Psychological Development

So far in this course, we have focused our attention on mental processes in mature, adult humans. We have taken mind and personality as givens, and asked two basic questions:

Now we take up a new question:Where do mind and personality come from?

Views of Development

In psychology, there are two broad approaches to the question of the development of mind:

The ontogenetic point of view is concerned with the development of mind in individual species members, particularly humans. This is developmental psychology as it is usually construed. Reflecting the idea that mental development, like physical development, ends with puberty, developmental psychologists have mostly focused on infancy and childhood. More recently, it has acquired additional focus on development across the entire "life span" from birth to death, resulting in the development of new specialties in adolescence, middle age, and especially old age.

The phylogenetic point of view is concerned with the development of mind across evolutionary time, and the question of mind in nonhuman animals. It includes comparative psychology, which (as its name implies) is concerning with studying learning and other cognitive abilities in different species (comparative psychology is sometimes known as cognitive ethology); and evolutionary psychology (an offshoot of sociobiology) which traces how human mental and behavioral functions evolved through natural selection and similar processes.Go to a discussion of the phylogenetic view.

The cultural point of view is concerned with the effects of cultural diversity within the human species. This approach has its origins in 19th-century European colonialism, which asked questions, deemed essentially racist today, about whether the "primitive" mind (meaning, of course, the minds of the colonized peoples of Africa, Asia, and elsewhere) was lacking the characteristics of the "advanced" mind (meaning, of course, the minds of the European colonizers).  Edward Burnet Tylor, generally considered to be the founding father of anthropology (he offered the first definition of "culture" in its modern sense of a body of knowledge, customs and values acquired by individuals from their native environments), distinguished between those cultures which more or less "civilized".  Stripped of the racism, the sub-field of anthropology known as anthropological psychology is the intellectual heir of this work. More recently, anthropological psychology has been replaced by cultural psychology, which studies the impact of cultural differences on the individual's mental life, without the implication that one culture is more "developed" than another.Go to a discussion of the cultural view.

The Ontogenesis of Personhood

Phylogenesis has to do with the development of the species as a whole. Ontogenesis has to do with the development of the individual species member. In contrast to comparative psychology, which makes its comparisons across species, developmental psychology makes its comparisons across different epochs of the life span.

Mostly, developmental psychology focuses its interests on infancy and childhood -- a natural choice, given the idea that mental development is correlated with physical maturation. At its core, developmental psychology is dominated by the opposition of nature vs. nurture, or nativism vs. empiricism:

Of course, neither physical nor mental development stops at puberty. More recently, developmental psychology has acquired an additional focus on development across the life span, including adolescence and adulthood, with a special interest in the elderly.

The Human Genome

The human genetic endowment consists of 23 pairs of chromosomes. Each chromosome contains a large number of genes. Genes, in turn, are composed of deoxyribonucleic acid (DNA), itself composed of a sequence of four chemical bases: adenine, guanine, thymine, and cytosine (the letters A, G, T, and C which you see in graphical descriptions of various genes). Every gene is located at a particular place on a specific chromosome. And since chromosomes come in pairs, so do genes.

For a nice historical survey of genetics, see The Gene: An Intimate History by Siddhartha Mukherjee (2016).

Corresponding pairs of genes, contain information about some characteristic, such as eye color, skin pigmentation, etc. While some traits are determined by single pairs of genes, others, such as height, are determined by several genes acting together. In either case, genes come in two basic categories.Dominant genes (indicated by upper-case letters) exert an effect on some trait regardless of the other member. For example, in general the genes for brown eyes, dark hair, thick lips, and dimples are dominant genes.Recessive genes (indicated by lower-case letters) affect a trait only if the other member is identical. For example, in general the genes for blue eyes, baldness, red hair, and straight noses are recessive. The entire set of genes comprises the organism's genotype, or genetic blueprint.

One of the most important technical successes in biological research was the decoding of the human genome -- determining the precise sequence of As, Gs, Ts, and Cs that make us, as a species, different from all other species. And along with advances in gene mapping, it has been possible to determine specific genes -- or, more accurately, specific alleles, or mutations of specific genes, known as single-nucleotide polymorphisms, or SNPs -- that put us at risk for various diseases, and which dispose us to various personality traits. The most popular method for this purpose is a genome-wide association study (GWAS),s introduced earlier in our discussion of sexual orientation.  In what is essentially a multiple-regression analysis, employing very large samples, GWAS examines the relationship between every allele and some characteristic of interest -- say, heart disease or intelligence.  Of course, a lot of these correlations will occur just by chance, but there are statistical corrections for that.  

Some sense of this situation is given by the "genetic" autobiography (A Life Decoded, 2007) published by Craig Venter, the "loser" (in 2001, to a consortium government-financed academic medical centers) in the race to decode the human genome -- though, it must be said, the description of the genome produced by Venter's group is arguably superior to that produced by the government-financed researchers. Anyway, Venter has organized his autobiography around his own genome (which is what was sequenced by his group in the race), which revealed a number of such genes, including:

None of these genes means that Venter is predestined for a heart attack or Alzheimer's disease, any more than he was genetically predestined to be a surfer. It's just that these genes are more common in people who have these problems than in those who do not. They're risk factors, but not an irrevocable sentence to heart disease or dementia.

Steven Pinker, a distinguished cognitive psychologist and vigorous proponent of evolutionary psychology (see below) went through much the same exercise in "My Genomic Self", an article in the New York Times Magazine (01/11/2009).  While allowing that the DNA analyses of personal genomics will tell us a great deal about our health, and perhaps about our personalities, they won't tell us much about our personal identities.  Pinker refers to personal genomics, of the sort sold on the market by firms like 23andMe, as mostly "recreational genomics", which might be fun (he declined to learn his genetic risk for Alzheimer's Disease), but doesn't yet, and maybe can't, tell us much about ourselves that we don't already know.  He writes:

Even when the effect of some gene is indubitable, the sheer complexity of the self will mean that it will not serve as an oracle on what the person will do....   Even if personal genomics someday delivers a detailed printout of psychological traits, it will probably not change everything, or even most things. It will give us deeper insight about the biological causes of individuality, and it may narrow the guesswork in assessing individual cases. But the issues about self and society that it brings into focus have always been with us. We have always known that people are liable, to varying degrees, to antisocial temptations and weakness of the will. We have always known that people should be encouraged to develop the parts of themselves that they can (“a man’s reach should exceed his grasp”) but that it’s foolish to expect that anyone can accomplish anything (“a man has got to know his limitations”). And we know that holding people responsible for their behavior will make it more likely that they behave responsibly. “My genes made me do it” is no better an excuse than “We’re depraved on account of we’re deprived.”

A prominent GWAS by Daniel Benjamin, a behavioral economist, and his colleagues studied 1.1 million people and identified 1,271 SNPs which were, taken individually, significantly associated with educational attainment measured in years of schooling (Lee et al., Nature Genetics, 2018).  Taken together, these variants accounted for about 11-13% of the variance in educational attainment.  However, in order to maximize the likelihood of revealing genetic correlates (essentially, by reducing environmental "noise"), the study was confined to white Americans of European descent.  When the same analysis was repeated in a sample of African-Americans, the same genetic variants accounted for less than 2% of the variance in educational attainment.  None of these SNPs should be thought of as dedicated to educational achievement, especially in populations other than those in which the GWAS was conducted in the first place.  In the first place, formal schooling arose too recently to be subject to genetic selection.  In fact, the actual function of these SNPs is unknown, and it is likely that many of them control traits whose association with educational attainment is rather indirect.  So (to take an example from a story in The Economist, 03/31/2018), children who inherit weak bladders from their parents may do poorly on timed examinations.  In any event, the fact that there are SNPs associated with educational attainment, or any other psychosocial characteristic, doesn't mean that these traits are directly heritable -- or that focusing on genes is the best way to understand them, much less improve them.  Remember the lessons of the search for the "gay gene", recounted above.

Notice the historical progression here.  To begin with, the "intelligence gene" is the Holy Grail of behavior genetics.  Psychologists have been interested in finding the genes that cause schizophrenia, or depression, or some other form of mental illness, but the search for the intelligence gene goes back to the 19th century -- that is, when Sir Francis Galton made the observation that educational achievement and financial success, both presumed markers of intelligence, tended to run in families (in Hereditary Genius, first published in 1869).  You might think, maybe, that in 19th century England rich families were more likely to send their children to Oxford and Cambridge and to get elected to Parliament or elevated to the House of Lords, than poor ones.  You might think that.  But Galton concluded that intelligence was hereditary, and was passed from one generation to another through bloodlines.  Galton knew nothing of genes: Gregor Mendel's papers setting out the principles of heredity had been published in an obscure journal in 1865 and 1866, but only garnered serious attention when they were discovered and republished in 1900; and Wilhelm Johannsen didn't coin the word "gene" until 1905). 

The molecular structure of DNA wasn't decoded until 1963 by Watson and Crick (with not-so-little but very much uncredited help from Rosalind Franklin).  But even the discovery of the "double helix" didn't advance our understanding of the genetic basis of intelligence.  Because, for the most part, we didn't know how many genes we had, or where particular genes were located on the 23 pairs of chromosomes that make up the human genome.  All we had was evidence from twin studies, of the sort reviewed in the lectures on Thinking, which clearly established the genetic contribution to individual differences in intelligence -- though the same studies also established the importance of the environment, particularly the nonshared environment.

Even twin studies can be misleading.  For example, one group of behavior geneticists used variants on the twin-study method (technically, the adoption-study method) to show that there is a genetic contribution to the amount of time that young children spend viewing television (Plomin et al., Psych. Science, 1990). But let's be clear: there is no "television-viewing gene".  TV wasn't invented until the 1920s, so there hasn't been enough time to evolve such a gene.  But there may be genetic contributions to traits that influence television viewing (for example, the tendency of some children to just sit around the house; or the tendency of some parents to ignore their children).  But any such influence is going to be indirect.  Moreover, even in this study the heritability of television-viewing dropped from roughly 50% at age 3 to less than 20% at age 5, so there are clearly other things going on.

At first, the hypothesis was that there was one, or only a few, genes that coded for intelligence, and that these genetic determinants could be identified by correlating specific genes with IQ scores (or some proxy for IQ, like years of schooling).  That didn't work out.  For example, Robert Plomin's group at the University of Texas (Chorney et al., Psych. Sci., 1998) reported that a specific gene known as IGF2R (for insulin-like growth factor-2 receptor), located on Chromosome 6, was associated with high IQ scores.  Plomin is a highly regarded behavior geneticist, and justly so, and he knows how to do these studies right, so before he submitted his research for publication he replicated his finding in a second sample.  As it happens, I was the editor of Psychological Science, the journal to which Plomin's research group submitted their paper.  I was skeptical of the finding, and told Plomin so: although I recognize that there is a genetic component to intelligence (the twin studies reviewed in the lectures on Thinking establish that definitively) these kinds of findings, which identify a specific gene with a specific psychological trait, rarely hold up when replicated in a new sample.  Still, there was nothing wrong with the study, and I agreed with Plomin that  I'ostensibly m sceptical of these kinds of studies: I was a graduat when the first "depression gene", or at least the chromosome on which it was located, was identified, and that fin

It should also be remembered that there is a great deal of "genetic" material on our chromosomes besides genes, and the GWAS technique doesn't discriminate between gene variants -- sequences of nucleotides that code for some trait (like intelligence) or disease (like schizophrenia) -- and SNPs, which also come in variants (hence the term polymorphism), but which aren't actually genes.  Anyway, with new powerful computers, we can correlate individual SNPs with traits or diseases, just as earlier investigators did with genes themselves.  And the result is that, indeed,  some SNPs are correlated with some traits.  But -- and this is a big but -- the correlations are very rarely statistically significant.  For example, the presence of one SNP, known as "rs11584700" (don't ask), adds two days to educational attainment.  However, if you aggregate across a couple dozen (or hundred) nonsignificant correlations, the overall correlation between the entire package of SNPs -- known as a polygenic risk score -- can rise to statistical significance.  Even so, the correlations, while they may be statistically significant because of the large numbers of subjects involved, may be practically trivial.  More important, however, remember the lesson of IGF2R.  An array of SNPs may correlate with intelligence (or years of schooling) in one sample, but this relationship may not replicate in another sample.

This is not to say that there are no genetic correlates of intelligence (or even educational attainment) or other socially significant traits.  I'm prepared to learn they are.  But the really important questions for me, when looking at genetic correlations, are:

• How strong is the association?  Is it statistically significant?
  
• If significant, is it replicable?
• If replicable, is the association strong enough to be of any practical significance?
• If so, what do you plan to do with this knowledge?
  

For an enthusiastic summary of the prospects for using GWAS to identify the genetic underpinnings of intelligence, and a defense of the search for the genetic causes of psychosocial phenotypes, see The Genetic Lottery: Why DNA Matters for Social Equality by Kathryn Paige Harden (2022), who directs the Twin Project at the University of Texas.  Her book was given a scathing review by M.W. Feldman and Jessica Riskin, two Stanford biologists, in the New York Review of Books ("Why Biology is Not Destiny", 04/21/2022).  Feldman and Riskin are appropriately skeptical about GWAS, and behavior genetics in general (although if they had taken this course, they wouldn't be quite so skeptical about psychologists' ability to measure personality traits, and they'd know where openness to experience comes from).  Like good interactionists, F&R argue that it's impossible to separate genetic and environmental influences, "because the environment is in the genome and the genome is in the environment".  Harden replied in a subsequent issue (06/09/2022), followed by a rejoinder by Feldman and Riskin.  The whole exchange is very valuable for setting out the terms of the debate. 

From Genotype to Phenotype

Of course, genes don't act in isolation to determine various heritable traits. One's genetic endowment interacts with environmental factors to produce a phenotype, or what the organism actually looks like. The genotype represents the individual's biological potential, which is actualized within a particular environmental context. The environment can be further classified as prenatal (the environment in the womb during gestation), perinatal (the environment around the time of birth), and postnatal (the environment present after birth, and throughout the life course until death).

Because of the role of the environment, phenotypes are not necessarily equivalent to genotypes. For example, two individuals may have the same phenotype but different genotypes. Thus, of two brown-eyed individuals, one might have two dominant genes for brown eyes (BB), while another might have one dominant gene for brown eyes and one recessive gene for blue eyes (Bb). Similarly, two individuals may have the same genotype, but different phenotypes. For example, two individuals may have the same dominant genes for dimples, (DD), but one has his or her dimples removed by plastic surgery.

The chromosomes are found in the nucleus of each cell in the human body, except the sperm cells of the male and the egg cells of the female, which contain only one element of each pair. At fertilization, each element contributed by the male pairs up with the corresponding element contributed by the female to form a single cell, or zygote. At this point, cell division begins. The first few divisions form a larger structure, the blastocyst. After six days, the zygote is implanted in the uterus, at which point we speak of an embryo.

Embryological Development

What are the mechanisms of embryological development? At one point, it was thought that the individual possesses adult form from the very beginning -- that is, that the embryo is a kind of homunculus (little man), and that the embryo simply grew in size. This view of development is obviously incorrect. But that didn't stop people from seeing adult forms in embryos!

In the 19th century, with the adoption of the theory of evolution, the homunculus view was gradually replaced by the recapitulation view, based on Haeckel's biogenetic law that

"Ontogeny recapitulates phylogeny".

What Haeckel meant was that the development of the individual replicates the stages of evolution: that the juvenile stage of human development repeats the adult stages of our evolutionary ancestors. Thus, it was thought, the human embryo first looks like an adult fish; later, it looks like adult amphibians, reptiles, birds, mammals, and nonhuman primates. This view of development is also incorrect, but it took a long time for people to figure this out.

The current view of development is based on von Baer's principle of differentiation. According to this rule, development proceeds from the general to the specific. In the early stages of its development, every organism is homogeneous and coarsely structured. But it carries the potential for later structure. In later stages of development, the organism is heterogeneous and finely built -- it more closely represents actualized potential. Thus, the human embryo doesn't look like an adult fish. But at some point, human and fish embryos look very much alike. These common structures later differentiate into fish and humans. A good example of the differentiation principle is the development of the human reproductive anatomy, which we'll discuss later in the course.

Neural Development in the Fetus, Infant, and Child

At the end of the second week of gestation, the embryo is characterized by a primitive streak which will develop into the spinal cord. By the end of the fourth week, somites develop, which will become the vertebrae surrounding the spinal cord -- the characteristic that differentiates vertebrates from invertebrates.

Bodily asymmetries seem to have their origins in events occurring at an early stage of embryological development. Studies of mouse embryos by Shigenori Nonaka and his colleagues, published in 2002, implicate a structure known as the node, which contains a number of cilia, or hairlike structures. The motion of the cilia induce fluids to move over the embryo from right to left. These fluids contain hormones and other chemicals that control development, and thus cause the heart to grow on the left, and the liver and appendix on the right -- at least for 99.99% of people. On rare occasions, a condition known as situs inversus, the position of the embryo is reversed, so the fluids flow left to right, resulting in a reversal of the relative positions of the internal organs -- heart on the right, liver and appendix on the left. It is possible that this process is responsible for including the hemispheric asymmetries associated with cerebral lateralization -- although something else must also be involved, given that the incidence of right-handedness is far greater than 0.01%.

All three processes -- cortical development, synaptic development, and myelinization -- continue for the rest of fetal development, and even after birth. In fact, myelinization is not complete until late in childhood.

Studies of premature infants indicate that an EEG signal can be recorded at about 25 weeks of gestation. At this point, there is evidence of the first organized electrical activity in the brain. Thus, somewhere between the 6th and 8th month of gestation the fetal brain becomes recognizably human. There are lots of the folds and fissures that characterize the human cerebral cortex. And there is some evidence for hemispheric specialization: premature infants respond more to speech presented to the left hemisphere, and more to music presented to the right. At this point, in the 3rd semester of gestation, the brain clearly differentiates humans from non-humans.

Interestingly, survivability takes a big jump at this point as well. If born before about 24 weeks of gestation, the infant has little chance of survival, and then only with artificial life supports; if born after 26 weeks, the chances of survival are very good. If born at this point, the human neonate clearly has human physical characteristics, and human mental capacities. In other words, by some accounts, by this point the fetus arguably has personhood, because its has actualized its potential to become human. At this point, it makes sense to begin to talk about personality -- how the person actualizes his or her potential for individuality. 

