through from start to finish, but do make sense if we assume that people are using specific brain modules for quick reasoning and don’t necessarily cross-check their results with each other. For example, when we are angry because someone has done something bad to us, we tend to snap at people who had nothing to do with it. Our brains are in “fight and punish mode” and latch on to the nearest person as the person who most likely committed the offense, even if we consciously know they weren’t involved.

Political discussions are often marred by folks running what ought to be logical arguments through status signaling, emotional, or tribal modules. The desire to see Bad People punished (a reasonable desire if we all lived in the same physical community with each other) interferes with a discussion of whether said punishment is actually useful, effective, or just. For example, a man who has been incorrectly convicted of the rape of a child will have a difficult time getting anyone to listen sympathetically to his case.

In the case of white South African victims of racially-motivated murder, the notion that their ancestors did wrong and therefore they deserve to be punished often overrides sympathy. As BBC notes, these killings tend to be particularly brutal (they often involve torture) and targeted, but the South African government doesn’t care:

According to one leading political activist, Mandla Nyaqela, this is the after-effect of the huge degree of selfishness and brutality which was shown towards the black population under apartheid. …

Virtually every week the press here report the murders of white farmers, though you will not hear much about it in the media outside South Africa.In South Africa you are twice as likely to be murdered if you are a white farmer than if you are a police officer – and the police here have a particularly dangerous life. The killings of farmers are often particularly brutal. …

Ernst Roets’s organisation has published the names of more than 2,000 people who have died over the last two decades. The government has so far been unwilling to make solving and preventing these murders a priority. …

There used to be 60,000 white farmers in South Africa. In 20 years that number has halved.

The Christian Science Monitor reports on the measures ordinary South Africans have to take in what was once a safe country to not become human shishkabobs, which you should pause and read, but is a bit of a tangent from our present discussion. The article ends with a mind-bending statement about a borrowed dog (dogs are also important for security):

My friends tell me the dog is fine around children, but is skittish around men, especially black men. The people at the dog pound told them it had probably been abused. As we walk past house after house, with barking dog after barking dog, I notice Lampo pays no attention. Instead, he’s watching the stream of housekeepers and gardeners heading home from work. They eye the dog nervously back.

Great, I think, I’m walking a racist dog.

Module one: Boy South Africa has a lot of crime. Better get a dog, cover my house with steel bars, and an extensive security system.

Module two: Associating black people with crime is racist, therefore my dog is racist for being wary of people who look like the person who abused it.

And while some people are obviously sympathetic to the plight of murdered people, “Cry me a river White South African Colonizers” is a very common reaction. (Never mind that the people committing crimes in South Africa today never lived under apartheid; they’ve lived in a black-run country for their entire lives.) Logically, white South Africans did not do anything to deserve being killed, and like the golden goose, killing the people who produce food will just trigger a repeat of Zimbabwe, but the modes of tribalism–“I do not care about these people because they are not mine and I want their stuff”–and punishment–“I read about a horrible thing someone did, so I want to punish everyone who looks like them”–trump logic.

Who dies–and how they die–significantly shapes our engagement with the news. Gun deaths via mass shootings get much more coverage and worry than ordinary homicides, even though ordinary homicides are far more common. homicides get more coverage and worry than suicides, even though suicides are far more common. The majority of gun deaths are actually suicides, but you’d never know that from listening to our national conversation about guns, simply because we are biased to worry far more about other people killng us than about ourselves.

Similarly, the death of one person via volcano receives about the same news coverage as 650 in a flood, 2,000 in a drought, or 40,000 in a famine. As the article notes:

Instead of considering the objective damage caused by natural disasters, networks tend to look for disasters that are “rife with drama”, as one New York Times article put it4—hurricanes, tornadoes, forest fires, earthquakes all make for splashy headlines and captivating visuals. Thanks to this selectivity, less “spectacular” but often times more deadly natural disasters tend to get passed over. Food shortages, for example, result in the most casualties and affect the most people per incident5 but their onset is more gradual than that of a volcanic explosion or sudden earthquake. … This bias for the spectacular is not only unfair and misleading, but also has the potential to misallocate attention and aid.

There are similar biases by continent, with disasters in Africa receiving less attention than disasters in Europe (this correlates with African disasters being more likely to be the slow-motion famines, epidemics and droughts that kill lots of people, and European disasters being splashier, though perhaps we’d consider famines “splashier” if they happened in Paris instead of Ethiopia.)

From Personality and Political Attitudes: “Conservatives are hard-working, organized, closed-minded, and emotionally stable. Liberals are lazy, disorganized, open-minded, and neurotic. Let’s see how the punditocracy spins that one.”

From a neuropolitical perspective, I suspect that patterns such as the Big Five personality traits correlating with particular political positions (“openness” with “liberalism,” for example, or “conscientiousness” with “conservativeness,”) is caused by patterns of brain activity that cause some people to depend more or less on particular brain modules for processing.

For example, conservatives process more of the world through the areas of their brain that are also used for processing disgust, (not one of “the five” but still an important psychological trait) which increases their fear of pathogens, disease vectors, and generally anything new or from the outside. Disgust can go so far as to process other people’s faces or body language as “disgusting” (eg, trans people) even when there is objectively nothing that presents an actual contamination or pathogenic risk involved.