What are the implications of fetal neural development for the mind and behavior of the fetus?  Here are some things we know. 

The importance of early experience in neural development cannot be overemphasized.  Socioeconomic status, nurturance at age 4 (as opposed to neglect, even if benign), the number of words spoken to the infant, and other environmental factors all are correlated with various measures of brain development.

Nature and Nurture in Personality Development

So where does personality come from? We have already seen part of the answer: Personality is not a given, fixed once and for all time, whether by the genes or through early experience. Rather, personality emerges out of the interaction between the person and the environment, and is continuously constructed and reconstructed through social interaction. A major theme of this interactionist view of personality is that the person is a part of his or her own environment, shaping the environment through evocation, selection, behavioral manipulation, and cognitive transformation. In the same way, development is not just something that happens to the individual. Instead, the individual is an active force in his or her own development.

The development of the individual begins with his or her genetic endowment, but genes do not act in isolation. The organism's genotype, or biological potential (sometimes referred to as the individual's "genetic blueprint", interacts with the environment to produce the organism's phenotype, or what the organism actually "looks like" -- psychologically as well as morphologically.

The individual's phenotype is his or her genotype actualized within a particular environmental context:

More broadly, it is now known that genes are turned "on" and "off" by environmental events.

The point is that genes act jointly with environments to produce phenotypes. These environments fall into three broad categories:

In light of the relation between genotype and phenotype, the question about "nature or nurture" is not whether some physical, mental, or behavioral trait is inherited or acquired. Better questions are:

The Twin-Study Method

Many basic questions of nature and nurture in personality can be addressed by using the techniques of behavior-genetics are used to analyze the origins of psychological characteristics. Perhaps the most interesting outcome of these experiments is that, while initially intended to shed light on the role of genetic factors in personality development. These behavior-genetic analyses show the clear role of genetic determinants in personality, but also reveal a clear role for the environment.

The most popular method in behavior genetics is the twin study -- which, as its name implies, compares two kinds of twins in terms of similarity in personality:

A qualification is in order. A vast proportion of the human genome, about 90%, is the same for all human individuals -- it's what makes us human, as opposed to chimpanzees or some other kind of organism. Only about 10% of the human genome actually varies from one individual to another. So, when we speak of DZ twins having "50%" of their genes in common, we really mean that they share 50% of that 10%. And when we speak of unrelated individuals having "no" genes in common, we really mean that they don't share any more of that 10% than we'd expect by chance.

Of course, you could compare triplets, quadruplets, and the like as well, but twins are much more convenient because they occur much more frequently in the population. Regardless of twins or triplets or whatever, the logic of the twin study is simple: To the extent that a trait is inherited, we would expect MZ twins to be more alike on that trait than DZ twins.

Born in Ontario in 1934, the Dionne Quintuplets (all girls, all genetically identical) were the first quintuplets to live past infancy.  Their parents already had four children, and the birth of five more was completely unexpected.  The family's financial straits led their parents to sign over custody of the girls to the Canadian Red Cross, which built a special hospital and "observatory" for them across the street from their family home; their parents also agreed to display them at the Chicago World's Fair.  "Quintland" actually became a tourist attraction, complete with bumper stickers.  All of this was in an attempt to protect the children -- and of course it misfired badly, creating, among other things, a sharp division between the quints on the one hand, and their parents and four older siblings on the other.  Their story has been told in many books: the most recent, The Miracle and Tragedy of the Dionne Quintuplets by Sarah Miller (2019), traces their lives.  Spoiler alert: the tragedy is that they were exploited anyway; the miracle is that each girl grew up with her own individual personality and interests.

 

In 2009, the Crouch quadruplets, Ray, Kenny, Carol, and Martina, all received offers of early admission to Yale ("Boola Boola, Boola Boola: Yale Says Yes, 4 Times" by Jacques Steinberg, New York Times, 12/19/2009).  Now you don't have to be twins to share such outcomes (and the Crouch quadruplets obviously aren't identical quadruplets anyway!).  Consider The 5 Browns, sibling pianists from Utah, all of whom, successively, were admitted to study at Julliard.  How much of these outcomes is due to shared genetic potential?  How much to shared environment?  How much to luck and chance?  That's why we do twin and family studies.

The April 2018 issue of National Geographic magazine, devoted to race, featured Marcia and Millie Biggs, 11-year-old fraternal twin sisters living in England (shown here with their father, Michael; you'll see their photo from the magazine cover later in these lectures).  Their mother, Amanda Wanklin, calls them her "one in a million miracle".  The girls' differences in physical appearance are due entirely to the vicissitudes of genetic chance.  But while Millie is a "girlie" girl, Marcia is more of a "tomboy" (those are Marcia's words, not mine).  Where those psychological differences come from is a much more complex, and much more interesting, story -- as we'll see in what follows.

Twins in space!  Einstein's theory of special relativity predicts that a twin traveling through space at high speed will age less rapidly than his or her earth-bound counterpart.  That hypothesis hasn't been tested yet, because we don't have the ability to put a twin in a spacecraft that travels at close to light-speed.  But still, there are reasons to think that space travel, including exposure to microgravity and ionizing radiation, not to mention the stress and absence of a circadian clock, might have substantial physiological and psychological effects.  Taking advantage of the presence of two identical twins, Scott and Mark Kelly, on the roster of American astronauts, the National Aeronautics and Space Administration (NASA) subjected the pair to a battery of physiological and psychological tests before, during, and after one of them (Scott) completed a 1-year mission aboard the International Space Station (ISS).  The results, reported in 2019 (Garrett-Bakelman et al., Science 04/12/2019) did indeed uncover a number of physiological changes induced by a year in space.  Interestingly, Scott's telomeres -- portions of the chromosome that shorten with age -- actually did lengthen during spaceflight: score one, maybe, for Einstein.  Psychologically, frankly, the study was sort of impoverished, but revealed no significant decrements in cognitive speed, accuracy, or efficiency (speed-accuracy trade-off) inflight, compared to pre-flight; post-flight, however, there was some diminution in performance.

As far as personality goes, the usual technique in twin studies is to administer some personality inventory, like the MMPI or CPI or NEO-PI to a large sample of MZ and DZ twins, thus obtaining scores representing each individual's standing on each personality trait measured by the inventory. Then we measure the similarity of the twins on each trait. The most common measure of similarity is the correlation coefficient, which summarizes the direction and strength of the relationship between two variables -- for example, between extraversion in one twin and extraversion in the other.

If we assume that a personality trait (or a physical trait like eye color, for that matter) is solely determined by the genes, and the environment has no effect, we would expect that following pattern of correlations:

An alternative measure of similarity is the concordance rate. Assuming that a person either has a trait or does not, on the hypothesis of exclusively genetic determination we would expect a concordance rate of 100% for MZ twins, and a concordance rate of about 50% for DZ twins (we will meet up with concordance rates again in the lectures on Personality and Psychopathology, when we discuss the origins of mental illness).

More generally, to the extent that a trait is inherited, we expect that MZ twins will be more similar to each other than DZ twins -- regardless of whether similarity is measured by the correlation coefficient or the concordance rate.

Once in a while, you'll read some extraordinary story about identical twins, separated at birth, who turn out remarkably similar as adults.  For examples, see Born Together -- Reared Apart: The Landmark Minnesota Twin Study by Nancy L. Segal (2012).  For example, one of the pairs of subjects in the Minnesota study were separated at birth, but both were named "Jim" by their adoptive parents, both married women named "Linda", both subsequently divorced, and both later married women named "Betty".  Both were chain-smokers, both drove Chevrolet cars, both were employed as deputy sheriffs, and both preferred to vacation at the same beach in Florida.  These coincidences are fun to read about, but let's be clear: there's no gene for driving Chevrolets, or for marrying women named "Linda".  Genes operate at an entirely different level.

For more on twins, especially identical twins, see How To Be Multiple: The Philosophy of Twins (2024) by the philosopher Helena de Bres, herself a twin, reviewed by Parul Sehgal in "Double Vision", New Yorker, 01/29/2024, from which the following quotations are taken.  Also Twinkind: The Singular Significance of Twins (2024) by William Viney (also a twin, and also mentioned by Sehgal).  Sehgal writes: "De Bres invokes twins from life and legend... to examine how multiples complicate our notions of personhood, attachent, and agency.  Twins have been critical to our understanding of ourselves,[de Bres] argues....  And they continue to unsettle our notions abut where bodies end and begin, about whether personalities, even fates, are forged or found....  For de Bres, to be a twin was to be seen.  It was, indeed, the social currency she possessed...; how, in school, being a twin meant that everyone knew who they were ('though not necessarily who we each were'), giving them 'a lifetime backstage pass to semicoolness'.  De Bres herself writes: "We identical twins, then, are tricky, disruptive, even deditious creatures.  We are the perfect crime.  Most people run into us only occasionally, but the experience of doing so, or the simple idea of twins, can enflame broader anxieties about the fragility of everyone's capacity to identify anyone."

Genes, Environments, and the "Big Five"

In fact, a twin study of the "Big Five" personality traits by Loehlin and his colleagues (1992) showed that for each dimension of the Big Five, MZ twins were more alike than DZ twins. Studies using other personality inventories, such as the MMPI or the CPI, have yielded similar sorts of findings. Taken together, this body of research provides prima facie evidence for a genetic contribution to individual differences in personality.But genes aren't the only forces determining individual differences in personality. If they were, then:

Physical characteristics like eye color and hair texture may come close to these values. Height and weight show high MZ correlations, but even these aren't perfectly correlated. The reason is that genotype alone is never sufficient to determine phenotype. Phenotypes always result from the interaction of genotypes with environmental factors. The typical MZ correlations for physical traits range upward from 0.50, suggesting high if not perfect heritabilities. By contrast, the typical MZ correlations for psychological traits range between 0.25 and .50 -- suggesting that genetic influences on personality are relatively weak, and environmental influences are correspondingly strong.

Note, however, that heritability can be misleading.  I owe the following example to Eric Turkheimer, a prominent behavior geneticist at the University of Virginia who has studied the heritability of both IQ and major mental illnesses like schizophrenia.  Consider the human trait of having two arms and two legs.  This is without any doubt completely determined by our genetic heritage (as Plato pointed out, we are featherless bipeds).  But if you look at the concordance between MZ and DZ twins for "armedness", you'll find almost perfect concordance: virtually 100% of the siblings of people with two arms also have two arms.  If then you just compare MZ and DZ correlations, you get a heritability of zero (0), or very close to it.  But we know that, barring rare incidents of accident or surgery, the number of arms is completely determined by the genes.

Of course, even perfect MZ correlations of 1.00 wouldn't be enough to clinch the case of exclusive heritability. Twins, especially MZ twins, share more than genes. They also share environments, and it is possible that MZ twins live in environments that are more alike than DZ twins (perhaps because MZ twins are of the same sex, or perhaps simply because they look more alike). This raises the question: How do we tease apart the genetic and environmental contributions to personality?

One way to address this question is to study identical twins who have been separate at birth and reared apart -- meaning that they share genes but not environments. Such cases do exist, and they are interesting, but the fact is that there are not enough of them to make a satisfactory sample. Moreover, many twins ostensibly "reared apart" really aren't. For example, twins might be "separated" for economic reasons, because their parents can't afford to raise both of them, and one of them reared by an aunt and uncle down the road. Even when twins are actually adopted out, adoption agencies often try to place adoptees with foster parents who resemble their biological parents in terms of age, educational levels, occupational status, and the like. Such twins probably share more of their environment than not.

It's important to recognize that heritability estimates are accurate only for the environment from which a population was drawn. For example, as discussed in the lectures on Thinking, the heritability of IQ is higher in high-SES populations than it is in low-SES populations. Apparently, high socioeconomic status gives freer rein to genetic influences, allowing people "to be all they can be", while low SES constrains them, confronting individuals with artificial ceilings on attainment.

Separating Genes and Environment(s)

The genetics-versus-environment question can also be approached in the context of standard twin studies. But the issue gets a little complicated, because it turns out that the "environment" comes in two basic forms:

As it happens, the relative strength of both environmental components of personality, as well as the genetic component, can be estimated from the observed pattern of MZ and DZ correlations  Falconer & Mackay, 1996).

Consider, first the entire distribution of a trait within a population, from those individuals with the lowest scores on Extraversion or Neuroticism to those with the highest scores on these traits. This distribution is typically represented by a more-or-less "normal" distribution -- the famous "bell curve". Each person's score on a trait measure -- Neuroticism, Extraversion, whatever -- is a measure of the person's phenotype -- how he or she "turned out" with respect to that dimension of personality. The entire distribution of individual scores within a population is the total variance on the trait(s) in question.

This total variance in the trait (100%, or a proportion equal to 1.0) is the sum of genetic variance (i.e., variance in the trait that is accounted for by to genetic variability, or individual differences in genotypes) and environmental variance (i.e., variance in the trait that is accounted for by environmental variability, or individual differences in environments):

The environmental variance, in turn, is the sum of variance due to the shared environment and variance due to the nonshared environment:

First, consider the comparison between MZ and DZ twins. By definition, MZ and DZ twins are identical with respect to the shared environment -- they are raised by the same parents in the same household. But MZ and DZ twins differ genetically: MZ twins are identical genetically, while DZ twins are no more alike, genetically speaking, than any two non-twin siblings. Thus, any difference in similarity between MZ and DZ twins must be due to genetic differences.

Genetic variance is a function of the difference between MZ and DZ correlations: the greater the MZ correlation compared to that of DZ, the more we can attribute similarity to shared genes than to shared environments (don't worry about where the "2" comes from: this is a technical detail):

G = 2 * (MZ - DZ).

Next, consider MZ twins raised together. By definition, MZ twins are identical with respect to both genes and the shared environment. They are the product of a single fertilized egg, and they are raised by the same parents in the same household. If the only contributions to variance were from the genes and the shared environment, they ought to be identical in personality. Therefore, any departure from a perfect correlation of 1.00 must reflect the contribution of the nonshared environment.

Variance due to the nonshared environment is a function of the MZ correlation: MZ twins share both genes and (shared) environment, so any MZ correlation less than a perfect +1.0 must reflect the contribution of the nonshared environment:

E_NS = 1 - MZ.

Once we've estimated the contributions of the genes and the nonshared environment, variance due to the shared environment is all that's left, so it can be estimated simply by subtracting G and E_NS from 1:

E_S= 1 - G -E _NS.

Here are some illustrative examples:

If the correlation for MZ twins is 1.00, and the correlation for DZ twins is .50, we have the situation described earlier: all the variance on the trait is attributable to genetic variance, and no variance is attributable to either sort of environmental factor, shared or nonshared.

If we reduce the MZ correlation substantially, but keep the DZ correlation pretty much the same, most of the variance is now attributed to the environment. There is some genetic effect, and some effect of each sort of environment.

If we increase the MZ correlation, but also increase the DZ correlation, most of the variance is still attributable to the environment, but the strength of the genetic contribution diminishes markedly.

If we use MZ and DZ correlations that roughly approximate those found in Loehlin's study of the Big Five, we find evidence of a substantial genetic component of variance, but also a substantial environmental component. Most important, we find that the contribution of the nonshared environment is much greater than that of the shared environment. In fact, the effect of the shared environment is minimal.

In fact, if we use Loehlin's data for Extraversion, that's exactly the pattern we find.The MZ correlation is about twice as high as the DZ correlation. The estimate for G is 48%, for E_NS it's 52%, with nothing left over for E_S.

And also for Neuroticism.

A variant on the twin study is the adoption study, which compares the similarity between adopted children and their biological parents and siblings (with whom they share 50% of their genes, on average) and adopted children and their adoptive parents and siblings (with whom they share no particular genes). To the extent that genes contribute to individual differences on some variable, biologically related individuals should be more alike than biologically unrelated ones.

The two methods can be combined, in a way, by comparing identical twins reared together (who share genes and environment) and identical twins reared apart (who, ostensibly, share genes but not environment). I say "ostensibly", because some twins "raised apart" simply live in different households (like with grandparents or aunts and uncles) for economic reasons, but still have a great deal of family, school, and social life in common. So it's not necessarily the case that identical twins reared apart don't share a family environment.

The Big Five

Considering all five traits examined in Loehlin's (1992) study of The Big Five, the results of actual twin studies of personality reveal that genetic factors account for approximately 40% of the variance on the Big Five traits; the nonshared environment accounts for approximately 50% of variance; and the shared environment accounts for less than 10% of variance. Apparently the family environment is not decisive for adult personality, and the nonshared environment is far more important.

To summarize, for each of The Big Five dimensions of personality, there is:

And now, perhaps, we have some idea of where some of those genes are.  In 2024, Daniel Levey, a psychologist at Yale, and his colleagues published an extensive genetic study of the Big Five personality traits (Gupta et al., Nature Human Behavior, 2024).  "Extensive" is actually an understatement.  They drew on the the Veterans Administration's "Million Veterans Project" (MVP), which has collected health data, including DNA information, from a large number of US military veterans receiving healthcare through the Veterans Administration (VA) system (hence its name).   A subsample of these individuals, numbering about 270,000 -- including completed an inventory of the Big 5 personality traits; Levey and his colleagues then used GWAS methodology to identify specific genetic loci that are associated with each of the Big 5 dimensions.  These are not quite the same as genes, but rather designate specific regions on chromosomes where genes or genetic markers are located.  Close enough for our purposes.  The MVP sample included about 240,000 subjects of European ancestry (EUR), and another 30,000 subjects of African ancestry (AFR).  The EUR portion of the MVP sample was also combined with other large samples collected in England and elsewhere to create a huge database of almost 700,000 subjects of European ancestry. 

Gupta, Levey, et al.'s initial analysis of the MVP data revealed 34 genomic loci that were significantly associated with one or another of the Big5 traits in the EUR subsample -- 11 each for Neuroticism and Extraversion, 3 for Agreeableness, 2 for Conscientiousness, and 7 for Openness.  Analysis of the AFR subsample yielded only 2 significant loci, both for Agreeableness.  Interestingly, none of the EUR loci were significant in the AFR sample, and neither of the AFR loci were significant in the EUR sample -- a point to which I'll return later.

Gupta et al. then combined the MVP-EUR sample with three other large samples of European ancestry, increasing the sample size to roughly 682K for Neuroticism and roughly 250-300K for the other Big5 traits -- thus increasing the power of the analysis, and their ability to detect statistically significant correlations between the Big 5 traits and various loci.  For Neuroticism, the enhanced power yielded an enormous increase in the number of significant loci, from 11 the MVP sample to 208 in the combined sample.  The combined sample also yielded 3 new loci for Extraversion (plus the 11 from MVP-EUR, for the total of 14; 2 loci for Agreeableness (where they had found 3 Agreeableness loci in the MVP-EUR sample alone; 2 for Conscientiousness (where MVP-EUR had also yielded 2), and 7 for Openness (where MVP-EUR had also found 7). 

Why the gains for Neuroticism were so much greater than for the other four Big 5 traits isn't clear.   It might have been due to the fact that the sample for Neuroticism almost trebled in size, while the samples for the other Big 5 traits increased by only about 7-25% (long story here, too technical to get into in this context).  Still, it also isn't clear why the increased power afforded by the larger sample resulted in a reduction in the number of significant loci for Extraversion.

It would also be important to know the extent to which the loci identified in the MVP-EUR sample were also identified in the other samples, and vice-versa. Recall, from the GWAS study of IQ discussed in the lectures on Thinking and Reasoning, Judgment and Decision-Making, that, for all the care and statistical power that went into that experiment, the significant association between IQ and the IGF2R gene, was not replicated in a subsequent study.  xxxxx

By aggregating the correlations between individual loci and their corresponding traits, Gupta et al. estimated heritabilities between 4-8% for each of the Big5 traits.  But recall the finding from Loehlin's twin study, just discussed that genes account for about 40% of the variance in Big5 traits, the nonshared environment about 50%, and the shared environment less than 10%.  In 2005, Loehlin and his colleagues -- including UCB's own Prof. Oliver John (J. Res. Pers., 2005) -- obtained somewhat slightly higher estimates for G and E_NS, and even lower estimates for E_S.  Now, heritability estimates can differ depending on precisely how they are obtained.  In the domain of IQ, for example, studies of identical twins reared apart tend to produce somewhat higher estimates of heritability, compared to the more common twin-study method (Loehlin & Plomin, Beh. Gen. 1989); but 45% vs. 5% -- now, that's a big difference.  Possibly, the heritabilities estimated from twin studies are just wrong, and the heritability of the Big 5 traits is a low lower than previously thought.  Alternatively, the remaining 40% may be accounted for by the cumulative effects of other loci (there are, after all, about 8,000 of them, and roughly 20,000 genes)?  Stay tuned, as the story of the genetics of behavior and personality continues to unfold.

Temperament

A somewhat different pattern occurs for individual differences in temperament, which some theorists consider to be the most "innate" of all personality characteristics. Allport (1961, p.34) defined temperament as follows:

Temperament refers to the characteristic phenomena of an individual's nature, including his susceptibility to emotional stimulation, his customary strength and speed of response, the quality of his prevailing mood, and all the peculiarities of fluctuation and intensity of mood, these being phenomena regarded as dependent on constitutional make-up, and therefore largely hereditary in origin.,

Based on this definition, Buss, Plomin, and Willerman (1973) devised the EASI scale to measure individual differences in four aspects of temperament in MZ and DZ twins aged 4 months to 16 years:

The general finding was that all four dimensions were, as predicted by Allport, indeed, highly heritable, with heritability coefficients averaging .58 -- though there were some differences by age and gender.

Attachment

Temperament" is often considered to be an innate characteristic of personality, so it is not particularly surprising that it has a relatively large genetic component.  Other characteristics of the child, however, show evidence of a clear environmental contribution.  A case in point is attachment style, a term coined by John Bowlby, a British psychiatrist who was influenced by both Darwin's theory of evolution and Freud's psychoanalytic theory of personality.  From Darwin, Bowlby got the idea that the even the very young children have to learn how to survive in their environment -- an environment composed principally of their parents, and especially their mothers as their primary caretakers.  From Freud, he got the idea that the parent-child relationship determined the character of the child's later relationships with other adults, particularly their spouses (and their own children).  Bowlby's attachment theory argues that attachment security is an important feature of personality.  At first glance, it might seem reasonable to assess attachment security on a single dimension, from insecure to secure, in fact attachment theory describes four main types of attachment: "secure" and three different types of "insecure" attachment.  These are typically measured by a behavioral procedure known as the Strange Situation developed by Mary Ainsworth, an American-Canadian psychologist who worked closely with Bowlby. 

The Strange Situation assessment is conducted over a series of phases:

1. The child (typically an infant, between 1 and 2 years old) and parent (usually the mother) are brought into a room. 
   
2. The infant is allowed to explore the room while the parent sits by passively. 
   
3. A a stranger enters the room and converses briefly with the parent.
4. First Separation Episode: The parent leaves the room, and the stranger begins interacting with the infant.
5. First Reunion Episode: Infants are typically distressed at the absence of the parent, and in any event the parent soon returns to the room. 
   
6. Second Separation Episode: The parent and stranger both leave the infant alone in the room. 
   
7. The stranger returns to the room and interacts with the infant.  
   
8. Second Reunion Episode: Finally, the parent returns, picks up the infant, and the stranger leaves.

During the Separation and Reunion episodes, various aspects of the infant's behavior are observed and coded:

• The child's exploratory activity.
• The child's reactions to the departure of the caregiver.
• The child's anxiety level when alone with the stranger.
• The child's behavior during reunions with the caregiver.

On the basis of these behavioral observations, the child is classified into one of four categories:

• Securely Attached: The child approaches the caregiver when s/he returns, and responds positively to the caregivers comforting behavior.
• Insecure Anxious: The child may approach the caregiver, but when s/he does approach, doesn't appear to have been comforted.
• Insecure Avoidant: The child does not become distressed when the caregiver leaves, and they do not respond when the caregiver returns.
  
• Insecure Disorganized: The child shows anxious and avoidant responses in a haphazard fashion.

Obviously, attachment is a two-way street: between the child and the caregiver and between the caregiver and the child.  Therefore, we'd expect the environment to play an important role in individual differences in attachment style.  And this is just what is revealed by twin studies of attachment style.  The question is a little more complicated to answer than for extraversion, IQ, or temperament, because attachment style is a categorical variable, not a continuous variable, but the logic of the analysis is the same: are MZ twins more likely to share the same attachment style as DZ twins?  There have been a number of studies addressing this question (reviewed by Gervai, Child & Adolescent Psychiatry & Mental Health, 2009).

• Finkel & Matheny (2000) found that genetic factors (G) accounted for 25% of population variance, the shared environment (E_S) for 0%, and the nonshared environment (E_N) for 75%.
• O'Connor & Croft (2001): G = 14%, E_S = 32%, E_N = 53%.
• Roisman & Fraley (2008): G = 17%. E_S = 53%, E_N = 30%.

Roissman & Fraley concluded that the most important determinant of attachment style was the quality of parenting received by the child.  Interestingly, given the relatively large contribution of the nonshared environment, the quality of parenting differs even between identical twins!

Most recently, Dugan et al. (JPSP:PPID, 2024) confirmed the importance of the nonshared environment in a large-scale study of elderly adults (N = 678 twin pairs) who had been enrolled in the Minnesota Twin Registry as children, and who completed a questionnaire measure of attachment styles (known as "Experiences in Close Relationships -- Relationship Structures") as elderly adults.  The ECR-RS measures two different aspects of adult attachment: attachment avoidance (e.g., "I do not feel comfortable opening up to people") and attachment anxiety (e.g., "I worry that people may abandon me").  In addition to examining these attachment styles in general, these investigators also assessed the subjects' specific attachments to their mothers, fathers, romantic partners, and best friends.  No matter which of the 10 aspects of attachment was being measured (2 styles x 5 targets), the results were strikingly consistent: genetic factors accounted for approximately 36% of the variance in attachment style, while the nonshared environment accounted for the remaining 64%.  Although the specific formula that Dugan et al. used did not provide an estimate of the shared environmental variance, it's pretty clear (do the arithmetic) that it was negligible.  Similar findings were obtained from the Relationship Scales Questionnaire, another measure of attachment styles in adulthood: genetic effects accounted for 28% of the variance in attachment anxiety, and 36% of the variance in attachment avoidance.  All of the remaining variance on both scales was accounted for by the nonshared environment.

IQ and Education

There's a different picture for individual differences in intelligence (as measured by standard IQ tests) and education.

Family studies show that the correlation between family members' IQ is, in turn, correlated with genetic resemblance.

But there's also a clear family influence:

Aggregating over the best available studies, a reasonable estimate for the MZ correlation for IQ is .86, and the corresponding estimate for DZ is .60. Plugging these figures into the equations, that means that about 52% of population variance in IQ is attributable to genetic variance. About 34% is attributable to the shared environment, and about 14% to the nonshared environment.

When you look at educational attainment, you get somewhat similar correlations, and somewhat similar estimates.

Sex and Suicide

Here are two other examples of how behavior genetics can shed light on the role of nature and nurture in development.

A study by Herden et al. (2007) looked at age of "sexual debut" -- that is, the age at which people had their first sexual intercourse. Here, genes account for about 31% of population variance, but the nonshared environment accounts for a whopping 59% of variance.

And another study, by Fu et al. (2002), examined genetic and environmental contributions to suicidal behavior, defined broadly to include both suicidal ideation and actual attempts at suicide. Genes were more influential on suicidal ideation than suicide attempt (probably due to a genetic influences on depression), but in both cases the contribution of the nonshared environment was much stronger than that of the genes, or the shared environment.

Political Attitudes

There also appears to be a genetic contribution to political attitudes. Evidence for this comes from the Virginia 30K Twin Study, which followed 29,080 subjects residing in Virginia, and their first-degree relatives, including 2,648 MZ twins and 1,748 DZ twins. Among other questionnaires, these subjects were administered the Wilson-Patterson Attitude Inventory of opinions on various socio-political issues, such as school prayer, property taxes, busing, and abortion. Half the questions were posed so that an endorsement indicated liberal attitudes, and half were worded in the conservative direction. 28 of the items were expressly political in nature. The subjects were also asked to reveal their political party affiliation, if any.

The investigators devised two scales of liberalism-conservatism, plus a scale of "opinionation" indicating how strong their opinions were. On each scale, MZ twins were more alike than DZ twins.

Calculating the components of variance, the investigators identified a substantial genetic contribution to both liberal-conservative attitudes and opinionation, though not so much to party affiliation as such (it's possible to be a relatively liberal Republican, and a relatively conservative Democrat). The shared environment was a relatively strong determinant of party affiliation: apparently, children tend to join their parents' political party. But in three out of four cases, the contribution of the nonshared environment was stronger than that of either the genes or the shared environment. Which just goes to show you how important individual experiences are to things like this.

Hatemi et al. (Behavior Genetics, 2014) obtained similar results in an even larger study.  Again employing the twin-study method, these investigators examined "left-right" or "liberal-conservative" political attitudes based on surveys of more than 12,000 twin pairs living in five democracies (Australia, Denmark, Hungary, Sweden, and the United States), surveyed over four decades (1980-2012).  They found that genetic factors accounted for about 40% of the variance in political attitudes; the shared environment, 12%; and the nonshared environment, 42%.  The only departure from this pattern is political-party affiliation, which is overwhelmingly determined by the shared environment.  Democrats tend to beget Democrats, and Republicans Republicans; but peoples attitudes toward specific issues, such as abortion or same-sex marriage, tend not to be passed from one generation to the next. 

Hatemi et al. went even further, conducting a genome-wide association study in three of the samples in an attempt to identify particular gene variants that might mediate the genetic contribution.  They found no plausible candidates, and concluded that individual differences in political attitudes, like individual differences in intelligence and other aspects of personality, were the product of large number of genes, each of which played a small role.  But from our point of view the most important finding is that environmental factors outweighed genetic factors, and the nonshared environment outweighed the shared environment.

As Irving Kristol once joked, "A neoconservative is a liberal who has been mugged by reality". If his identical twin hasn't been similarly mugged, he'll probably stay a liberal.

To sum up: Twin studies reveal genetic influences on personality and attitudes, and these are interesting, but by far the most surprising finding of behavior-genetics research is the evident power of the nonshared environment. We've been taught, at least since Freud (though that should have been our first clue that something might be wrong!) that the way children are treated in the family determines how they'll grow up. In fact, this widespread belief appears to be incorrect. There is an extensive literature on child-rearing practices, considering such things as age of weaning and toilet training (the sorts of things that interested Freud a great deal), and these aspects of childhood seem to have little influence on adult personality (Sears, Maccoby, & Levin, 1957).

Happiness (Subjective Well-Being)

One of the enduring puzzles in personality psychology is what accounts for individual differences in happiness or life satisfaction. Current theory favors the view that each of us has a sort of baseline level of happiness - -think of it as a sort of "happiness set point" around which we fluctuate, depending on what is going on in our lives at the moment (e.g., Kahneman et al., 1999). But where does this baseline level of happiness come from? The obvious answers -- like, "the richer you are the happier you are" aren't sufficient. Research shows a surprisingly low correlation between socioeconomic measures like income or wealth and self-rated happiness. A pioneering study by Lykken and Tellegen (1996) suggested that baseline happiness levels were substantially heritable -- meaning that we are born with a tendency to be happy or unhappy. But Lykken and Tellegen never claimed that happiness was all in the genes.

De Neve (2011; De Neve et al., 2011) examined the genetic and environmental contributions to happiness -- that is, one's sense of well-being and satisfaction with life. in the National Longitudinal Study of Adolescent Health (known as the Add Health Study), a survey of almost 27,000 American students enrolled in 80 high schools in that began in 1994-1995, and continues with regular follow-up interviews. At one point, the subjects were asked to indicate "How satisfied are you with your life as a whole?" -- the standard formulation used in such studies. Most respondents said they were at least fairly satisfied with their lives, but the important point has to do with the determinants of these ratings.

Using a somewhat more complicated formula than our "double the difference" heuristic, De Neve et al. calculated the following components of variance:

So, these findings are consistent with the general pattern found with the Big Five: a significant genetic component, but also a significant contribution from the nonshared environment.

Rumination and Depression in Adolescence

Depression is a serious problem in adolescent (and adult) mental health, and rumination -- perseverative thinking about one's problems and feelings -- is a known risk factor for depression.  Moore and her colleagues studied genetic and environmental contributions to both rumination and depression.  Adolescents enrolled in the Wisconsin Twin Project completed a variety of questionnaires, yielding the following correlations and estimates of components of variance.

Moore et al. used a slightly different procedure to estimate the components of variance.  In each case, it's obvious that there was a significant genetic contribution to variance.  The contribution of the nonshared environment was equally substantial, and the contribution of the shared environment was virtually nil.

Economic Behavior

The financial crisis of 2008 led some social scientists to try to understand how some bankers and investors, working in such a "rational" environment as the economy, could possibly have been so reckless. And, naturally, the question got frames in terms of nature and nurture, genes and environment. A study by Cronqvist and Siegel, two American business professors, analyzed data on savings habits in a sample of almost 15,000 Swedish identical and fraternal twins -- the largest twin registry in existence (again, you can get this kind of data in a country like Sweden, which has a comprehensive and efficient national healthcare system). Of course, Swedes are human, and so there was wide variation in savings behavior -- that is, how much of an individual's disposable income was saved for the future, instead of being spent in the present. The primary measure chosen was change in the individual's net worth between 2002 and 2006, adjusted for individual differences in gross income.

Comparing identical and fraternal twins, they got the following correlations:

Plugging these values into our rough-and-ready formula, we get the following estimates:

Actually, Cronqvist and Siegel applied a number of more sophisticated mathematical models to their data, but they all led to the same conclusion:

The old joke goes: "How do you get to Carnegie Hall?".  and the punchline is: "Practice, practice, practice!".  And it's true: Anders Ericsson and his colleagues (1993), surveying elite musicians, estimated that it took about 10,000 hours of disciplined practice in order to become an expert instrumentalist: this study gave rise to the "10,000 hour rule" popularized by Malcolm Gladwell in his book, Outliers (2008). 

But it's not all practice.  A group led by Miriam Mosing, a behavior geneticist at the Karolinska Institute in Sweden, finds that musical ability is a product of both genes and the environment -- and the critical environment is not the shared environment (e.g., whether the subject's parents were themselves musicians); it's the nonshared environment. 

It makes sense that the shared environment is a stronger determinant of practice than of musical ability per se.  It was probably the subjects' parents who encouraged them to take up music in the first place, and then insisted that they practice -- at least so long as they were living at home!.  But still, the nonshared environment trumps the shared environment.

Putting it All Together

The most comprehensive study of heritability was published by Tinca Polderman, Danielle Posthuma, and their associates in Nature Genetics for 2015.  Indeed, it is the most comprehensive genetic study of anything conceivable, because it's a meta-analysis of virtually all twin studies published from 1958 to 2012 -- covering 2,748 publications, 17,804 traits, and 14,558,903 twins (many of whom appeared in more than one study). These investigators used a slightly different method of estimated heritability, and the contribution of the shared and nonshared environment, than presented in this course, but the results they obtained are entirely compatible with those discussed here.  Still, they determined that, across all domains, the average heritability was 49%, and the contribution of the shared environment 17%, with the non-shared environment accounting for most of the rest of variation.  The largest heritabilities were associated with biological traits, with smaller heritabilities associated with psychosocial traits such as those we're concerned with in this course.  and the shared environment also had the largest influence in the biological domain.  However, all of the traits showed significant heritabilities -- not a single trait, biological or psychosocial, had an average heritability whose confidence interval included zero (0).  Most of the findings were consistent with an "additive" genetic model in which each trait is influenced by a number of different genes.  The table at the right shows the twin correlations, and estimates of heritability (h²) and shared environmental variance (c²) for the "Top 20" most-investigated traits in the literature.

There's a pattern here. For most psychologically interesting behaviors, genes may account for a significant proportion of individual differences, but by far the most important determinant is the nonshared environment.

Genes "for" Personality?

We'll turn to the nonshared environment in a moment, but first, let's explore the genetic component a little more. Is there really a "happiness gene"? Not exactly, but De Neves et al. did identify a particular genetic polymorphism that does seem to be involved in the genetic component.