Similarly, people who feel more guilt in one area of their life often feel guilt in others–eg, “White guilt was significantly associated with bulimia nervosa symptomatology.” The arrow of causation is unclear–guilt about eating might spill over into guilt about existing, or guilt about existing might cause guilt about eating, or people who generally feel guilty about everything could have both. Either way, these people are generally not logically reasoning, “Whites have done bad things, therefore I should starve myself.” (Should veganism be classified as a politically motivated eating disorder?)

I could continue forever–

Restrictions on medical research are biased toward preventing mentally salient incidents like thalidomide babies, but against the invisible cost of children who die from diseases that could have been cured had research not been prevented by regulations.

America has a large Somali community but not Congolese, (85,000 Somalis vs. 13,000 Congolese, of whom 10,000 hail from the DRC. Somalia has about 14 million people, the DRC has about 78.7 million people, so it’s not due to there being more Somalis in the world,) for no particular reason I’ve been able to discover, other than President Clinton once disastrously sent a few helicopters to intervene in the eternal Somali civil war and so the government decided that we now have a special obligation to take in Somalis.

–but that’s probably enough.

I have tried here to present a balanced account of different political biases, but I would like to end by noting that modular thinking, while it can lead to stupid decisions, exists for good reasons. If purely logical thinking were superior to modular, we’d probably be better at it. Still, cognitive biases exist and lead to a lot of stupid or sub-optimal results.

Advertisements

Testosterone metabolization, autism, male brain, and female identity

I began this post intending to write about testosterone metabolization in autism and possible connections with transgender identity, but realized halfway through that I didn’t actually know whether the autist-trans connection was primarily male-to-female or female-to-male. I had assumed that the relevant population is primarily MtF because both autists and trans people are primarily male, but both groups do have female populations that are large enough to contribute significantly. Here’s a sample of the data I’ve found so far:

A study conducted by a team of British scientists in 2012 found that of a pool of individuals not diagnosed on the autism spectrum, female-to-male (FTM) transgender people have higher rates of autistic features than do male-to-female (MTF) transgender people or cisgender males and females. Another study, which looked at children and adolescents admitted to a gender identity clinic in the Netherlands, found that almost 8 percent of subjects were also diagnosed with ASD.

Note that both of these studies are looking at trans people and assessing whether or not they have autism symptoms, not looking at autists and asking if they have trans symptoms. Given the characterization of autism as “extreme male brain” and that autism is diagnosed in males at about 4x the rate of females, the fact that there is some overlap between “women who think they think like men” and “traits associated with male thought patterns” is not surprising.

If the reported connection between autism and trans identity is just “autistic women feel like men,” that’s pretty non-mysterious and I just wasted an afternoon.

Though the data I have found so far still does not look directly at autists and ask how many of them have trans symptoms, the wikipedia page devoted to transgender and transsexual computer programmers lists only MtFs and no FtMs. Whether this is a pattern throughout the wider autism community, it definitely seems to be a thing among programmers. (Relevant discussion.)

So, returning to the original post:

Autism contains an amusing contradiction: on the one hand, autism is sometimes characterized as “extreme male brain,” and on the other hand, (some) autists (may be) more likely than neurotypicals to self-identify as transwomen–that is, biological men who see themselves as women. This seems contradictory: if autists are more masculine, mentally, than the average male, why don’t they identify as football players, army rangers, or something else equally masculine? For that matter, why isn’t a group with “extreme male brains” regarded as more, well, masculine?

(And if autists have extreme male brains, does that mean football players don’t? Do football players have more feminine brains than autists? Do colorless green ideas sleep furiously? DO WORDS MEAN?)

*Ahem*

In favor of the “extreme male brain” hypothesis, we have evidence that testosterone is important for certain brain functions, like spacial recognition, we have articles like this one: Testosterone and the brain:

Gender differences in spatial recognition, and age-related declines in cognition and mood, point towards testosterone as an important modulator of cerebral functions. Testosterone appears to activate a distributed cortical network, the ventral processing stream, during spatial cognition tasks, and addition of testosterone improves spatial cognition in younger and older hypogonadal men. In addition, reduced testosterone is associated with depressive disorders.

(Note that women also suffer depression at higher rates than men.)

So people with more testosterone are better at spacial cognition and other tasks that “autistic” brains typically excel at, and brains with less testosterone tend to be moody and depressed.

But hormones are tricky things. Where do they come from? Where do they go? How do we use them?

According to Wikipedia:

During the second trimester [of pregnancy], androgen level is associated with gender formation.[13] This period affects the femininization or masculinization of the fetus and can be a better predictor of feminine or masculine behaviours such as sex typed behaviour than an adult’s own levels. A mother’s testosterone level during pregnancy is correlated with her daughter’s sex-typical behavior as an adult, and the correlation is even stronger than with the daughter’s own adult testosterone level.[14]

… Early infancy androgen effects are the least understood. In the first weeks of life for male infants, testosterone levels rise. The levels remain in a pubertal range for a few months, but usually reach the barely detectable levels of childhood by 4–6 months of age.[15][16] The function of this rise in humans is unknown. It has been theorized that brain masculinization is occurring since no significant changes have been identified in other parts of the body.[17] The male brain is masculinized by the aromatization of testosterone into estrogen, which crosses the blood–brain barrier and enters the male brain, whereas female fetuses have α-fetoprotein, which binds the estrogen so that female brains are not affected.[18]

(Bold mine.)