And it's certainly possible that there are genes "for" certain basic individual differences in temperament, such as speed and strength of emotional response. There may even be genes for individual differences in some of the "Big Five" personality traits, such as extraversion and neuroticism. But there are unlikely to be genes for all the important individual differences in personality, for the simple reason that there aren't enough genes. The human genome contains about 22,500 genes -- and remember that about 90% of these don't vary from one individual to another, which leaves about 2,250 genes to work with. And when you consider that, in order to get the smooth bell-shaped normal distributions of personality traits, you need a model of polygenetic inheritance -- lots of genes making tiny contributions to each individual-difference variable -- you can see that the number of genes mounts up pretty quickly, and you pretty quickly run out of genes. So there's got to be something else going on, and that "something else" is going to be powerful, not trivial.

Moreover, consider the nature of some of the individual differences studied by behavior geneticists.

Epigenetics

Identical twins, whether raised together or apart, are strikingly similar on a host of psychological variables. But they're also strikingly different, even when they were raised together. Why should this be the case?

One answer lies in the complexities of genetic inheritance: it turns out that individuals with the same DNA might not be genetically identical after all. According to epigenetic theory, genes carry epigenetic tags, such as histones and methyl groups, that do not change an individual's DNA, but rather enhance or suppress the activity of particular genes -- turning them on or off, if you will. It's by virtue of epigenesis that embryological development happens -- how the single cell of the newly conceived zygote begins to differentiate into the specific cells that make up the various body parts of the developing fetus. And it's the basic mechanism for many forms of gene therapy -- inserting a gene into one (biochemical) environment will have one effect, while inserting the same gene into a different biochemical environment can have a quite different effect.

But the scope of epigenesis goes beyond embryological development. The result of epigenesis is that while two individuals can have the same genotype, they can have different phenotypes even at the cellular level of analysis. The result can be -- here I'm taking an extreme example for purposes of illustration -- that two MZ twins who share the genotype for blue eyes, but one might end up with brown eyes. Further, according to theory, environmental factors like stress (war, child abuse, even social prejudice), poor nutrition, or other forms of deprivation (such as the economic deprivation of poverty), which can alter the chemical environment of the gene, might affect the activity of these epigenetic tags.

The interesting thing is that some of these environment-induced changes in gene expression -- essentially, turning some genes on or off -- can themselves be heritable.  So, for example, the effect of stress on one identical twin can alter that individual's genome, and perhaps the genes that he or she passes on to his or her offspring -- providing a genetic mechanism for the intergenerational transmission of the acquired characteristic.

Here's an example of how epigenetic influences are studied (in mice, in this case; obviously you can't do this kind of study in humans), in an elegant experiment by Darlene Francis, a neuroscientist in UCB's School of Public Health, and her colleagues (Nature Neurosci. 2003). 

Epigenetics is sometimes thought to prove Lamarck's idea of the inheritance of acquired characteristics -- that "acquired" modifications to the body (or, for that matter, behavior) can be passed on from parents to children.  It doesn't and they can't.  The prenatal environment can affect whether certain certain genes are turned on or off, and in this way what happens to a parent can affect the consequences of a child's genetic endowment.  But that's not the same thing as the Lamarckian vision -- in which giraffes grow long necks to eat the tender leaves from the high branches of trees.  Besides, it's not at all clear that the altered genome can be transmitted genetically.  The altered genome will be preserved during mitosis, or ordinary cell division, but it is not at all clear that it will also be preserved during meiosis, in which cell division produces sperms and eggs.  Whatever intergenerational transmission occurs may be by virtue of pure environmental mechanisms -- biochemical, maybe, but environmental nonetheless.

In its broadest sense, epigenetics refers to everything that determines an individual's phenotype, other than his genotype. As such, it describes the effects of the environment on gene expression, and might be counted as an example of the person-by-situation interaction discussed at length in the lectures on Personality and Social Interaction.  That is, some environmental factor, such as stress levels, interacts with some aspect of the individual's genetic endowment, to alter some phenotypical aspect of behavior.  ,

In general, the synergistic interplay of genes and environments appears in two broad forms (for examples in the context of attachment, see Dugan et al., JPSP:PPID, 2024):

• Gene-Environment Correlations (known as rGE in the trade) occur when some gene -- or, more precisely, individual differences in some heritable trait -- directs the individual toward exposure to particular environments.  For example, a person carrying a gene for aggressiveness tends to find him- or herself in environments where aggression is likely to occur. 
  
• In passive rGE, the genetic and environmental contributions to an individual's behavior have the same source -- e.g., parents who have a gene for hostility may also create a hostile home environment.
  
• In active rGE, for example, individuals actively select environments that are compatible with their genetic dispositions -- e.g., an introverted individual may prefer quiet, solitary environments like libraries.
• In evocative rGE, individuals evoke behavior from others that create an environment that is compatible with their genetic tendencies -- e.g., an extraverted person may lead others to engage in boisterous behavior at parties.
  
• Gene-Environment Interactions (GxE) occur when the presence or absence of a particular gene alters or moderates the effect of the environment on the individual, or vice-versa -- when a particular environment affects (favors or inhibits) the expression of a particular gene. 
  
• A person who is genetically predisposed to alcohol abuse may be more likely to develop alcoholism when exposed to an environment in which there is a lot of heavy drinking, compared to someone who assiduously avoids such settings.
• Alternatively, an environment characterized by a great deal of hostile, aggressive behavior may be more likely to bring about hostile and aggressive behavior in an individual who is genetically disposed to hostility and aggressiveness, compared to an environment which encourages passivity and quietude.
  

We'll see more examples of the gene-by-environment interaction in the lectures on Psychopathology and Psychotherapy.  But epigenetics is a biological construct, and it has only the vocabulary of biochemistry -- histones, methyl groups, stress hormones -- to describe that environment. But as psychologists, we are primarily interested in a different level of analysis -- the environment construed in psychological terms, as the person's mental representation of the environment. In psychological terms, what is important about stress are the levels of unpredictability and uncontrollability that cause it to occur; and whether the person subjectively experiences an environment as stressful -- regardless of whether it is "objectively" stressful. And what goes for stress also applies to other aspects of the environment: psychologists are always centrally interested in the individual's mental representation of the environment. How that mental representation is represented neurally, and what effects it will have biochemically, reflect a different level of analysis.

For a good account of epigenesis, see The Epigenetics Revolution: How Modern Biology is Rewriting Our Understanding of Genetics, Disease, and Inheritance by Nessa Carey (2012). See also Carey's articles on "Epigenetics in Action" in Natural History magazine: Part 1, "Deciphering the Link Between Nature and Nurture", appeared in the April 2012 issue; Part 2, appeared in May 2012.

See also "Epigenetics: The Evolution Revolution" by Israel Rosenfield and Edward Ziff, New York Review of Books, 06/07/2018 -- a sequel of sorts to their earlier essay, "Evolving Evolution", NYRB, 05/11/2006.

For a more technical overview, see "Effects of the Social Environment and Stress on Glucocorticoid Receptor Gene Methylation: A Systematic Review" by Gustavo Turecki and Michael Meaney, Biological Psychiatry, 2016.

In a somewhat similar manner, so-called jumping genes are segments of DNA that can replicate and insert themselves in new places in the genome, altering the activity of their neighbor genes. Jumping genes appear to be particularly active in the brain, leading some neurogeneticists to suggest that they are the key to human uniqueness: even identical twins are not, precisely, genetically identical, and these small differences in the genome are responsible for the differences in personality observed in MZ twins -- including differences in the concordance rates for various forms of mental illness, such as depression.

Well, maybe. Epigenetic theory is plausible, and it might even be true, in some cases. Epigenetic factors have been invoked to account for the relatively low concordance rates between MZ twins for certain mental illnesses, such as schizophrenia and autism (as discussed in the lectures on Psychopathology and Psychotherapy). But note that epigenetic theory defines "the environment" in purely biological, even biochemical terms -- not to put too fine a point on it, as the microenvironmental soup that the gene sits in. As psychologists, we're primarily interested in the social and cultural macroenvironment in which the individual person lives.

Frankly, by focusing on epigenetic factors, behavior genetics sometimes seem to want to have it all -- to be able to explain even differences between genetically identical individuals in genetic, biochemical terms. I noted a similar trend with respect to junk DNA.  Put another way, there is a certain class of biologically oriented individuals who want to discount the effects of the social and cultural environment at all costs. That's fine, if you're a biologist, because it's the very nature of the biological level of analysis to construe the environment in physical, chemical, and biological terms. But psychology is a social as well as a biological science, and so psychologists take a broader view of the environment -- one which considers social and cultural influences on their own terms, without reducing them to biochemistry.

Or, as E.B. White and Carl Rose (and, later, Irving Berlin) put it, "I say it's spinach (and the hell with it)". Let's look at the macroenvironment, or the social and cultural world in which the individual develops, from birth across the entire lifespan.

For an accessible discussion of epigenetic theory, see:

• "Hidden Switches in the Mind" by Eric J. Nestler,Scientific American, 12/2011.
• The Epigenetics Revolution: How Modern Biology is Rewriting Our Understanding of Genetics, Disease, and Inheritance (2012) by Nessa Carey.  Excerpted in Natural History magazine, April-June 2012.
  

For a discussion of jumping genes, see:

• "What Makes Each Brain Unique" by Fred H. Gage (a distant relative of the famous Phineas Gage) and Alysson R. Muotri Scientific American, March 2012.

For excellent coverage of the modern science of heredity and genetics, see:

• She Has Her Mother's Laugh: The Powers, Perversions, and Potential of Heredity by Carl Zimmer, a prominent science journalist.

Genetic Nurture -- or, "Unto the Next Generation"

One of the paradoxes of behavior genetics is that the same study that provides evidence of genetic contributions to behavior also can give evidence of environmental contributions.  So, for example, the twin-study method allows us to estimate the effects of heredity, the shared environment, and the nonshared environment.  Similar insights can come from the gene-wide association study (GWAS) method described earlier, which allows us to identify particular genes -- or, more precisely, particular alleles -- associated with particular characteristics.  For example, a recent study by Augustine Kong and his colleagues (Science, 2018), employing a huge sample of Icelandic probands (as subjects are commonly called in genetics research), to identify a number of different alleles which were significantly associated with educational achievement (EA, measured by years of education completed).  Recall that children receive half their genes from their fathers and half from their mothers.  Therefore, it follows that some of these alleles will be passed down from parents to their children, but others will not.  For each subject, Kong et al. identified which EA genes (as we'll call them) were transmitted by each parent to each of his or her children, and then examined the contribution of both transmitted and non-transmitted EA genes to EA in their children.  They found that direct genetic transmission from parent to child accounted for a significant proportion of the variance in the children's EA, as would be expected from the twin studies summarized earlier.  But there was also a significant effect of the parents' non-transmitted EA genes on their children's EA -- an effect about 1/3 as strong as the direct genetic effect.  Kong et al. call this effect genetic nurture -- that is, by virtue of the EA genes that they possess, the parents create an environment that affects EA in their children over and above the direct effect of the EA genes the parents have passed down to them.  This genetic nurture effect is one component in the shared environment of the children. 

If you think about it, genetic nurture effects involve more than parents and children.  Children get roughly 1/4 of their genes from each of their grandparents, too.  So some EA alleles are transmitted from grandparents to grandchildren directly, through the parents, but the non-transmitted genes also affect the environment in which the parents were raised, and also -- to the extent that the children are in contact with their grandparents -- the environment for the children.  And remember that children in the same family share about 50% of their genes in common.  So some parental EA genes, will get passed to proband John, but not to his siblings Jean and Don, and vice-versa.  But by virtue of the EA genes they have received, Jean and Don will create an environment that affects John's EA.  And vice-versa.  

Remember that children get half of their genes from each parent on average.  Now imagine that John didn't get his fair share of EA genes, but that Jean and Don got more than their fair share.  John might be genetically disadvantaged in this sense, but Jean and Don will use their genetic advantage to create an environment that promotes John's EA.  And if John got more than his fair share, and Jean and Don got less, they might create an environment that holds John's EA back. This is one of the ways in which the family environment ostensibly shared by siblings is not the same for each of them.  Now let's explore some other ways.

The Nonshared Environment

The fact is that, with respect to the major dimensions of personality, variability within a family -- i.e., the variance among children raised in the same family environment -- is almost as great as the variability between families -- i.e., the variance between children raised in different families. This result is sometimes misinterpreted as meaning that parents have no influence on their children. But it doesn't mean that at all.

What it means, first, is that there are other forces at work besides the parents, and these become increasingly important as the child begins to move beyond the family (e.g., by going to school or joining the soccer league)

In addition, it means that parents don't have the same effects on each of their children.

These differences in the nonshared environment are critical for our individual uniqueness. Put another way, the most important environmental determinants of personality are the unique experiences that we have in our lives.

The Nonshared Environment

One source of the nonshared environment, of course, is the extra-familial environment. Even if all children within a family are treated precisely alike by their parents, once they begin to interact with the world outside the family they will have experiences that tend to make them different.

As examples of extra-familial influence, consider a series of studied by David Rowe and his colleagues (e.g., 1992), using behavior-genetic methods, of the sources of various aspects of adolescent behavior.

Rowe and his colleagues found similar causal patterns for alcohol consumption, delinquency, sexual behavior, and pregnancy. In each case, the shared environment had relatively little impact on behavior, but peer groups had a powerful effect on whether adolescents experimented with tobacco, alcohol, sex, and misbehavior.

Similarly, a study by Kindermann (1993) of academic motivation among elementary-school children found that students tended to group themselves into cliques such as "brains" and "slackers" -- but also that membership in these cliques tended to be somewhat unstable, with individual children moving back and forth from one group to another. Interestingly, the children's attitudes toward school changed as they changed cliques. "Brains" lost interest in school if they moved to a "slacker" clique, and "slackers" gained interest if they moved to a "brains" clique -- despite the fact that their IQs remained constant, and parental influences presumably did so as well. It was the peer group that caused the changes to occur.

Based on results such as these, Harris (1995, 1998, 2006) has proposed a theory of group socialization which argues that peer groups and peer cultures, not parents, are the most powerful socialization forces impinging on a child. In fact, Harris argues that socialization is context-specific, and that children may behave quite differently depending on whether they are at home or away, and depending on which particular extra-familial context they're in.

As an example, she points to "code switching" among bilingual and bicultural children (as discussed in the lectures on "Language"):

Even within the family context, however, there are important differences among siblings. Harris (1995, 1999, 2006) has classified these within-family differences into four categories:

In addition to these within-family differences, there are also extra-familial influences that influence the development of personality and social behavior.

Harris on Parental Influence In The Nurture Assumption: Why Children Turn Out the Way They Do (1998), Harris reproduced a famous bit of verse by Philip Larkin (1922-1985), an English poet.  Now, Larkin had issues: he once quipped that "Sex is too much fun to share it with others".  In "This Be the Verse" (1971/1974) Larkin reflected the dominant view of personality development in his time, influenced as it was by Freudian psychoanalysis -- that parental influence dominates the development of personality. They fuck you up, your mum and dad. They may not mean to, but they do. They fill you with the faults they had And add some extra, just for you. But they were fucked up in their turn By fools in old-style hats and coats, Who half the time were soppy-stern And half at one another’s throats. Man hands on misery to man. It deepens like a coastal shelf. Get out as early as you can, And don’t have any kids yourself. To which, Harris replied with a bit of verse of her own: Poor old Mum and Dad: publicly accused by their son, the poet, and never given a chance to reply to his charges. They shall have one now, if I may take the liberty of speaking for them: How sharper than a serpent’s tooth To hear your child make such a fuss. It isn’t fair—it’s not the truth— He’s fucked up, yes, but not by us. Then by whom?  Peers and other elements of the nonshared environment.

Child-Driven Effects

Child-driven effects, also known as reactive effects, relate to the fact that each child brings certain physical and behavioral characteristics into the family, which in turn affect how he or she is treated by parents and others.

Some reactive effects reflect the environment's response to the physical appearance of the child. These reflect the evocation mode of the person-by-situation interaction.

Other reactive effects are instigated by the behavior of the child, not just his or her physical appearance. These reflect the manipulation mode of the person-situation interaction: there is something the child does, however unwittingly or involuntarily, to alter the environment in which he or she is raised.

Of course, such child-driven effects may extend beyond the child's interactions with his or her parents, as children begin to venture outside the home to school, playgroups, sports programs, and the like. Thus:

In general, child-driven effects are unpredictable, because they greatly depend on the response to the child by the people who make up the child's environment -- their own personalities, beliefs, attitudes, goals, and the like.

Parent-Driven Effects

Parents don't just react to their children's appearance, behavior, or temperament. To some extent, parental behavior is independent of the physical, mental, and behavioral characteristics of the child. For example:

Relationship-Driven Effects

In statistical terms, we can think of child-driven and parent-driven effects as main effects.  So, with two main effects, come the interaction between them.  Relationship-driven effects have to do with the "fit" between child and parent in terms of appearance and temperament. For example:

Relationship-driven effects are related to the selection mode of the person-by-situation interaction, in that they involve the degree of "fit" between the child and the other people who make up his or her social environment.

For a journalistic account of children whose characteristics did not "match" with those of their parents, either by virtue of some disability (e.g., deafness, Down syndrome, autism, or schizophrenia) or some talent, see Far from the Tree by Andrew Solomon (2012; reviewed by Nathan Heller in "Little Strangers", New Yorker, 11/19/2012; also by Julie Myerson in "Coming Into Their Own", New York Times Book Review, 11/25/2012).  Parents and children adjust to each other, and that's Solomon's point, but it isn't always easy.  For one thing, children may have an identity that is different from their parents.  The parents might be able-bodied,. the child disabled in some way; the parents might be hearing, the child might be deaf; might be straight, but the child gay; the parents might be Catholic, but the child has converted to Islam.  These parent-child differences in horizontal identity can make it hard for parents and children to get along.

An interesting twist on relationship-driven effects is illustrated by a recent behavior-genetic study which shows how child personality can affect parenting behavior (Ayoub et al., Social Psychological and Personality Science, 2018).  In a study employing 1,411 children, twins and triplets, enrolled in the Texas Twin Project, the researchers (using an alternative genetic model to the one discussed earlier in these lectures), first confirmed that there is a significant genetic contribution to the Big Five personality traits -- but also that the nonshared environment accounted for 62-72% of the variance in childhood personality.  But in a variant on the usual twin-study method, the investigators also examined MZ and DZ correlations on parental warmth and stress.  The finding was that identical twins received more similar parenting than fraternal twins did, especially with respect to parental stress variable.  For example, children who scored high on agreeableness received more parental warmth and less parental stress than did those who scored low.  The children elicited different parental behaviors by virtue of personality characteristics that were partly heritable.  Childhood personality accounted for only a relatively small proportion of variance in parenting warmth and stress, but the effect was significant.  As the authors conclude, "parenting is a dyadic and dynamic process, whereby both parents and children influence each other".

Family-Context Effects

Family context effects relate to the children's "microenvironments" within a family. For example, in my family, there was a father, a mother, a girl and two boys. Therefore, my family microenvironment consisted of my parents, my sister, and my older brother. But my brother's family environment was different -- it consisted of my parents (who of course were also his parents), but it also included my sister and me, his younger brother. Similarly, my sister's family environment consisted of our parents and my brother and me. Different people in each environment. Put bluntly, I grew up in an environment that included a very popular cheerleader and a varsity basketball player; my sister and brother didn't. It's the same for every child in every family.

Birth-Order Effects?

Among the most controversial family-context effects involve birth order -- that is, systematic differences in personality between first-born and latter-born siblings in a family. Because there are no systematic genetic differences between first-borns and latter-borns (all brothers and sisters share a random 50% of their genes in common), any systematic differences between them must be due to their position in the family constellation.

But are there any such systematic differences owing to family constellation?

Until recently, most researchers held that birth-order effects were weak or inconsistent (Schooler, 1966; Ernst & Young, 1983). To be sure, there were occasional studies that demonstrated personality differences between first-borns and latter-borns, but there were lots of confounding variables that made the studies difficult to interpret:

Birth Order and Personality

None of these problems are intractable, however, provided that your sample sizes are large enough. And there are reasons for being interested in the possibility of birth-order effects on personality. For example, Frank Sulloway (1996), a historian of science who dabbles in evolutionary psychology, has argued that, in Darwinian terms, siblings complete with each other for their place in the family environment -- just like species and organisms compete for their environmental niches in nature.  This is known as the Family Niche Theory.

Sulloway, trained as a historian of science, found that scientists who made revolutionary contributions to their fields tended to be latter-borns. This led him to become interested in the wider psychological literature on birth-order and personality. In fact, a "meta-analysis" (i.e., a form of quantitative literature review that summarizes and aggregates the outcomes of many studies) performed by Sulloway, revealed systematic birth-order effects on personality:

Technically speaking, Sulloway counted the number of comparisons on each dimension that gave positive, negative, or null findings with respect to his Darwinian hypotheses. For each Big Five dimension, he found that far more studies supported his hypothesis than contradicted it. Lots of studies yielded unambiguous findings, though, leading to some controversy over his interpretations.

The power of Sulloway's meta-analysis comes from the fact that he made his predictions in advance, based on his reading of Darwinian theory. However, his analysis involved a lot of interpretation, and it would be nice to see his findings confirmed in a study expressly designed to test his hypotheses.

Such a study was performed by Paulhus et al. (1999), based on samples drawn from student populations in both California and Canada. Paulhus asked subjects to think about themselves and their siblings, and to nominate who in their family was the "achiever" and who was the "rebel". In both samples, Paulhus found that subjects were more likely to nominate the first-born child in their family as the "achiever", and a laterborn child as the "rebel", than we would expect by chance.

In two other studies, Paulhus and his colleagues also found other respects in which firstborns differed from laterborns. For example, there were more firstborns nominated as the "scholastic achiever" in the family, and more laterborns nominated as the "liberal" in the family, than we would expect by chance. These departures from statistical expectations are sometimes small, but these small effects accumulate to provide significant support for Sulloway's hypotheses.

Note that Sulloway and Paulhus found significant personality differences between first- and laterborn children, but these differences do not necessarily validate Sulloway's "Darwinian" theory. The personalities of first- and laterborn children may differ in significant ways for reasons that have nothing to do with competition for an environmental niche. There may be other expectations. For example, parents may impose their own expectations more strongly on the firstborn, and give laterborns more freedom. Royal families (like the House of Windsor in England) want to produce "an heir and a spare", but once the heir proves up to his assigned job, his younger sibling may be given a great deal of freedom to pursue his own interests. In England, Edward looks like a rebel only because Charles is doing his duty.

Birth Order and Intelligence

Perhaps the most controversial claim about birth-order is that firstborns are more "intelligent", as measured by standard "IQ" tests, than laterborns. It was this hypothesis that was specifically rejected by the Schooler (1966) and Ernst & Young (1983) studies cited earlier. However, a provocative study by Zajonc and his colleagues has revealed an interesting (if small) effect of birth order on general intelligence. These effects, however small, has led Zajonc to develop a confluence model of development that recapitulates the major themes of these lectures:

The first study involved an analysis of data collected in the Netherlands as part of a study of the effects of the Dutch famine of 1944 (Zajonc & Markus, 1974). As part of routine testing for the military draft, the Dutch government administered a nonverbal IQ test (Ravens Progressive Matrices) to every Dutch male who reached 19 in the years 1963-1966. Zajonc and Markus then plotted mean IQ scores as a function of both family size and birth order.

The results revealed a significant interaction of birth order with family size on IQ. Specifically:

Before we go on, please note that the effects just described are very small. Note the Y-axis on the accompanying figure: the difference between the top and bottom means amounts to only about 10 IQ points. The differences noted above achieve statistical significance only by virtue of the huge sample size involved.

Given the fact that individual differences in IQ are only weakly related to social outcome in the first place, the differences revealed in Zajonc & Markus's analysis are of no practical significance. However, as we will see, they are of considerable theoretical import. Big theories can be built on small effects, and that is as true in psychology as it is in physics.

The Confluence Model

In order to explain the joint effects of family size and birth order on IQ, Zajonc and Markus proposed a confluence model of intellectual development. This model traces the mutual intellectual influences among children, and their parents, as they develop. The major features of the model are as follows:

The actual effects of dilution and growth depend on the spacing of the siblings.

• If siblings are spaced closely together, the dilution effect is increased.
• If siblings are spaced farther apart, the dilution effect is weakened.
• In large families, some earlyborns are much older than some laterborns. Therefore, the dilution effect is weakened for these laterborn children.

The theory makes the interesting prediction that twins and triplets should be even more disadvantaged than only children, because their birth produces a big dilution effect. This is, in fact the case, but of course the outcome depends on details of birth order, spacing, and the like. Some other implications of the theory:

The confluence theory tells only part of the story of intellectual development . There are lots of other things going on, both genetic and environmental. But the major assumptions of the theory illustrate the basic points made throughout these lectures on personality, social interaction, and psychological development:

The person is a part of his or her own environment.

The person is an agent of his or her own development.

The nonshared environment is such a powerful force in personality development because everyone creates a unique environment for him- or herself, in interaction with other people, through the various modes of person-by-situation interaction:

The uniqueness of the individual's environment is the sum total of many different effects, acting alone and in combination. And the uniqueness of the environment, as it is shaped by the individual him- or herself, in turn contributes to the uniqueness of the individual.

There is a sort of paradox in development, which is that the universal processes of development are universal, but these widely shared processes come together in such a way as to produce a unique individual.

The implications of these interactions for development are profound, because they mean that, in psychological terms:

Every child is born to different parents, raised in a different family,

lives in a different neighborhood, attends a different school, and worships in a different church.

In medieval times, philosophers debated the idea of contingentia mundi -- the idea that the world world we actually live in is not the only one that is possible.  And we now know, scientifically, that this is the case: if that comet had not struck the Earth, killing the dinosaurs, things would have been very different for us humans (and other mammals!).  What goes for the Earth goes for the individual as well.  The people we are are not the only ones that are possible.  Each of us is contingent: we have been shaped by a whole host of forces, each of which could have been different from what it was; and we have shaped ourselves, and each of us could have made different choices than the ones we made.  Personality, like the world itself, is contingent.
 

Gender Dimorphism

The interaction of nature and nurture can be seen when we look at the development of gender dimorphism:

Some role differences, known as the procreative imperatives, are built into us by our biology:

But gender role goes beyond the demands of reproduction to include, at least in this culture:

It is now very clear that this developmental process is not a simple matter of genetic determination. Rather, it reflects a complex interaction between genetic/biochemical processes (the phyletic imprimatur) and social/environmental processes (the social imprimatur) -- an interaction that is made more complex by the fact that

the developing child is both a target and an instigator of his or her own development.

Gender Differentiation in Fetal Development

As noted earlier, the normal human cell possesses 46 chromosomes, arranged in 23 pairs. Two of these, making up a single pair, are the sex chromosomes, known as X and Y. Normally, males carry one X and one Y chromosome (XY), while females carry two X chromosomes (XX). Genes for sex-linked traits are located on these sex chromosomes. Note that because each parent contributes one chromosome of each pair to his or her offspring, and the mother can contribute only X chromosomes to this process, in the final analysis the father determines the sex of his child: if he contributes an X chromosome, the child will be genetically female (XX); if he contributes a Y chromosome, the child will be genetically male (XY).

Although the fetus is genetically male or female from the beginning (because its cells carry the XX or XY chromosome pairs from the beginning), early in gestation the fetus is otherwise undifferentiated with respect to gender. That is, although it carries the XX or XY chromosomal endowment, it has no outward appearance of being male or female. This is because at this stage the fetus's gonadal tissue is undifferentiated. . Remember the debate between recapitulation and differentiation as basic themes in development? In these terms, the undifferentiated gonadal tissue of the early fetus is a primordial structure which will eventually differentiate into the more complex structures representing the male and female reproductive anatomy).

In technical terms, the structures in this undifferentiated gonadal tissue contain the anlagen (or foundation) of the male and female reproductive systems:

After about six weeks of gestation, sexual differentiation begins. In response to genetic messages (carried on the X and Y chromosomes), one set of structures begins to develop while the other one becomes vestigial. If the fetus carries the XY genotype, the inner medulla will grow into the testes of the male, and the outer cortex regresses; if the fetus carries the XX genotype, the outer cortex will grow into the ovaries of the female, while the inner medulla vestigiates.

The genes themselves appear to play no further role in what happens. Rather, further sexual differentiation occurs by virtue of hormones secreted by the gonads -- and in particular, those male hormones secreted by the testes. There may be a role for the female hormones in genetically XX fetuses, but this is not clear at present. As a rough approximation, further sexual differentiation appears to reflect what the biologists call "nature's rule":

add something to masculinize.

Without masculinization instigated by the male gonadal hormones, the remaining gonadal tissue will naturally differentiate into the female reproductive system. So, in a sense, at this point the program for sexual dimorphism passes from the genes to the hormones.

The hormones secreted by the testes have effects on other structures in the initially undifferentiated gonadal tissue (again, there may be independent effects of female hormones in genetically XX individuals, but this is a controversial point in endocrinology).

When everything goes as programmed, after nine months of gestation a human baby is born with a set of external genitalia that are recognizably male or female, and a corresponding set of male or female internal reproductive organs.

Anomalies of Gender Differentiation

But sometimes things don't run quite the way they're programmed, and the child is born sexually ambiguous -- individuals who are known technically as pseudohermaphrodites.

Chromosomal XX Individuals. If a genetic female somehow experiences an environment to which androgen has been added, she will be born with female internal genitalia, but most likely an enlarged clitoris and fused vaginal labia; rarely, such a girl will be born with a normal penis and scrotum (of course, the scrotum will be empty, because there are no testes to descend into it). This occurs in two principal ways.

In both cases, the children are raised as girls.

Chromosomal XY Individuals. In a genetic male, the failure of the Mullerian-inhibiting substance can leave the fetus with a set of male external genitalia, but both the female and male internal reproductive systems. Note, however, that such an individual has only testes (the gonadal tissue becomes either testes or ovaries); thus, he cannot menstruate or gestate. The children are raised as boys.

Guevodoces. An interesting syndrome, originally discovered in an isolated area of the Dominican Republic (but also documented in an isolated village in Puerto Rico), involves genetically male individuals (chromosomal XY) who are born with a particular defect in their androgen system known as 5-alpha-reductase deficiency syndrome. Because they do not undergo masculinzation in utero, these children are born with apparently female external genitalia; if the condition is undiagnosed, they are raised as girls. At puberty, however, the flow of natural testosterone induces masculinization: the voice deepens, the child develops a typically masculine muscular structure, breasts do not develop as expected and -- surprise! -- the child's scrotal tissue balloons, testes descend, and what originally appeared to be a clitoris enlarges into a functioning penis. Hence the popular name for this condition --guevodoces.

What's especially interesting about this syndrome is that the children readily shift their gender identities, and corresponding gender roles, from feminine to masculine. This is because their culture is prepared for the possibility of a spontaneous "sex change" from previous cases known in the village -- it's as if they say "Oh, that happened to Uncle Jose, too!". The boys' adolescent and adult behavior, including sexual behavior, is not are appreciably different from that of "normal" males.

Before the arrival of modern medicine, the condition went undiagnosed until adolescence. Now, the condition is diagnosed at birth, either through chromosomal testing or through palpation of the groin (which reveals the undescended testes), and the children are identified and raised as boys, from birth -- appearances to the contrary notwithstanding.

Jeffrey Eugnedies' novel Middlesex (Farrar, Straus & Giroux, 2002; he also wrote The Virgin Suicides) uses the fictional life of Cal (nee Calliope) Stephanides, a Greek-American with "guevodoces syndrome", as a metaphor for the identity crises of immigrant, "hyphenated" ethnic Americans, as well as places "like Berlin, like Korea, like all the other places in the world that were no longer one thing or the other".

Klinefelter's Syndrome, 47 XXY. In another condition, a chromosomal male has an extra X chromosome, thus 47XXY rather than 46XY (this occurs in fewer than 1/500 male births). Common consequences include "feminized" physique, infertility, delayed motor and speech development, difficulties with reading and writing. As adults, these individuals often gradually lose both sexual potency and interest. Many of these symptoms can be reversed with hormone replacement therapy, replacing the testosterone that is missing naturally.

These anomalies of gender differentiation sometimes raise the question of gender identity: is a person male or female? And not just a question about how the individual identifies him- or herself -- but also issues of how s/he is identified by other people. Because of previous gender-related controversies -- including the fact that, in the 1936 Berlin Olympic Games, the Germans cajoled a male athlete, Hermann Ratjen, into living as a woman for three years before entering "her", renamed Dora, into the high jump competition ("she" lost); and the practice, in certain countries of Communist Eastern Europe, of doping female athletes with testosterone and other steroids in order to enhance their performance -- in the 1970s the International Olympic Committee began testing female athletes to confirm their "femaleness" by inspecting their chromosomal material for the presence of a telltale X chromosome. But Stella Walsh, a Polish sprinter who also competed in the 1936 Olympic games, apparently suffered from androgen-insensitivity syndrome -- although chromosomally male, she identified herself as a woman and had lived as a woman all her life. Under current rules, she would have been disqualified from competition. At the same time, since 2004 the IOC's rules have allowed transsexual women -- that is, chromosomally male individuals who identify themselves as women and have undergone sex-reassignment surgery and post-operative hormone-replacement treatment -- to complete as women (though as of 2008, no openly transsexual individuals have qualified for the competition). Like Stella Walsh, these individuals would also have failed a chromosomal test of gender [see "The XY Games" by Jennifer Finney Boylan,New York Times 08/03/08.].

All of which raises the questions: What is the proper criterion for being male or female? Chromosomal sex? Body morphology? Gender identity?  And how many categories of gender are there, anyway?  The usual two? Or are there at least two more, to cover conditions like androgen-insensitivity?

Hormonal Effects on Mind and Behavior

Pseudohermaphroditic children are of interest because they are genetic females who experience the effects of male hormones, and genetic males who do not experience these effects. Therefore, at least in principle, these the children provide an opportunity to observe the effects of prenatal hormones on behavior -- observations that might provide evidence of a hypothetical "masculinization of the brain" underlying the differences between masculine and feminine gender roles. And, in fact, there is some evidence for hormonal effects on behavior. Thus, for example, girls with the female adrenogenital syndrome and progestin-induced pseudohermaphroditism generally appear more vigorous and aggressive -- in terms of gender role stereotypes, more "tomboyish" -- than control girls; by comparison, the genetic males with the androgen-insensitivity syndrome, who are raised as girls, are behaviorally indistinguishable from control girls. This evidence is controversial, however, and shouting matches regularly erupt when it is presented and discussed at scientific meetings.

Sometimes, pregnant women who suffer from severe diabetes will be treated with exogenous estrogen and progesterone to prevent miscarriage. These female hormones do suppress the action of the androgen that would normally circulate to genetically and hormonally male fetuses, but appears to have no effects on the external or internal reproductive anatomy. These children are, of course, raised as boys. They are relevant to this discussion only because some early literature indicated that they were somewhat less aggressive and competitive than other boys. However, it should also be noted that the mothers of these children are also sicker than the mothers of control boys, and this by itself may inhibit vigorous play -- and effect which has sometimes been attributed to the lack of "masculinization of the brain". When we add controls for race, age, social class, and especially maternal illness during pregnancy, however, the difference disappears. Therefore, the behavioral differences appear to be caused by environmental, rather than biological factors.

In any event, this process, sometimes called the phyletic imprimatur, leaves the developing fetus and newborn child with a set of external genitalia, and an internal reproductive system, that are more or less recognizably male or female. At this point, the influence of biological factors stops, temporarily, and the individual's biographical history takes over as the parents and others structure an environment corresponding to their conceptions of how boys and girls should be raised -- an important part of gender-role socialization referred to as the social imprimatur.

In other words, the program for sexual differentiation (or gender dimorphism) passes from the hormones to the (social) environment). However, the program will be passed back to the hormones later, at the time of adolescence (for both sexes), and again at the time of menopause (for females). Actually, from birth on, it is probably best to think of the program being passed back and forth between the hormones and the environment. This is what is meant by the interaction of nature and nurture.

Postnatal Hormonal Influences

The effects of the sex hormones do not stop with the differentiation of the internal and external genitalia. They come back on the scene at least two more times, at puberty and in old age, each time interacting with the social environment.