Let’s re-read that: the male brain is masculinized by the aromatization of testosterone into estrogen.

If that’s not a weird sentence, I don’t know what is.

Let’s hop over to the scientific literature, eg, Estrogen Actions in the Brain and the Basis for Differential Action in Men and Women: A Case for Sex-Specific Medicines:

Burgeoning evidence now documents profound effects of estrogens on learning, memory, and mood as well as neurodevelopmental and neurodegenerative processes. Most data derive from studies in females, but there is mounting recognition that estrogens play important roles in the male brain, where they can be generated from circulating testosterone by local aromatase enzymes or synthesized de novo by neurons and glia. Estrogen-based therapy therefore holds considerable promise for brain disorders that affect both men and women. However, as investigations are beginning to consider the role of estrogens in the male brain more carefully, it emerges that they have different, even opposite, effects as well as similar effects in male and female brains. This review focuses on these differences, including sex dimorphisms in the ability of estradiol to influence synaptic plasticity, neurotransmission, neurodegeneration, and cognition, which, we argue, are due in a large part to sex differences in the organization of the underlying circuitry.

Hypothesis: the way testosterone works in the brain (where we both do math and “feel” male or female) and the way it works in the muscles might be very different.

Do autists actually differ from other people in testosterone (or other hormone) levels?

In Elevated rates of testosterone-related disorders in women with autism spectrum conditions, researchers surveyed autistic women and mothers of autistic children about various testosterone-related medical conditions:

Compared to controls, significantly more women with ASC [Autism Spectrum Conditions] reported (a) hirsutism, (b) bisexuality or asexuality, (c) irregular menstrual cycle, (d) dysmenorrhea, (e) polycystic ovary syndrome, (f) severe acne, (g) epilepsy, (h) tomboyism, and (i) family history of ovarian, uterine, and prostate cancers, tumors, or growths. Compared to controls, significantly more mothers of ASC children reported (a) severe acne, (b) breast and uterine cancers, tumors, or growths, and (c) family history of ovarian and uterine cancers, tumors, or growths.

Androgenic Activity in Autism has an unfortunately low number of subjects (N=9) but their results are nonetheless intriguing:

Three of the children had exhibited explosive aggression against others (anger, broken objects, violence toward others). Three engaged in self-mutilations, and three demonstrated no aggression and were in a severe state of autistic withdrawal. The appearance of aggression against others was associated with having fewer of the main symptoms of autism (autistic withdrawal, stereotypies, language dysfunctions).

Three of their subjects (they don’t say which, but presumably from the first group,) had abnormally high testosterone levels (including one of the girls in the study.) The other six subjects had normal androgen levels.

This is the first report of an association between abnormally high androgenic activity and aggression in subjects with autism. Although a previously reported study did not find group mean elevations in plasma testosterone in prepubertal autistic subjects (4), it appears here that in certain autistic individuals, especially those in puberty, hyperandrogeny may play a role in aggressive behaviors. Also, there appear to be distinct clinical forms of autism that are based on aggressive behaviors and are not classified in DSM-IV. Our preliminary findings suggest that abnormally high plasma testosterone concentration is associated with aggression against others and having fewer of the main autistic symptoms.

So, some autists have do have abnormally high testosterone levels, but those same autists are less autistic, overall, than other autists. More autistic behavior, aggression aside, is associated with normal hormone levels. Probably.

But of course that’s not fetal or early infancy testosterone levels. Unfortunately, it’s rather difficult to study fetal testosterone levels in autists, as few autists were diagnosed as fetuses. However, Foetal testosterone and autistic traits in 18 to 24-month-old children comes close:

Levels of FT [Fetal Testosterone] were analysed in amniotic fluid and compared with autistic traits, measured using the Quantitative Checklist for Autism in Toddlers (Q-CHAT) in 129 typically developing toddlers aged between 18 and 24 months (mean ± SD 19.25 ± 1.52 months). …

Sex differences were observed in Q-CHAT scores, with boys scoring significantly higher (indicating more autistic traits) than girls. In addition, we confirmed a significant positive relationship between FT levels and autistic traits.

I feel like this is veering into “we found that boys score higher on a test of male traits than girls did” territory, though.

In Polymorphisms in Genes Involved in Testosterone Metabolism in Slovak Autistic Boys, researchers found:

The present study evaluates androgen and estrogen levels in saliva as well as polymorphisms in genes for androgen receptor (AR), 5-alpha reductase (SRD5A2), and estrogen receptor alpha (ESR1) in the Slovak population of prepubertal (under 10 years) and pubertal (over 10 years) children with autism spectrum disorders. The examined prepubertal patients with autism, pubertal patients with autism, and prepubertal patients with Asperger syndrome had significantly increased levels of salivary testosterone (P < 0.05, P < 0.01, and P < 0.05, respectively) in comparison with control subjects. We found a lower number of (CAG)n repeats in the AR gene in boys with Asperger syndrome (P < 0.001). Autistic boys had an increased frequency of the T allele in the SRD5A2 gene in comparison with the control group. The frequencies of T and C alleles in ESR1 gene were comparable in all assessed groups.