Puberty. At puberty, the program for gender dimorphism passes back to the hormones, as indicated by such milestone events as menarche (onset of menstruation) in girls and nocturnal emissions ("wet dreams") in boys. The most obvious post-natal effects of the sex hormones are the development of such secondary sex characteristics as the deepening of the voice in males and the development of breasts in females, and the masculinization or feminization of overall body shape. These physical changes are instigated by sex hormones, testosterone and estrogen, secreted by the testes and ovaries, respectively.

Interestingly, there is now evidence that puberty may begin long before adolescence. In girls, for example, the pituitary hormones associated with puberty, along with secretions from the ovaries, are known to begin at about age 9, and breast development often begins between 10 and 11 years of age. Martha McClintock and Gilbert Herdt, researchers at the University of Chicago, have noted that children experience a spurt in physical growth around age 6, accompanied by the appearance in the skin of oil-producing sebaceous glands similar to those associated with pimples in adolescence (Current Directions in Psychological Science, 12/96). They also reported that signs of sexual attraction, heterosexual or homosexual, can be observed in children as young as 9 or 10 years of age; interestingly, this is also about the time that girl-boy teasing begins. This attraction should not be confused with sexual desire, much less sexual activity; these kick in later, as the individual approaches and enters adolescence. Rather, at this early age there appears to be only a more or less clear "leaning" toward one sex or the other. McClintock and Herdt suggest that these changes are related to the secretion by the adrenal glands of a form of androgen known as dihydroepioandrosterone (DHEA), which begins at about age 6 and increases to a critical level at about age 10 -- a point which they call arenarche, by analogy to menarche. DHEA reaches adult levels at about age 18 before diminishing over the rest of the life course.

In any event, the hormonal effects of puberty interact with the social imprimatur as parents and others impose culturally bound standards for adolescent behavior. In some cultures, for example, sexual experimentation is permitted, even encouraged; in others, sex is strictly prohibited until marriage. In some cultures, boys and girls are permitted to date, and even engage in light sexual activity ("petting"); in other cultures, boys and girls can meet only under conditions of strict supervision; in still other cultures, marriages are arranged by the parents, and the engaged couple may have only minimal contact with each other before their wedding day. In all cultures, parents and others scrutinize adolescents for signs of "normality", and the adolescents' gender identities, gender roles, and erotic orientations are strengthened and challenged.

Middle and Old Age. Later in life, there are further dramatic changes in hormone levels. These are most obvious in women, particularly the sudden drop in estrogen levels, and cessation of menstruation, known as menopause. Recent evidence suggests that there may be a male version of menopause as well, known as partial androgen deficiency in aging men (PADAM), or andropause. Although men produce sperm throughout their adult lives, they experience a gradual decline in testosterone levels as they age (about 0.5% per year after age 30), which in turn may be associated with fatigue, loss of muscle tone and bone density, and "decreased libido" (a term derived from Freud's term for the sexual drive). Note, however, that the diagnostic criteria for andropause overlap greatly with those for depression. In the absence of laboratory tests showing abnormally low levels of testosterone, it is not at all clear that andropause is a legitimate diagnosis, or that HRT is a legitimate treatment.

Here again, the hormones interact with the social environment. For example, the cessation of menstruation, and the loss of child-bearing capacity, may challenge women's gender identities and roles. Similarly, age-related erectile difficulties may challenge those of men.

In summary, the "program" for gender dimorphism of identity and role begins with the sex chromosomes (XY or XX), which differentiate the gonadal tissue into testes or ovaries; and continues with hormones secreted by the testes, which differentiate the internal and external reproductive anatomy. This phyletic imprimatur, a process of gender dimorphism that is common to all humans, and is under direct biological control, endows the fetus with characteristically male or female reproductive anatomy.

At birth the program for gender dimorphism passes from the genes and the hormones to the social environment, as the parents classify the newborn child as male or female, and begin the process of raising him or her according to cultural concepts of masculinity and femininity. Early in childhood, the child recognizes his or her own gender, identifies him- or herself as male or female, and begins to model his or her attitudes, beliefs, and behavior on others of his or her "own kind". These social learning processes, which are under environmental control and vary from one culture to another, is known as the social imprimatur. Everyone undergoes gender dimorphism of identity and role, but the outcome differs from one individual to the next, and from one culture to the next, depending on the details of the social imprimatur.

The social imprimatur encompasses many different elements:

The point of all this is that gender dimorphism extends far beyond the obvious matters pertaining to reproductive anatomy. There are at least three other aspects of gender dimorphism that we have to consider.

Gender Identity

Gender role socialization is not simply imposed on the child from outside: the child is also an active agent of his or her own gender socialization, as he or she acquires an identity as a little boy or little girl.

Beginning at about age 2, children notice (as it were) their own genitalia, and identify themselves as boys or girls. (The cartoon on the left captures the situation beautifully: In the original, the caption reads "There is a difference!".  I first encountered this cartoon in the 1960s in a "QSL" card mailed by an amateur radio operator, whose identity I have since forgotten.  Apparently, it's now available as a tattoo.)

This self-identification has a number of consequences

Children's recognition of gender is perfected by about age 3. By this time, they strongly prefer objects labeled as "for" their own gender.

Differences in gender-role behavior are not reliably observed before age 2, but they are well established by the time the child goes to school:

Actually, children learn both gender roles, and adopt the role that is appropriate to their gender identity as male or female. Even so, there is some asymmetry in their preferences:

The active participation of the child in his or her own gender-role socialization illustrates the principle that the child is an active agent of his or her own development. Once the child has categorized him- or herself as male or female, he or she begins the active process of learning and performing the roles deemed appropriate for his or her gender. In this way, gender role socialization illustrates the complex interactions that play out between nature and nurture, and between the person and the environment.

Gender Role

In western society, gender roles have traditionally been divided into two major dimensions.

Link to an interview with Janet Taylor Spence.

In many ways, it would seem that masculinity and femininity are polar opposites -- that they anchor opposite ends of a single dimension of gender role. And, indeed, that's how these personality characteristics are measured in traditional questionnaire measures of personality, such as the Mf (Masculinity-Femininity) scale of the Minnesota Multiphasic Personality Inventory (MMPI) or the Fe (Femininity) scale of the California Psychological Inventory (CPI).

But, more recently, some feminist psychologists have argued convincingly for an alternative conception in which masculinity and femininity are construed as independent dimensions of personality (recall that a similar debate took place with respect to the relation between positive and negative emotionality). Such a scheme yields four basic categories of gender role:

Androgyny was a new scientific concept in the 1970s, but it was foreshadowed by one of the earliest documents of feminist theory: Woman in the Nineteenth Century (1845) by Margaret Fuller, the lone woman in the "Transcendentalist Club" that included Ralph Waldo Emerson and Henry David Thoreau. As Judith Thurman writes (reviewing a number of biographies of Fuller):

Margaret was a strapping girl who preferred boys' strenuous activities to girls' decorous ones.... [Her platonic but] amourous friendships informed Fuller's prescient notion of gender as a bell curve -- the idea that there are manly women, womanly men, and same-sex attractions, all of which would be considered perfectly natural in an enlightened society.... It was an "accursed lot", Fuller concluded, to be burdened with "a man's ambition" and "a woman's heart", though the ambition, she wrote elsewhere, was "absolutely needed to keep the heart from breaking". ("An Unfinished Woman", by Judith Thurman, New Yorker , 04/01/2013.)

The terms "tomboy" and its rough male equivalent, "sissy", have long been used to label gender-nonconforming girls and boys (though, in practice, "sissy" is generally more pejorative.  For a historical survey of "tomboyism" (with occasional forays into sissyhood), see Tomboy: The Surprising History and Future of Girls Who Dare to be Different (2020) by Lisa Selin Davis (reviewed by Lisa Damour in "'Tomboy' Looks at Gender Roles, and Role-Playing, Through the Ages", New York Times Book Review, 11/08/2020).

How Big Are Gender Differences, Really?

Men and women, and boys and girls, differ from each other psychologically in obvious and subtle ways.  In addition to obvious differences in masculinity (agency) and femininity (communality), it's also been claimed that there are gender differences in cognitive ability: males are generally held to be superior in mathematical and spatial ability, for example, and females to be superior in verbal ability.  All of these differences fit the cultural stereotypes (at least so far as Western culture is concerned), but how big are these differences, really.  Are they big enough to justify obvious differences in social outcomes, such as the fact that there are many more men than women teaching math and science at the college level? 

Oddly enough, for all the studies -- and there are umpteen thousands of them -- documenting gender differences in performance on this or that task, it was only relatively recently that anyone viewed this literature from the standpoint of effect size -- which, as you'll remember from the lectures on Methods and Statistics, measures the strength of an effect -- in this cases, differences between two groups classified by gender.  The first comprehensive review of gender differences was published by Eleanor Macoby and Nancy Jacklin in 1974, but these researchers did not have the statistical tools of meta-analysis available to them at the time, and they had to present only a verbal, impressionistic summary of the literature.  Since then, however, the technique of meta-analysis has been developed, allowing researchers to combine the results of a large number of studies, and summarize their findings in a single quantitative score representing effect size -- that is, taking all of the studies together, the magnitude of the difference between males and females on various measures.  These meta-analyses of the literature confirm that many of these gender differences exist -- but they also show how remarkably small most of them really are.

Among the first researchers to notice this, and to make a big deal out of it, was Janet Shibley Hyde, a professor at the University of Wisconsin (whose PhD was from UC Berkeley). 

OK, so there are differences, on average, in spatial, verbal, and mathematical abilities between males and females.  The economist Lawrence Summers, who was Secretary of the Treasury during the Clinton Administration and later President of Harvard University (and author of a paper famously entitled "There Are Idiots"), pointed to this last difference, especially, in explaining why there were so few women among the math and science faculty and graduate students at Harvard -- implying that women just didn't have the quantitative chops to succeed at the highest level in these fields (in this he was supported by Steven Pinker, a distinguished Harvard psychologist; in the resulting brouhaha, Summers was forced to resign from the presidency, though he remains a professor in the Economics Department).  But Summers left out two important points.

Hyde's papers set of an avalanche of meta-analyses of gender differences in various abilities and traits.  In 2005, Hyde summarized the results of these analyses with her gender similarities hypothesis, "that males and females are similar on most, but not all, psychological variables. That is, men and women, as well as boys and girls, are more alike than they are different" (p. 581).  Surveying the 46 published meta-analyses available to her at the time, she found a mean (unsigned) difference of d = .21 between males and females.  The figure at left depicts what a d of 0.21 looks like, in terms of the normal distribution.  Fully 78% of the psychological gender differences uncovered in the literature were, in terms of effect size, "very small" or "small" in magnitude; only 9% counted as "large" or "very large".  As Hyde pointed out, these differences hardly justify the title (and argument) of John Gray's 1992 best-selling book, Men Are From Mars, Women Are From Venus! (see also Hyde, 2007).  

And to put the icing on the cake, Zell and his colleagues (2015) reviewed a total of 106 meta-analyses, and confirmed Hyde's essential findings: an unweighted average d score of 0.21, right in the middle of the range of "small" effects; fully 85% of these effects classified as "very small" or "small", and fewer than 3% classified as "large" or "very large". 

Zell et al. conclude, "We utilized data from over 20,000 individual studies and over 12 million participants to reevaluate the gender similarities hypothesis and found that its core proposition receives strong support" (p. 18).

So, males and females are "far more similar than different" (Zell et al., p. 18) -- except on a few variables, the most salient being gender role: masculinity and femininity.  So now let's ask where these differences come from.

The debate over gender differences just won't go away.  In 2017, James Damore, a software engineer at Google, took a page from Lawrence Summers's book and argued that calls for increased representation of women in silicon Valley ignored scientific evidence of gender differences in cognitive abilities relevant to software engineering.  His manifesto quickly went viral.  Damore was subsequently fired on grounds of incompatibility with Google's culture, but the editors of The Economist argued (08/12/2017) that Larry Page, the co-founder of Google and the chief executive officer of Alphabet, Google's parent company, should have written a "ringing, detailed rebuttal" to Damore's manifesto instead.  Page didn't do that, but The Economist did it for him (08/19/2017), essentially echoing the points made here.  

Heredity and Environment in Gender Role

A number of twin studies have been conducted to determine the various sources of individual differences in gender role. Unfortunately, these studies have construed masculinity and femininity as polar opposites, not independent dimensions of gender role, but their results are still interesting.

In a pioneering study, Irving Gottesman et al. (1965, 1966), administered the masculinity-femininity scales of the MMPI and CPI to a sample of adolescent twins. Robert Dworkin and his colleagues repeated this testing with the same sample some 10 years later, when the subjects were adults. although these investigators did not calculate components of variance, the fact that the correlations for MZ twins were consistently higher than those for DZ twins gives prima facie evidence for a genetic contribution to individual differences in gender role. This isn't terribly surprising. But the relatively low magnitude of the MZ correlations suggests that the environment also plays a role in shaping this aspect of personality. In fact, if you apply the formulas discussed earlier, it's clear that the nonshared environment is by far the most powerful determinant of gender role. So much for the easy equation of biological sex (which, of course, is almost completely determined by the genes) and psychosocial gender.

A more recent study by Loehlin et al., with a larger, more representative sample of subjects, came to much the same conclusions (Loehlin et al. tested both Americans and Europeans, but for purposes of comparison I show only the American results here). MZ twins were more similar than DZ twins, in terms of masculinity-femininity, but the nonshared environment was a much more powerful

So, it seems that the biological and social imprimaturs interact to produce the individual's gender identity and gender role, but (setting aside the issue of the masculinization of the brain) the chief effects of the genes and hormones are anatomical and physiological, not psychological. They endow the developing fetus with reproductive anatomy that is more or less recognizably male or female, and that is just about it.

In these and other ways the environment constrains and supports the development of "appropriate" gender roles.

Evidence with respects to gender comes from classic studies of child-rearing practices (summarized by Maccoby and Jacklin,The Psychology of Sex Differences, 1974).

Link to an interview with Eleanor Maccoby.

In the wake of the women's liberation movement that began in the late 1960s, the social environment has become somewhat less rigidly structured according to gender. Many people would like to think that gender-role socialization has been loosened in these more "enlightened" times. But while traditional concepts of masculinity and femininity may have been loosened, they have not been abolished.

As an example, consider these toddlers' "training pants" bought in 2000 -- more than 30 years after the feminist revolution swept America. The girls' version, in pink, portrays Minnie Mouse singing songs with Daisy Duck; The boys' version, in blue, portrays Mickey Mouse driving a car and flying a plane with Donald Duck. These differences have nothing to do with the physical differences between the sexes (which might well make structural differences in training pants desirable), and nothing to do with behavioral differences either. They are a simple, and not too subtle, reminder of the social differences -- the differences in gender role -- between boys and girls.

A 2012 doctoral dissertation by Elizabeth Sweet, a sociology graduate student at UC Davis, surveyed advertisements for toys in the Sears catalog over the 20th century.  She summarized her findings as follows ("Guys and Dolls No More?", New York Times, 12/23/2012):

Gender has always played a role in the world of toys.  What's surprising is over the last generation, the gender segregation and stereotyping of tows have grown to unprecedented levels.  We've made great strides toward gender equity over the past 50 years, but the world of toys looks a lot more like 1952 than 2012.  Gender was remarkably absent from the two ads at the turn of the 20th century but played a much more prominent role in toy marketing during the pre- and post-World war II years.  However, by the early 1970s, the split between "boys' toys" and "girls' toys" seemed to be eroding....  I found that in 1975, very few toys were explicitly marketed according to gender, and nearly 70% showed no markings of gender whatsoever.  In the 1970s, toy ads often defied gender stereotypes by showing girls building and playing airplane captain, and boys cooking in the kitchen.  But by 1995, the gendered advertising of toys had crept back to mid-century levels, and it's even more extreme today.  In fact, finding a toy that is not marketed either explicitly or subtly (through the use of color, for example) by gender has become incredibly difficult...  For example, last year [2011] the ego Group, after two decades of marketing almost exclusively to boys, introduced the new "Friends" line for girls....  Critics pointed out that the girls' sets are more about beauty, domesticity and nurturing than building -- undermining the creative, constructive value that parents and children alike place in the toys.

A 2014 "Big Data" analysis of anonymous Google searches revealed persisting gender differences in parents' concerns and expectations ("Google, Tell me.  Is My Son a Genius?" by Seth Stevens-Davidowitz, New York Times, 01/19/2014).

Other modes of the person-by-situation interaction require more than the mere presence and appearance of the person. In these modes, the person must do something:

Erotic Orientation

The last aspect of gender dimorphism is erotic or sexual orientation -- whom one is attracted to as a sexual partner. Again, the simpleminded story is that genetic males identify themselves as boys, grow up to become masculine men, and are sexually attracted to women; and genetic females identify themselves as girls, grow up to become feminine women, and are sexually attracted to men. But everything we've discussed about about gender dimorphism so far has proved to be more complicated than that. Genetic males don't always grow up with the corresponding reproductive anatomy; they don't always identify themselves as little boys; and they don't always become stereotypically masculine. Genetic females are no different in these respects. And so, as we might expect, erotic orientation is also pretty complicated.

There was a time when homosexuality was classified as abnormal behavior, a sign of mental illness, and homosexuals were frequently sent to psychiatrists and psychologists in an attempt to "cure" them of their deviance. In 1970, following a vote of the members of the American Psychiatric Association, homosexuality was removed from the listing of "sexual disorders" in the 3rd edition of the Diagnostic and Statistical Manual for Mental Disorders (DSM-III). Individuals can still seek psychotherapy if they are bothered by their homosexual orientation (or by their heterosexual orientation, for that matter). But that is not the same thing as considering homosexuality per se to be a mental illness.

Transsexualism, or transgender, has undergone a similar evolution.  In DSM-II, published in 1968, transgender identity was listed as a "sexual deviation".  In DSM-III (1980), the edition which eliminated homosexuality as a category of mental illness, transsexualism was listed as a "psychosexual disorder".  In DSM-IV (1994) it was listed as a "sexual and gender identity disorder".  In DSM-5 (2013), the listing was changed to "gender dysphoria" and limited to individuals who distress or dysfunction with respect to their gender identity, male or female, trans or not.  In 2017 the World Health Organization proposed to remove transsexualism, or transgender identity from its International Classification of Diseases (ICD), the worldwide equivalent of the DSM.

Homosexuality was also commonly considered to be criminal behavior -- which is why so many homosexuals, especially homosexual men, stayed "in the closet" until relatively recently. As recently as 2003, 14 states still had "sodomy laws" on their books, proscribing such "unnatural" sexual behaviors as oral or anal sex -- although, frankly, these had rarely been enforced when practiced by heterosexual couples). These remaining laws were invalidated in a landmark Supreme Court ruling in Lawrence v. Texas (2003).

The origins of homosexuality are a persisting puzzle for psychologists and biologists. At first blush, homosexuality would appear to be maladaptive, because homosexuals do not procreate, the trait doesn't contribute to the survival of the species, and one would think it would have been erased from the human genome by now. For that reason, evolutionary psychologists twist themselves inside out trying to concoct "just so stories" to explain how homosexuality is adaptive after all.

Evolution of Homosexuality

Homosexuality poses problems for evolutionary psychology, because, like altruism, it seems maladaptive. If evolution favors traits that increase reproductive fitness, how could a trait evolve that doesn't lead to reproduction at all? Over the years, a number of hypotheses have been offered to explain how a genetic basis for homosexuality might have evolved:

These hypotheses are all very interesting, but they all appear to be predicated on the same adaptationist fallacy -- the notion that traits must be adaptive to evolve, and that traits evolve by virtue of their adaptive value.

The answer to the mystery of homosexuality may be simply that it is a mystery. If you think about it, setting reproductive issues aside, the sexual attraction that two people of the same sex feel for each other may be no different, no more mysterious, than the sexual attraction that two people of opposite sex feel for each other.

As Hanne Blank puts in in her book,Straight: The Surprisingly Short History of Heterosexuality (2012):

We don't know much about heterosexuality. No one knows whether heterosexuality is the result of nature or nurture, caused by inaccessible subconscious developments, or just what happens when impressionable young people come under the influence of older heterosexuals".

Development of Homosexuality

In many cases, the precursors of homosexuality can be seen long before adulthood, or even adolescence, in childhood behavior patterns (Bailey & Zucker, 1995).  As a group, both male and female homosexuals tend to show cross-sex-typed role behaviors fairly early in childhood.  These children have sex-appropriate gender identities, in that chromosomal boys identify themselves as boys and chromosomal girls identify themselves as girls.  It's just that they tend to display characteristics associated with the other gender role -- the boys more "feminine", the girls more "masculine" in sports, play, toy choice, and "pretend" play.  However, not too much should be made of these signs of "prehomosexuality".  Children also engage in a fair amount of role-experimentation, as they begin the process of figuring out who they are. 

Let's look at our tried-and-true way of examining the origins of some trait, which is the twin study.  Michael Bailey and his colleagues reviewed 12 large twin studies, looking at the concordance rate for homosexuality.  The figures differ a little depending on whether the study specifically recruited homosexual subjects, which injects some bias into the sample, or rather was based on more representative samples of the population.  In either case, the MZ concordance rate is higher than the DZ concordance rate, which again provides prima facie evidence for a genetic contribution to homosexuality.  When the researchers calculated an overall estimate, correcting for sampling bias, it's very clear that when it comes to erotic orientation, as with almost everything else we know about personality and attitudes, genes are important but not decisive: the environment, and particularly the nonshared environment, makes a big difference.

More recently, Andrea Ganna and her colleagues documented genetic influences on homosexuality -- or, at least, same-sex sexual behavior -- using the GWAS methods described earlier (Science, 08/30/2019, from which the graphics are taken).  Based on two large samples from the UK (the Biobank database of 408,995 individuals) and the US (68,527 individuals who subscribed to the "23and Me" service), as well as three smaller samples, they looked for specific gene loci (technically, technically, single-nucleotide polymorphisms, or SNPs; see more below) associated with subjects reporting that they had ever had sex with a partner of the same sex.  But the investigators also had information about the proportion of same-sex to total sexual partners, sexual attraction, and sexual identity.  After all this, they found 5 such SNPs two for men (on chromosomes 11 and 15), one for women (on chromosome 4), and two for both men and women (on chromosomes 7 and 12).  Of course, having sex with a partner of the same sex isn't the same thing as being homosexual -- there's a fair amount of same-sex sexual experimentation that never gets beyond that.  Interestingly, none of these loci were on the X chromosome -- a region known as Xq28 famously identified as the "gay gene" by Dean Hamer, a researcher at the National Institutes of Health, in 1993 (one wonders, then, how these 5 SNPs will fare in the next study). That's five -- 5 -- genes out of 20,000 in the entire human genome.  And taken together, these genes account for only 1% of population variance in same-sex behavior.  As the authors put it, "same-sex sexual behavior, like most complex human traits, is influenced by the small, additive effects of very many genetic variants, most of which cannot be detected at the current sample size". 

The researchers also examined the "genetic correlations" between same-sex sexual behavior and other personal characteristics -- that is, the amount of variance that two characteristics share due to genetic influences.  Among the strongest associations were marijuana use, openness to experience (one of The Big Five personality traits) and the number of sexual partners.  Interestingly, the genetic correlation between same-sex sexual behavior and the ratio of the lengths of the subjects' 2nd and 4th fingers -- the "2D:4D digit ratio", famously reported to be a physical trait correlated with androgen levels in men (leading perhaps millions of men to get out their rulers and measure themselves -- again).

It is possible that hundreds or thousands of genes make a contribution to same-sex behavior, but only these five -- 5! -- turned up significant in a GWAS involving half a million people.  And again, they account for only 1% of the variance.  That's a long way from "the gay gene".  And when the investigators took account of all the genome-wide correlations, even the ones that didn't reach statistical significance, they obtained a "SNP-based" heritability coefficient ranging between 8 and 25%.  That is, they estimated that all genetic influences, when aggregated, accounted for as little as 8%, or as much as 25%, of population variance in same-sex sexual behavior, broadly construed to include fantasies, attraction, and identity as well as actual behavior.  That's about the same as for other personality characteristics, as estimated by more conventional twin studies (and without all the high-tech hoopla).  And, it leaves the vast bulk of population variance to be explained by environmental influences, both shared and nonshared.  Some of the nonshared environment may, in turn, be influenced by genetic tendencies.  For example, there is a genetic contribution to the tendency to have multiple sexual partners -- and the more sexual partners one has, arguably, the more likely at least one of them will be of the same sex.  And there is a genetic contribution to openness to experience, and one of those experiences might be same-sex sex.

The influence of genetics has been used by some advocates to argue that homosexuality is not a choice -- it's given by nature, just like our other physical characteristics; and therefore, homosexuality should not be criminalized, and homosexuals should not be discriminated against (for example, they should have the same rights to marry as heterosexual individuals).  And that's a point.  But does the role of the (nonshared) environment indicate that there's a role for personal choice as well -- that, in the final analysis, homosexuals choose to be that way?  Not necessarily.

At the same time, it's pretty clear that nature and nurture interact to create an individual's erotic orientation. How could this happen? We have no idea, but think about how, as we've shown, nature and nurture interact to produce other aspects of gender dimorphism -- biological sex, gender identity, gender role -- and erotic orientation as well.

Here are two very interesting theories that suggest how nature and nurture could interact to determine whether one is heterosexual or homosexual. Even if these theories are wrong in detail, they are important because they suggest a way of thinking about the origins of homosexuality that escapes the rigid confines of thinking of it as a biologically determined trait on the one hand, or as a personal choice on the other.

Michael Storms (1981) based his theory on the phenomenon of imprinting, and the notion of a critical period, discussed in the lectures on Learning. Storms began with the assumption, for which there is some supporting data, that gay men tend to reach puberty earlier than heterosexual men (this claim is controversial, but it doesn't matter for the purposes of this example). That is to say, some boys reach sexual maturity, start getting sexually attracted to anything, at a time in their lives when they're hanging mostly with other boys (because, at that age, girls are still yucky creatures to be avoided). Just as a gosling follows the first thing that moves after it hatches, so (the theory goes) a boy who enters puberty when there are mostly other boys in his environment will become erotically oriented to other males. A similar explanation would hold for lesbian women.

More recently, Daryl Bem (1996, 1998) proposed a theory of erotic orientation that is similar to Storms' in form, but differing in detail. Bem began with the assumption that we are sexually attracted to people who are very different from ourselves. Yes, similarity breeds liking, but -- to quote Bem's phrase -- "the exotic becomes erotic". So, a boy with stereotypically masculine personality characteristics will, when he reaches puberty, begin to be attracted to someone with stereotypically feminine personality characteristics -- which is, in all likelihood, a female. But a boy with stereotypically feminine personality characteristics will, when he reaches puberty, be attracted to someone with stereotypically masculine personality characteristics -- which will, in all likelihood, be a male. the same kind of process unwinds for girls. You can work out other possibilities for yourself.

Note that both Storms' and Bem's theories are compatible with a genetic contribution to homosexuality -- but, critically, neither one assumes that there is anything like a "gene for homosexuality"; and, thus, neither gets caught up in the Darwinian paradox of same-sex attraction.

And also note that both theories are compatible with a large contribution of the nonshared environment.

It should be noted that, in part, the development of homosexuality has a cultural component to it.  It's been argued that gay people, much like deaf people -- for that matter, much like Catholics and Jews and New Yorkers, inhabit a distinctive culture that is different from the one inhabited by straights (Protestants, Muslims, San Franciscans).  Part of identifying yourself as a member of a group involves learning and partaking of that culture through the process of social learning.  This would be true even if homosexuality were completely determined by genetic endowment.  This is not to say that vulnerable children are recruited into the "gay lifestyle" by scoutmasters and gym teachers.  Wherever homosexuality, or any other part of one's identity and personality comes from, you have to learn how to be who you are.  It's also to say, paraphrasing David M. Halperin, an English professor at U. Michigan who has taught a course entitled "How to be Gay: Male Homosexuality and Initiation", which became infamous among conservative pundits, that "gay [people] acquire a conscious identity, a common culture, a particular outlook on the world, a distinctive sensibility....Queer [people] are different, and we should hold on to our culture" ("How to Be Gay", Chronicle of Higher Education, 09/07/2012; see also his 2012 book, How to Be Gay).

It should be noted that this general framework -- nature and nurture interacting, through genes and the nonshared environment -- is not specific to aspects of gender dimorphism. Other aspects of personality are probably acquired in much the same way. In other words, the development of gender differences in identity and role (not to mention erotic orientation) serve as models for the development of personality in general.

For a comprehensive overview of the development of homosexuality, and the implications of this research for public policy, see "Sexual Orientation, Controversy, and Science" by J. Michael Bailey et al., published in Psychological Science in the Public Interest (2016).

Gender Polymorphism

All of this should have made clear that, when it comes to issues of sex and gender, the story is not merely one of straightforward genetic determinism. Contrary to what Freud said, anatomy isn't destiny after all.

So, if you do the math, that's 5 x 4 x 4 x 4 = 320 different gender-related categories. And that's a minimum estimate -- especially when you consider that these aspects of gender dimorphism may not be organized as discrete categories at all, but rather as continuous dimensions, so that there is an infinite number of locations in the four-dimensional "gender space". Never mind that some children and adults are gender fluid, moving back and forth between male and female gender identity and masculine and feminine gender role.

And, to make things even more interesting, very recent research adds a fifth set of gender-related categories, what you might call erotic or sexual identity, analogous to gender identity, but having to do specifically with sexual attraction and behavior. So, just as you can have a person who is biologically male but has a feminine gender identity, or someone with a male gender identity who adopts a feminine gender role, so there appear to be individuals who are, say, homosexual in erotic orientation but who identify themselves as heterosexual (Haldeman, 2003, 2004;. That is, they acknowledge their homosexual leanings, but for some reason wish to identify themselves as heterosexuals -- and sometimes seek treatment to help them conduct themselves accordingly. This is not the same as forcing a homosexual to undergo "conversion therapy". Rather, this appears to be a matter of how individuals choose to identify themselves -- often for religious reasons, but sometimes for nonreligious personal reasons.

But it's complicated in very interesting ways. And these complications have implications for how we think about development, and especially personality development, in general.  For example, Hyde et al. (Am. Psych., 2018) have argued that emphasis on the "gender binary" -- the idea that there are only two types of people, male and female, has thoroughly distorted scientific research on all aspects of gender.  They identify five sets of research outcomes that challenge the gender binary, and should lead us to think differently about gender in the future (quoting from their abstract):

1. neuroscience findings that refute sexual dimorphism of the human brain;
   
2. behavioral neuroendocrinology findings that challenge the notion of genetically fixed, non-overlapping, sexually dimorphic hormonal systems;
   
3. psychological findings that highlight the similarities between men and women;
   
4. psychological research on transgender and nonbinary individuals’ identities and experiences; and
   
5. developmental research suggesting that the tendency to view gender/sex as a meaningful, binary category is culturally determined and malleable.

The Latest on Sex and Gender For more on biological, psychological, and sociocultural aspects of sex and gender, see the Special Issue on Sex and Gender published by Scientific American, 09/2017.  As the editors say on the cover, "It's Not a Women's Issue: Everybody has a stake in the new science of sex and gender".  For example: "Promiscuous Men, Chaste Women, and Other Gender Myths", by Cordelia Fine and Mark A. Elgar, takes issue with the classic position of evolutionary psychology that behavioral differences between males and females are hard-wired into the nervous (and endocrine) system by natural selection.  On the contrary, they show that environmental and experiential factors play a major role in many of these differences, that many of these differences, besides being small, are not immutable.  In their view, :progressive cultural shifts" "rewrite" nature. "Is There a 'Female' Brain?", by Lydia Denworth, answers its own question with a fairly vigorous "No", and that "most brains are a mosaic of male and female characteristics".  As with behavioral and psychological sex differences, their is considerable overlap between males and females in the distribution of various features of the brain (e.g., the volume of the left hippocampus). "When Sex and Gender Collide", by Kristina R. Olsen, summarizes the results of the TransYouth Project, a study which has followed more than 300 transgender and gender-nonconforming children for 20 years.  A major finding of the study is that a fairly firm "trans" identity develops fairly early in these children. "Beyond XX and XY", by Amanda Montanez, consists of a fabulous chart, much expanded from Money and Ehrhardt's charting of the phyletic and social imprimatur, showing all the different biological factors that can affect biological sex (click on the image below for a full-size reproduction).  "The Brilliance Trap", by Andrei Cimpian and Sara-Jane Leslie", takes on the question of whether the (small) sex differences in math ability justify the under-representation of women in STEM fields.  It doesn't -- not least because a whole host of psychosocial factors, including gender stereotyping (and the sex discrimination it leads to) and stereotype threat (and the self-handicapping it leads to) are probably more important.  They also make the interesting  argument that stereotypes about scientific or artistic "brilliance", coupled with the myth of gender difference (and the stereotype threat that comes with it) discourage girls and women from even entering some fields -- even though most of those who work successfully in these fields are not "brilliant" -- only very smart.  The bottom line is that there are plenty of women soldiers, pilots, and engineers to go around, if only the psychosocial context were more favorable. In their introduction, the Editors note that "Sex is supposed to be simple -- at least at the molecular level.  The biological explanations that appear in textbooks amount to X+X=[female\ and X+Y=[male], Venus or Mars, pink or blue.  As science looks more closely, however, it becomes increasingly clear that a pair of chromosomes do not always suffice to distinguish girl/boy -- either from the standpoint of sex (biological traits) or gender (social identity)." They're right.  The biology of sex and gender, taken alone, is complication enough.  Add the psychology, not to mention the anthropology, sociology, and all the other social sciences, and you've got plenty more.

Maturation: Development as Quantitative Change

The earliest theories of psychological development focused on maturation and learning. In general, these theories offered a view of the child as a short, stupid adult who grows smarter as he or she grows bigger. Viewed in this way, there is a continuum between childhood and adulthood, with no abrupt, qualitative changes.

Maturation may be defined as the progressive, inevitable unfolding of certain patterns of behavior under genetic control. These behavior patterns occur in a regular sequence, unaffected by practice or environmental change.

Maturation is a good description of certain developmental processes, such as walking. We speak of children "learning" to walk, but we know from the stepping reflex in infants (see the lecture supplement on Learning) that walking occurs naturally, requiring only that the child be able to support itself. Thus, walking occurs as soon as the skeletal musculature develops sufficiently to provide that support.

A classic study of maturation involved traditional Hopi and Navajo children, who are swaddled and bound to a cradle for the first year of life. This severely restricts motor behavior, but once released from the cradle there is little retardation in the emergence of walking.

In another classic study, by Arnold Gesell, some children were trained in walking and climbing stairs. They did, in fact, show these behaviors earlier than untrained children. But the untrained controls quickly caught up, and the further progress of the experimental group was not accelerated by their training. Both groups advanced beyond walking at the same pace.

The continuous view of development is exemplified by the measurement of intelligence in terms of IQ:

The implication is either method of measuring intelligence is that children continuously grow smarter as they grow older.

Despite debates over whether IQ is heritable, the classical "continuity" view is that the child gradually acquires knowledge through learning, where learning was construed as tantamount to classical or instrumental conditioning. John Locke, an English philosopher of the 18th century, famously argued that the infant is a tabula rasa, or "blank slate", which is "written on" by experience. In the Lockean view, development is a matter of learning more than you already know.

Whether the theoretical focus is on maturation or learning, the process of development is viewed as a matter of continuous, quantitative change: the infant starts out small, physically and mentally, and gets bigger, physically and mentally, as he or she grows up.

Development as Qualitative Change

Later theories of development focused on qualitative changes. The child is not just a short, stupid adult, but rather the young child is held to think differently than older children and adults. Thus, developmental differences are qualitative, differences in kind, not just quantitative, differences in amount. Children are not stupid, compared to adults, but their intelligence needs to be appreciated on its own terms. D

Despite debates over the heritability of IQ, the classical continuity view is that the child gradually acquires knowledge through learning, where learning was construed as tantamount to the sorts of things we studied in the lectures on classical and instrumental conditioning.  John Locke, an English philosopher of the 18th century, famously argued that the infant is a “tabula rasa”, or blank slate, which is written on by experience.  In the Lockean view, development is a matter of learning more than you already know.  This contrasted with the earlier view of Descartes, who assumed that some knowledge was innate, as a gift from God.  Whether the theoretical focus is on biological maturation or on learning, in either case the process of development is viewed as a matter of continuous, quantitative change.  The infant starts out small, physically and mentally, and gets bigger, physically and mentally, as he or she grows up.