What’s the significance of CAG repeats in the AR gene? Apparently they vary inversely with sensitivity to androgens:

Individuals with a lower number of CAG repeats exhibit higher AR gene expression levels and generate more functional AR receptors increasing their sensitivity to testosterone…

Fewer repeats, more sensitivity to androgens. The SRD5A2 gene is also involved in testosterone metabolization, though I’m not sure exactly what the T allele does relative to the other variants.

But just because there’s a lot of something in the blood (or saliva) doesn’t mean the body is using it. Diabetics can have high blood sugar because their bodies lack the necessary insulin to move the sugar from the blood, into their cells. Fewer androgen receptors could mean the body is metabolizing testosterone less effectively, which in turn leaves more of it floating in the blood… Biology is complicated.

What about estrogen and the autistic brain? That gets really complicated. According to Sex Hormones in Autism: Androgens and Estrogens Differentially and Reciprocally Regulate RORA, a Novel Candidate Gene for Autism:

Here, we show that male and female hormones differentially regulate the expression of a novel autism candidate gene, retinoic acid-related orphan receptor-alpha (RORA) in a neuronal cell line, SH-SY5Y. In addition, we demonstrate that RORA transcriptionally regulates aromatase, an enzyme that converts testosterone to estrogen. We further show that aromatase protein is significantly reduced in the frontal cortex of autistic subjects relative to sex- and age-matched controls, and is strongly correlated with RORA protein levels in the brain.

If autists are bad at converting testosterone to estrogen, this could leave extra testosterone floating around in their blood… but doens’t explain their supposed “extreme male brain.” Here’s another study on the same subject, since it’s confusing:

Comparing the brains of 13 children with and 13 children without autism spectrum disorder, the researchers found a 35 percent decrease in estrogen receptor beta expression as well as a 38 percent reduction in the amount of aromatase, the enzyme that converts testosterone to estrogen.

Levels of estrogen receptor beta proteins, the active molecules that result from gene expression and enable functions like brain protection, were similarly low. There was no discernable change in expression levels of estrogen receptor alpha, which mediates sexual behavior.

I don’t know if anyone has tried injecting RORA-deficient mice with estrogen, but here is a study about the effects of injecting reelin-deficient mice with estrogen:

The animals in the new studies, called ‘reeler’ mice, have one defective copy of the reelin gene and make about half the amount of reelin compared with controls. …

Reeler mice with one faulty copy serve as a model of one of the most well-established neuro-anatomical abnormalities in autism. Since the mid-1980s, scientists have known that people with autism have fewer Purkinje cells in the cerebellum than normal. These cells integrate information from throughout the cerebellum and relay it to other parts of the brain, particularly the cerebral cortex.

But there’s a twist: both male and female reeler mice have less reelin than control mice, but only the males lose Purkinje cells. …

In one of the studies, the researchers found that five days after birth, reeler mice have higher levels of testosterone in the cerebellum compared with genetically normal males3.

Keller’s team then injected estradiol — a form of the female sex hormone estrogen — into the brains of 5-day-old mice. In the male reeler mice, this treatment increases reelin levels in the cerebellum and partially blocks Purkinje cell loss. Giving more estrogen to female reeler mice has no effect — but females injected with tamoxifen, an estrogen blocker, lose Purkinje cells. …

In another study, the researchers investigated the effects of reelin deficiency and estrogen treatment on cognitive flexibility — the ability to switch strategies to solve a problem4. …

“And we saw indeed that the reeler mice are slower to switch. They tend to persevere in the old strategy,” Keller says. However, male reeler mice treated with estrogen at 5 days old show improved cognitive flexibility as adults, suggesting that the estrogen has a long-term effect.

This still doesn’t explain why autists would self-identify as transgender women (mtf) at higher rates than average, but it does suggest that any who do start hormone therapy might receive benefits completely independent of gender identity.

Let’s stop and step back a moment.

Autism is, unfortunately, badly defined. As the saying goes, if you’ve met one autist, you’ve met one autist. There are probably a variety of different, complicated things going on in the brains of different autists simply because a variety of different, complicated conditions are all being lumped together under a single label. Any mental disability that can include both non-verbal people who can barely dress and feed themselves and require lifetime care and billionaires like Bill Gates is a very badly defined condition.

(Unfortunately, people diagnose autism with questionnaires that include questions like “Is the child pedantic?” which could be equally true of both an autistic child and a child who is merely very smart and has learned more about a particular subject than their peers and so is responding in more detail than the adult is used to.)

The average autistic person is not a programmer. Autism is a disability, and the average diagnosed autist is pretty darn disabled. Among the people who have jobs and friends but nonetheless share some symptoms with formally diagnosed autists, though, programmer and the like appear to be pretty popular professions.

Back in my day, we just called these folks nerds.

Here’s a theory from a completely different direction: People feel the differences between themselves and a group they are supposed to fit into and associate with a lot more strongly than the differences between themselves and a distant group. Growing up, you probably got into more conflicts with your siblings and parents than with random strangers, even though–or perhaps because–your family is nearly identical to you genetically, culturally, and environmentally. “I am nothing like my brother!” a man declares, while simultaneously affirming that there is a great deal in common between himself and members of a race and culture from the other side of the planet. Your  coworker, someone specifically selected for the fact that they have similar mental and technical aptitudes and training as yourself, has a distinct list of traits that drive you nuts, from the way he staples papers to the way he pronounces his Ts, while the women of an obscure Afghan tribe of goat herders simply don’t enter your consciousness.

Nerds, somewhat by definition, don’t fit in. You don’t worry much about fitting into a group you’re not part of in the fist place–you probably don’t worry much about whether or not you fit in with Melanesian fishermen–but most people work hard at fitting in with their own group.

So if you’re male, but you don’t fit in with other males (say, because you’re a nerd,) and you’re down at the bottom of the highschool totem pole and feel like all of the women you’d like to date are judging you negatively next to the football players, then you might feel, rather strongly, the differences between you and other males. Other males are aggressive, they call you a faggot, they push you out of their spaces and threaten you with violence, and there’s very little you can do to respond besides retreat into your “nerd games.”

By contrast, women are polite to you, not aggressive, and don’t aggressively push you out of their spaces. Your differences with them are much less problematic, so you feel like you “fit in” with them.

(There is probably a similar dynamic at play with American men who are obsessed with anime. It’s not so much that they are truly into Japanese culture–which is mostly about quietly working hard–as they don’t fit in very well with their own culture.) (Note: not intended as a knock on anime, which certainly has some good works.)

And here’s another theory: autists have some interesting difficulties with constructing categories and making inferences from data. They also have trouble going along with the crowd, and may have fewer “mirror neurons” than normal people. So maybe autists just process the categories of “male” and “female” a little differently than everyone else, and in a small subset of autists, this results in trans identity.*

And another: maybe there are certain intersex disorders which result in differences in brain wiring/organization. (Yes, there are real interesx disorders, like Klinefelter’s, in which people have XXY chromosomes instead of XX or XY.) In a small set of cases, these unusually wired brains may be extremely good at doing certain tasks (like programming) resulting people who are both “autism spectrum” and “trans”. This is actually the theory I’ve been running with for years, though it is not incompatible with the hormonal theories discussed above.

But we are talking small: trans people of any sort are extremely rare, probably on the order of <1/1000. Even if autists were trans at 8 times the rates of non-autists, that’s still only 8/1000 or 1/125. Autists themselves are pretty rare (estimates vary, but the vast majority of people are not autistic at all,) so we are talking about a very small subset of a very small population in the first place. We only notice these correlations at all because the total population has gotten so huge.

Sometimes, extremely rare things are random chance.

Time Preference: the most under-appreciated mental trait

Time Preference isn’t sexy and exciting, like anything related to, well, sex. It isn’t controversial like IQ and gender. In fact, most of the ink spilled on the subject isn’t even found in evolutionary or evolutionary psychology texts, but over in economics papers about things like interest rates that no one but economists would want to read.

So why do I think Time Preference is so important?

Because I think Low Time Preference is the true root of high intelligence.

First, what is Time Preference?

Time Preference (aka future time orientation, time discounting, delay discounting, temporal discounting,) is the degree to which you value having a particular item today versus having it tomorrow. “High time preference” means you want things right now, whereas “low time preference” means you’re willing to wait.

A relatively famous test of Time Preference is to offer a child a cookie right now, but tell them they can have two cookies if they wait 10 minutes. Some children take the cookie right now, some wait ten minutes, and some try to wait ten minutes but succumb to the cookie right now about halfway through.

Obviously, many factors can influence your Time Preference–if you haven’t eaten in several days, for example, you’ll probably not only eat the cookie right away, but also start punching me until I give you the second cookie. If you don’t like cookies, you won’t have any trouble waiting for another, but you won’t have much to do with it. Etc. But all these things held equal, your basic inclination toward high or low time preference is probably biological–and by “biological,” I mean, “mostly genetic.”

Luckily for us, scientists have actually discovered where to break your brain to destroy your Time Preference, which means we can figure out how it works.

The scientists train rats to touch pictures with their noses in return for sugar cubes. Picture A gives them one cube right away, while picture B gives them more cubes after a delay. If the delay is too long or the reward too small, the rats just take the one cube right away. But there’s a sweet spot–apparently 4 cubes after a short wait—where the rats will figure it’s worth their while to tap picture B instead of picture A.

But if you snip the connection between the rats’ hippocampi and nucleus accumbenses, suddenly they lose all ability to wait for sugar cubes and just eat their sugar cubes right now, like a pack of golden retrievers in a room full of squeaky toys. They become completely unable to wait for the better payout of four sugar cubes, no matter how much they might want to.

So we know that this connection between the hippocampus and the nucleus accumbens is vitally important to your Time Orientation, though I don’t know what other modifications, such as low hippocampal volume or low nucleus accumbens would do.

So what do the hippocampus and nucleus accumbens do?

According to the Wikipedia, the hippocampus plays an important part in inhibition, memory, and spatial orientation. People with damaged hippocampi become amnesiacs, unable to form new memories.There is a pretty direct relationship between hippocampus size and memory, as documented primarily in old people:

“There is, however, a reliable relationship between the size of the hippocampus and memory performance — meaning that not all elderly people show hippocampal shrinkage, but those who do tend to perform less well on some memory tasks.[71] There are also reports that memory tasks tend to produce less hippocampal activation in elderly than in young subjects.[71] Furthermore, a randomized-control study published in 2011 found that aerobic exercise could increase the size of the hippocampus in adults aged 55 to 80 and also improve spatial memory.” (wikipedia)

Amnesiacs (and Alzheimer’s patients) also get lost a lot, which seems like a perfectly natural side effect of not being able to remember where you are, except that rat experiments show something even more interesting: specific cells that light up as the rats move around, encoding data about where they are.

“Neural activity sampled from 30 to 40 randomly chosen place cells carries enough information to allow a rat’s location to be reconstructed with high confidence.” (wikipedia)

"Spatial firing patterns of 8 place cells recorded from the CA1 layer of a rat. The rat ran back and forth along an elevated track, stopping at each end to eat a small food reward. Dots indicate positions where action potentials were recorded, with color indicating which neuron emitted that action potential." (from Wikipedia)
“Spatial firing patterns of 8 place cells recorded from the CA1 layer of a rat. The rat ran back and forth along an elevated track, stopping at each end to eat a small food reward. Dots indicate positions where action potentials were recorded, with color indicating which neuron emitted that action potential.” (from Wikipedia)

According to Wikipedia, the Inhibition function theory is a little older, but seems like a perfectly reasonable theory to me.

“[Inhibition function theory] derived much of its justification from two observations: first, that animals with hippocampal damage tend to be hyperactive; second, that animals with hippocampal damage often have difficulty learning to inhibit responses that they have previously been taught, especially if the response requires remaining quiet as in a passive avoidance test.”

This is, of course, exactly what the scientists found when they separated the rats’ hippocampi from their nucleus accumbenses–they lost all ability to inhibit their impulses in order to delay gratification, even for a better payout.

In other word, the hippocampus lets you learn, process the moment of objects through space (spatial reasoning) and helps you suppress your inhibitions–that is, it is directly involved in IQ and Time Preference.

 

So what is the Nucleus Accumbens?

According to Wikipedia:

“As a whole, the nucleus accumbens has a significant role in the cognitive processing of aversion, motivation, pleasure, reward and reinforcement learning;[5][6][7] hence, it has a significant role in addiction.[6][7] It plays a lesser role in processing fear (a form of aversion), impulsivity, and the placebo effect.[8][9][10] It is involved in the encoding of new motor programs as well.[6]

Dopaminergic input from the VTA modulate the activity of neurons within the nucleus accumbens. These neurons are activated directly or indirectly by euphoriant drugs (e.g., amphetamine, opiates, etc.) and by participating in rewarding experiences (e.g., sex, music, exercise, etc.).[11][12] …

The shell of the nucleus accumbens is involved in the cognitive processing of motivational salience (wanting) as well as reward perception and positive reinforcement effects.[6] Particularly important are the effects of drug and naturally rewarding stimuli on the NAc shell because these effects are related to addiction.[6] Addictive drugs have a larger effect on dopamine release in the shell than in the core.[6] The specific subset of ventral tegmental area projection neurons that synapse onto the D1-type medium spiny neurons in the shell are responsible for the immediate perception of the rewarding property of a stimulus (e.g., drug reward).[3][4] …

The nucleus accumbens core is involved in the cognitive processing of motor function related to reward and reinforcement.[6] Specifically, the core encodes new motor programs which facilitate the acquisition of a given reward in the future.[6]

So it sounds to me like the point of the nucleus accumbens is to learn “That was awesome! Let’s do it again!” or “That was bad! Let’s not do it again!”

Together, the nucleus accumbens + hippocampus can learn “4 sugar cubes in a few seconds is way better than 1 sugar cube right now.” Apart, the nucleus accumbens just says, “Sugar cubes! Sugar cubes! Sugar cubes!” and jams the lever that says “Sugar cube right now!” and there is nothing the hippocampus can do about it.

 

What distinguishes humans from all other animals? Our big brains, intellects, or impressive vocabularies?

It is our ability to acquire new knowledge and use it to plan and build complex, multi-generational societies.

Ants and bees live in complex societies, but they do not plan them. Monkeys, dolphins, squirrels, and even rats can plan for the future, but only humans plan and build cities.

Even the hunter-gatherer must plan for the future; a small tendril only a few inches high is noted during the wet season, then returned to in the dry, when it is little more than a withered stem, and the water-storing root beneath it harvested. The farmer facing winter stores up grain and wood; the city engineer plans a water and sewer system large enough to handle the next hundred years’ projected growth.

All of these activities require the interaction between the hippocampus and nucleus accumbens. The nucleus accumbens tells us that water is good, grain is tasty, fire is warm, and that clean drinking water and flushable toilets are awesome. The hippocampus reminds us that the dry season is coming, and so we should save–and remember–that root until we need it. It reminds us that we will be cold and hungry in winter if we don’t save our grain and spend a hours and hours chopping wood right now. It reminds us that not only is it good to organize the city so that everyone can have clean drinking water and flushable toilets right now, but that we should also make sure the system will keep working even as new people enter the city over time.

Disconnect these two, and your ability to plan goes down the drain. You eat all of your roots now, devour your seed corn, refuse to chop wood, and say, well, yes, running water would be nice, but that would require so much planning.

 

As I have mentioned before, I think Europeans (and probably a few other groups whose history I’m just not as familiar with and so I cannot comment on) IQ increased quite a bit in the past thousand years or so, and not just because the Catholic Church banned cousin marriage. During this time, manorialism became a big deal throughout Western Europe, and the people who exhibited good impulse control, worked hard, delayed gratification, and were able to accurately calculate the long-term effects of their actions tended to succeed (that is, have lots of children) and pass on their clever traits to their children. I suspect that selective pressure for “be a good manorial employee” was particularly strong in German, (and possibly Japan, now that I think about it,) resulting in the Germanic rigidity that makes them such good engineers.

Nothing in the manorial environment directly selected for engineering ability, higher math, large vocabularies, or really anything that we mean when we normally talk about IQ. But I do expect manorial life to select for those who could control their impulses and plan for the future, resulting in a run-away effect of increasingly clever people constructing increasingly complex societies in which people had to be increasingly good at dealing with complexity and planning to survive.

Ultimately, I see pure mathematical ability as a side effect of being able to accurately predict the effects of one’s actions and plan for the future (eg, “It will be an extra long winter, so I will need extra bushels of corn,”) and the ability to plan for the future as a side effect of being able to accurately represent the path of objects through space and remember lessons one has learned. All of these things, ultimately, are the same operations, just oriented differently through the space-time continuum.

Since your brain is, of course, built from the same DNA code as the rest of you, we would expect brain functions to have some amount of genetic heritablity, which is exactly what we find:

Source: The Heritability of Impulse Control
Source: The Heritability of Impulse Control, Genetic and environmental influences on impulsivity: a meta-analysis of twin, family and adoption studies

“A meta-analysis of twin, family and adoption studies was conducted to estimate the magnitude of genetic and environmental influences on impulsivity. The best fitting model for 41 key studies (58 independent samples from 14 month old infants to adults; N=27,147) included equal proportions of variance due to genetic (0.50) and non-shared environmental (0.50) influences, with genetic effects being both additive (0.38) and non-additive (0.12). Shared environmental effects were unimportant in explaining individual differences in impulsivity. Age, sex, and study design (twin vs. adoption) were all significant moderators of the magnitude of genetic and environmental influences on impulsivity. The relative contribution of genetic effects (broad sense heritability) and unique environmental effects were also found to be important throughout development from childhood to adulthood. Total genetic effects were found to be important for all ages, but appeared to be strongest in children. Analyses also demonstrated that genetic effects appeared to be stronger in males than in females.”

 

“Shared environmental effects” in a study like this means “the environment you and your siblings grew up in, like your household and school.” In this case, shared effects were unimportant–that means that parenting had no effect on the impulsivity of adopted children raised together in the same household. Non-shared environmental influences are basically random–you bumped your head as a kid, your mom drank during pregnancy, you were really hungry or pissed off during the test, etc., and maybe even cultural norms.

So your ability to plan for the future appears to be part genetic, and part random luck.

Bi-modal brains?

But... the second equation makes perfect sense.
But… the second equation makes sense.

So I have this co-woker–we’ll call her Delta. (Certain details have been changed to protect the privacy of the innocent.) Delta is an obviously competent, skilled worker who has succeeded at her job in a somewhat technical field for many years. She has multiple non-humanities degrees or accredidations. And yet, she frequently says things that are mind-numbingly dumb and make me want to bang my head on my desk.

To be fair, everybody makes mistakes and says incorrect things sometimes; maybe she thinks the exact same thing about me. Also, I have no real perspective on how dumb people think, because I haven’t spent much of my life talking to them. Even the formerly homeless people I know can carry on a layman’s discussion of quantum physics.

At any rate, I don’t actually think Delta is dumb. Instead, I think she has, essentially, two brain modes: Feeling Mode and Logic Mode.

Feeling Mode happens to be her default; she can do Logic Mode perfectly well, but she has to concentrate to activate it. If Logic Mode isn’t on, then things just get automatically processed through Feelings Mode and, as a result, don’t always make sense.

When Logic Mode is on, she does quite fine–her career, after all, is dependent on her rational, logical abilities, above-average math skills, etc. But her job is just that, not a passion, not something she’d do if it didn’t put food on the table. When she is in default mode, her brain just doesn’t make logical connections, notice patterns (especially meta-patterns), or otherwise understand a lot of the stuff going on around her. And her inability to judge distances/estimate sizes just makes me cringe.

My conversation topics typically go over like lead balloons.

In a recent Stanford Magazine article, Content to Code? in which Marissa Messina discusses her decision to major in computer science:

BEFORE STANFORD, I’d never heard the term “CS.” When my pre-Orientation mates used it repeatedly during our technology-free week of hiking in Yosemite prior to the start of freshman year, I had to ask them what it stood for. But their matter-of-fact response—”computer science”—was still a foreign concept to me. …

“Nonetheless, I celebrate my decision to develop my technical side. Although it does not come naturally to me, in Bay Area culture, knowing how to code feels like a prerequisite to existing. …

“I quickly learned through get-to-know-you conversations that being a “techie” was inherently cooler than being a “fuzzie,” and that social standard plus rumors of superior job prospects for engineers began to make me question my plan to major in psychology.

“Three years later, here I am, close to graduating and capable of coding. Now what?

“I certainly don’t imagine myself thriving as a professional programmer, because thinking in syntactically flawless computer-speak remains a wearisome process for me. … “

How on Earth does anyone arrive at Stanford without knowing that computer science exists?

Messina illustrates my theory rather well. She can go into logic mode, she can write code well enough to major in CS at Stanford, but it does not come naturally to her and she finds it rather unpleasant. She is only doing it because, back in freshman year, someone said her job prospects would be better with a CS degree. Now she realizes that she doesn’t actually want to do CS for a full-time job.

I suspect that most people operate primarily in Feelings Mode, and may be even worse than my co-worker at activating Logic Mode. Some may not have an operative Logic Mode at all; a few people may not have a Feeling Mode, but that seems less common. Feelings are instinctual, irrational, and messy. They exist because they are useful, but that does not mean they make logical sense.

For example, let’s suppose an out-of-control train is racing toward a group of schoolchildren who’ve been tied to the railroad tracks, but if you push a 9-foot tall man in heavy plate mail in front of the train, his death will save the children.

People operating in Logic Mode start debating the virtues of Kant’s Categorical Imperative verses Mill’s Utilitarianism.

People operating in Feelings Mode want to know what kind of psycho came up with a fucked up question like that. Children tied to the train tracks? Murdering an innocent bystander by pushing him in front of the train? Why are you fuckers debating this? Are you all sick in the head?

When Feeling people switch over into Logic Mode, I suspect it exerts some cost on them: that is, they can do it, but they don’t really like it. It’s uncomfortable, unpleasant, and sometimes exhausting. So most of the time, they prefer to be in default mode.

So there are things that they can understand in Logic Mode, but since they find the whole business unpleasant, they prefer to ignore such conclusions if they possibly can. This probably makes it very difficult to get people to make any kind of decisions involving unpleasant scenarios + data. The unpleasantness itself of the scenario breaks them out of Logic Mode and into Feeling Mode, and then the whole business is flushed down the toilet because someone goes into a screaming fit because you hurt their feelings with your data.

Earlier this morning, I happened across this “Systematizing Quotient” Quiz that HBD Chick linked to. Obviously the quiz has certain drawbacks, like user bias and the difficulty of comparing oneself to others (do I know more or less about car engines than other people? I probably know less about them than most men, but since I can diagram how an engine works and explain it, do I know more than the average woman? Where do I fall on a population scale? And what if I wouldn’t research something before buying it because I already know all about it, or because I think the brands available on the market are similar enough that the time spent resourcing would not be cost-effective?) but I thought I’d try it, anyway.

I scored in the 61-80 range, which is not terribly surprising. What’s weird is just how low everyone else scores, since the averages are 24 and 30 for women and men, respectively, and it’s not like the scale goes down to -50 or anything.

At any rate, when Delta started talking about how much she hates the Common Core math, well, I was curious. I did some digging and came up with problems like the one at the top of the screen, generally accompanied by a bunch of comments from parents like, “What are they even doing?” and “I have no idea what that is!” and “That makes no sense!” And I just look at them all like, Wow, you can’t figure out that 5+2+10+10+10=37?

Sure, math is a recently evolved trait and all, but those sorts of comments still vaguely surprise me.

IQ probably intersects the two modes via a separate axis. That is, a high-IQ Feelings Person might be able to concentrate enough of their mental resources to out-math a low-IQ Logic person, and vice versa, a high-IQ Logic Person might be able to concentrate enough mental resources to out-feel a Feeling Person. (For example, by reading a book about what various facial expressions mean and then using that knowledge in real life.) Delta, for example, could probably figure out the problem after a while, but would still say it’s a terrible problem.

There was a conversation around here somewhere about a recent paper that came out claiming that the discrepancy between the number of men and women in high-end mathematics was due to not enough girls taking rigorous math courses in middle school. Well, I don’t know about the middle schools where the paper was published, but my middle school only had one math class, and we all took it, so I don’t think that’s exactly the problem. More likely, cognitive differences just happen to be manifesting themselves in Middle School, and the math geniuses are starting to outshine people who are smart and hard working but not geniuses.

In the conversation, someone remarked that while women (or in this case, girls,) they’ve known can do math perfectly well, they tend not to enjoy it, and prefer doing other things, whereas the men they know are more or less forced to do it because their brains just happen to automatically look for patterns. This was the original inspiration for this post; the idea that someone might be able to switch back and forth between two modes, but would generally prefer one, while someone else might generally prefer the other. I might call it “Logic Mode” and The Guardian might call it “Systematizing Mode”, but they’re both basically the same.

If this is true, most people may not operate in Feeling Mode, but most women do. On the other hand, it may be that only a small sub-set of men operate primarily in Logic Mode, either, but they happen to be a larger sub-set than the sub-set of women who operate primarily in Logic Mode. Since I don’t talk to most people (no one possibly could,) and my real-life conversations are largely limited to other women, I am curious about your personal observations.