Piaget's Stage Theory of Cognitive Development

This was the view propounded by Jean Piaget, who argued that the development of intelligence proceeds through a sequence of stages, each defined by cognitive landmarks:

An important concept in Piaget's theory is the schema, a term which should be familiar from Bartlett's work on memory reconstruction and Neisser's concept of the perceptual cycle. For Piaget, a schema is an organized mental structure that produces a particular coordinated response to a certain class (or category) of stimulation. Thus, a schema is a kind of concept, that renders diverse objects functionally equivalent -- because they belong in the same class.

In a very real sense, cognitive development is the development of schemata (the plural of schema, although schemas is also an acceptable alternative) -- their elaboration and differentiation. Schemata are like categories, and guide the perception of (and thus response to) particular objects and events. The interaction between schemata and the objects with which they come into contact is characterized by the twin processes of assimilation and accommodation. By virtue of assimilation, the mental representation of the stimulus is altered so that it will fit into the schemata employed to process it; by virtue of accommodation, the schema itself is altered so that it can receive the stimulus. Thus, the final representation of the stimulus is a sort of compromise between what was expected and the actual stimulus itself. The neonate confronts the world with a primitive set of innate schemata; through assimilation and accommodation, the nature of these schemata gradually change. At certain points, however, the changes in mental schemata are so dramatic that they appear qualitative rather than merely quantitative; these shifts mark the child's move from one stage of cognitive development to another.

The first of Piaget's stages is sensory-motor intelligence (also known as the sensory-motor period). In this stage, which encompasses the first two years of life, the world of the child is one of unrelated sensory experiences and reflex-like motor reactions to them. The child has no ability to connect the present with the past or the future, and no ability to distinguish between self and other. At least, that is how the child starts out. To take a phrase from William James, for the sensory-motor infant the world is a "blooming, buzzing confusion". Over the first two years of life, these capacities develop; according to Piaget, their complete acquisition terminates this stage of cognitive development.

One of the major accomplishments of the sensory-motor period is the development of object permanence. The newborn's behavior is tied to what comes through its senses, which are processed by sensory-motor schemata. Out of sight is out of mind. Eventually, however, the child comes to behave as if they have internal representations of objects that are not actually present in their sensory environment. Early on, if a toy is hidden they will turn their attention to something else. Later, if a toy is hidden they will search for it. This searching behavior shows that the child has an idea of the object that persists despite its physical disappearance -- at this point, the child has acquired the capacity for forming internal, mental representations --memories -- of the outside world.

At this point, about 24 months after birth, the infant moves fully into the next stage of cognitive development, the preoperational period. At this point, the child is able to form and retain internal representations of objects and events, but these representations exist as individual mental units, unrelated to each other. The major achievement of the preoperational period is the ability to relate one representation to another, through higher-order schemata called operations. This takes about the next five years.

The development of operational modes of thought is marked by the emergence of conservation, which occurs by about the time the child is seven years of age. In the earliest portion of the preoperational period, the child does not conserve at all. If a short, wide cup of liquid is poured into a tall, thin glass, he or she may well say that there is more liquid in the former than in the latter. This is because the child can track changes in either height or diameter (actually, if you want to be technical, radius), but not both simultaneously. Thus, the child cannot understand that volume, which is a product of height and radius, remains constant when the liquid is poured from one container into the other. Similarly, if a young child is shown five objects lined up over a short distance, and later the same five objects arrayed over a longer line, he or she is likely to say that there are more objects in the latter case than in the former. Thus, the child confuses the number of the objects with the linear distance over which they are arrayed. At some point, however, the child no longer makes these mistakes: he or she has acquired the ability to consider height and radius, number and distance, simultaneously, and compensate for one with the other. At that point, the child has acquired the ability of conservation.

An analogue of conservation failure in the interpersonal domain is egocentrism, which is not to be confused with selfishness. From Piaget's point of view, egocentrism reflects the child's inability to take another's point of view, or to appreciate the viewpoints of other peoples. Thus, an egocentric child will say that other people view the world from the same perspective as he or she does; and that others will react to events as he or she does. By the age of seven, however, the child has abandoned the egocentric attitude, and is able to represent the world as others see it, as well as he or she sees it.

At about age seven, the child enters the stage of concrete operations, in which children are capable of thinking and reasoning about objects and events which they have actually experienced. Concrete-operational children are actually pretty powerful thinkers. They conserve, they can pay attention to objects other than the most salient ones, they can take another's point of view, they can take account of transformations in state, they can classify objects into groups based on shared properties, and they can generate and use hierarchical classification schemes. You can get along pretty well in life with nothing more than concrete operations, so long as you are reasoning about familiar problems involving familiar objects and events.

Unfortunately, concrete operations aren't always enough. It is often useful to go beyond our own past experiences, and to reason about things we haven't seen or touched, and about things that aren't visible or touchable. Concrete operations don't suffice for this purpose. Good thing, then, that around age twelve the child enters into the last of Piaget's stages,formal operations. In formal operations, the thinking can be purely symbolic, without referring to anything at all by way of concrete objects and events. It is necessary, for example, for the transition from arithmetic to algebra -- where x, for example, is just an abstract symbol and doesn't refer to anything at all.

The hallmark of formal operations is scientific thinking, which is what lies behind Piaget's notion of the child as a naive scientist. This is marked by four different qualities: (1) hypothetico-deductive reasoning, in which we can hypothesize about a certain state of affairs, and then reason deductively from that hypothesis (that is, we can assume that something is true without requiring that it actually be true); (2) inductive reasoning, in which the person generalizes from specific observations to general principles; (3) reflective abstraction, in which the child is able to reflect on their own thoughts to arrive a novel ideas; and (4) propositional logic, in which the child can reason about two or more abstract entities represented by statements such as "If there is a P, then there is a Q". Children who have developed the capacity for formal operations are able to deal with several abstract variables at the same time.

If concrete operations are analogous to arithmetic, formal operations are analogous to algebra.

Piaget traced cognitive development only up to adolescence. One aspect of the debate about his theory is whether cognitive development does in fact end with the acquisition of formal operations, or whether there are other, more advanced, stages of thought.

Another question is whether Piaget's stages are stages at all -- that is, whether it is true that children progress from sensory-motor intelligence through preoperational logic and concrete operations to formal operations in the way that he thought. An important part of this question concerns the lower boundaries of Piaget's stages. Is it really true that children younger than age 7 are generally incapable of abstract thought? A very large research tradition has developed out of questions like these, the general conclusion of which is that cognitive development is probably more continuous than Piaget thought, and that even very young children have amazing powers of thought, at least in limited domains. For example, it has been shown that five-month-old infants are capable of rudimentary arithmetic operations of addition and subtraction, so long as they are only asked to deal with very small number of objects. 

Of course, Piaget was not the only theorist to offer a conception of development in terms of stages.  Long before Piaget, Sigmund Freud had offered a stage theory of psychosexual development running essentially from birth to adolescence.  And in the 1960s, Erik Erikson, a follower of Freud's, offered a stage theory of psychosocial development that encompassed the entire life cycle from birth to death.  Piaget, Freud before him, and Erikson afterward had an enormous influence on thinking in developmental psychology.  Their idea of development as a progression through a succession of qualitatively different stages was received quite enthusiastically.  And you can see the legacy of these theorists in the proliferation of stage theories of everything in developmental psychology.

For a comprehensive account of 

Kohlberg's Theory of Moral Development

Following in Piaget's footsteps, Lawrence Kohlberg developed a theory of the stages of moral development -- by which he meant moral reasoning, not moral behavior. For Kohlberg, moral development followed a strict trajectory of 3 stages, each consisting of two sub-stages, through which the child proceeds from a heteronymous (other) to autonomous (self) orientation..

Kohlberg assessed an subjects' stage of moral reasoning by coding their responses to the Heinz dilemma -- a story about a man, named Heinz, who has a sick wife:

Like Piaget, Kohlberg believed that his stages were universal, but in fact his studies were largely confined to male subjects -- leading him to conclude, for example, that women generally failed to reach the highest stages of moral development (roughly equivalent to the views of a liberal Democrat).Later, Carol Gilligan -- who was first Kohlberg's student, and later his colleague at Harvard's Graduate School of Education -- argued that Kohlberg's procedures were flawed, and that women went through stages of moral development that were qualitatively different from those of men. In this way, Gilligan laid the foundation for "difference feminism", based on the "essentialist" view that women's mental lives follow different principles than those of men.

In any event, Gilligan revised Kohlberg's scheme, based on the idea that women are primarily motivated by a concern for others, rather than independence, and an ethic of care rather than an ethic of justice. Unlike Kohlberg (and Piaget), Gilligan did not assign particular age ranges to the stages, but she retained the assumption that girls and women moved through these stages in an invariant sequence. Whereas Piaget and Kohlberg assumed that the transition from one stage to the other is based on changes in cognitive capacity, Gilligan argues that the transition is mediated by changes in the person's sense of self.

Freud's Theory of Psychosexual Development

In the domain of personality, another theory postulating qualitatively different developmental stages is Freud's theory of psychosexual development. The fundamental assertion of Freud's psychoanalytic theory is that personality is rooted in conflict between certain biological instincts (sex, aggression) and environmental constraints and demands. This conflict must be resolved in some way. The child's adaptation to conflict interacts with other developmental events, and personality is formed from the resulting habitual adaptation.

For Freud, sexual and aggressive motives are at the center of personality. These instincts arise from the id and are controlled by the ego and superego. They are the urges that the defense mechanisms defend us against. Now, nobody would argue that sexual issues are not important for personality. Many of the issues that confront adolescents and adults are sexual in nature. Once you reach puberty, if not before, sex is an issue. But Freud went further, by stating that sex is the paramount issue, from birth. He believed that sexual impulses were present in the newborn child, and that they continued to seek expression and gratification until death. The theory of infantile sexuality was Freud's most radical hypothesis. However, the student should understand that for Freud, sexuality was not confined to intercourse and orgasm. Rather, Freud defined sex as anything that lead to pleasure. Thus, the theory of infantile sexuality is a portrait of the infant as an active seeker of pleasure. The precise form that this pleasure takes is determined by the child's stage of development.

For Freud, all instincts have their origins in some somatic irritation -- metaphorically, some itch that must be scratched. At any time, a particular portion of the body is the focus of that instinct -- the place where arousal and gratification occur. These somatic loci change systematically through childhood, and stabilize in adolescence. These systematic changes comprise the stages in psychosexual development, and the child's progress through these stages is decisive for the development of personality.

Freud believed that the individual's passage through these stages left its imprint on adult personality. If all goes well, the person develops a genital character, as reflected in full sexual satisfaction through orgasm. The genital character is able to effectively regulate his (or her) sexual impulses for the first time. The person need no longer adopt primitive defenses such as repression, though certain adaptive defenses are still operative. The person's emotions are no longer threatening, and can be expressed. The person is no longer ambivalent, and is able to love.

However, all doesn't usually go well -- else there wouldn't be a need for psychoanalysts! Freud believed that people rarely, if ever, passed through the psychosexual stages without incident, and people rarely develop the genital character spontaneously. Usually, the person experiences some sort of developmental crisis at an earlier stage -- a crisis that prevents growth, fulfillment, and the final achievement of genital sexuality.

By virtue of fixation and/or regression, the person -- that means you, and me, and everyone -- develops a particular neurotic character, depending on the developmental stage at which the person has fixated, or to which he or she has regressed.

As you can see, this is the kind of theory that only a particular kind of man could conjure up. In discussing Freud's psychoanalysis, the student is warned that there is hardly a shred of clinical or experimental evidence in support of the theory. And a lot of it will strike a contemporary reader as quaint, if not downright silly. Nevertheless, psychoanalysis has had such an enormous impact on our culture -- literature, cinema, and art -- that it would be criminal to ignore it entirely. So you get some introduction to psychoanalysis in this course, but this is it -- a couple of crummy paragraphs buried in a lecture supplement on development.

Erickson's Theory of Psychosocial Development

Traditionally, mental development was thought to stop with adolescence -- or, at latest, the entry to adulthood.

Even so, there has always been some sense that development was not complete at adolescence -- that change and growth were still possible in young adulthood, middle age, and old age.

This concern with development throughout the lifespan, from birth to death, is expressed in the psychosocial theory of Erik Erikson, a disciple of Freud's. Erikson argued that personality was the product of the social environment as well as of biology. He de-emphasized the instincts, and especially, infantile sexuality, and focused instead on the social conditions of child development and adult life. Mostly, Erikson focused on the issue of ego identity -- one's awareness of oneself, and one's meaning for other people. He also expanded the notion of development by arguing that there is, indeed, life after puberty. Not only did he propose stages of growth beyond the genital, but he also introduced a social reinterpretation of the classic Freudian stages -- hence the label "psychosocial", rather than "psychosexual".

In the end, Erikson gave us an epigenetic conception of development similar to Freud's. That is, the individual progresses through an inevitable sequence of stages; and, at each, meets and resolves some crisis. Each stage builds on the one(s) that went before. Each stage has several elements, including a crisis that must be met and a strength that develops during the crisis. The resulting stage theory is sometimes known as the "Eight Ages of Man".

In the previous stages, Erikson reinterpreted Freud (as some of their names imply). In the following stages, emerging after adolescence, Erikson added to the basic Freudian conception.

As he (actually, his wife and collaborator, Joan Erikson) entered his (her) 9th decade, Erikson (in The Life Cycle Completed, 1998) postulated a ninth stage, in which the developments of the previous eight stages come together at the end of life:

Observational studies have provided some evidence for this fourth stage, but Erikson's original "eight-stage" view is the classic theory of personality and social development across the life cycle. 

It should be noted that Erikson's theory, like Freud's, is highly impressionistic, and not necessarily based on a proper scientific analysis of lifespan development. Also, like Freud's, it is strongly based on a particular cultural experience, and a particular view of what is important in life. Erikson's theory is not presented as established scientific fact, but rather as a good example of how a stage concept of development can be applied throughout the life span. There are some pearls of wisdom here, but take the whole thing with a grain of salt. The theory has been extremely influential in popular culture, and it has fostered an entire new discipline of life-span developmental psychology.

Stage Theories of Everything

Piaget, Erikson, and Kohlberg, not to mention Freud, popularized stages (rather than continuities) as a framework for understanding psychological development, and pretty soon stage theories began to appear in lots of different domains, including of everything.

James Fowler proposed a stage theory of the development of religious faith.

Elisabeth Kubler-Ross proposed a stage theory of death and dying.

And, famously, "Bill W." and "Dr. Bob", the founders of Alcoholics Anonymous, proposed the "12 Steps" that people must take, in order, in order to recovery from alcoholism -- a "self-help" approach that has since been generalized to other forms of substance abuse.

As intuitively appealing as these stage theories may be, they share the problems which beset Piaget's theory (and, for that matter, Freud's and Erikson's and Kohlberg's).  So where does the intuitive appeal come from?  Mostly, I think, stage theories appeal to us because we like stories, and stage theories provide a kind of narrative structure that organizes what goes on in development.  Development proceeds from starting point (like Piaget's sensory-motor period) and proceeds to some endpoint (like formal operations).  The story has a beginning, a middle, and an end, with plot-points along the way like the acquisition of object permanence and the loss of egocentrism. 

The only problem is that stage theories aren't right.  Development just doesn't proceed in the lockstep stages envisioned by Piaget and Freud.

Still, Piaget's theory of cognitive development has been enormously influential.  Most current theoretical approaches to psychological development have their origins in extensions of, or reactions, to, Piaget's theory and research.  For a comprehensive account of Piaget's theory of cognitive development, there's nothing better than The Developmental Psychology of Jean Piaget by John Flavell (1963). For an overview of the development (pardon the pun) of developmental theory since Piaget, see "The Evolution of Developmental Theories Since Piaget" by Philippe Rochat, (Perspectives in Psychological Science, 2024).

Cognitive Development After Piaget

In all stage theories, the stages of development are universal,obligatory,stereotyped, and irreversible.

These are essentially hypotheses about the nature of development, and when they are tested, research has usually failed to confirm them. In particular, research testing Piaget's theory revealed a number of anomalies, which led investigators to refocus their theories on continuities in mental development.

The first of these, one noted by Piaget himself, was the phenomenon of decalage (pardon my French). The implication of Piaget's stage theory of development is that the child makes a wholesale, quantum leap from one stage to the next. But Piaget and others recognized that this transition wasn't as abrupt as initially believed. A child moving from the pre-operational period to concrete operations, for example, would conserve on some tasks but not others.

The second, and by far more critical, had to do with the lower boundaries of the various stages. Again, the implication of Piaget's theory is that there are certain kinds of tasks that a child of a certain age just cannot do. And so developmental psychologists began to ask whether, in fact, 3- and 4-year-olds might conserve -- if only they were tested in the right way.

Among the first to ask this question was Rochel Gelman, who examined the child's conception of number. Piaget had argued that pre-operational children just didn't have any concept of number, or quantity, they couldn't count, and they certainly couldn't do arithmetic. But Gelman, working with Randy Gallistel (a mathematical psychologist who was also her husband), analyzed the concept of number into three principles:

So, suppose you show a 4-year-old a set of 3 objects, and ask her how many there are, and she says "five". On first blush, it looks like she doesn't know how to count. But Gelman went further, and asked the child to count the objects out loud. A child might say "One, four, five -- five things!"; asked again, she repeats her count, "One, four, five -- five things". What this indicates is that the child understands one-to-one correspondence, has a stable order, and understands cardinality. She just doesn't assign the conventional words to various quantities. But she's clear got a concept of number, and she clearly has a grasp of counting principles. Gelman's study showed that "pre-operational" children typically displayed an understanding of these counting principles, even if they didn't count like the adults did.

Karen Wynn pushed the envelope even further, by asking if infants also had the ability to count. This is clearly impossible, according to Piaget, because numbers represent quantities, and children in the sensory-motor period don't have representations. Wynn's research was based on the idea that looking time -- the amount of time that children (and adults, for that matter) spend looking at something -- is an index of surprise and attention. she showed 4-5-month-old infants displays of the following sort:

Wynn found that infants were surprised -- they looked longer at the display -- when it gave the "wrong" answer -- when 1+1=1 rather than 2, or (especially) when 2-1=2 rather than 1. While the infants probably couldn't do differential calculus, they did seem to have at least a rudimentary ability to add and subtract. Not bad for babies who haven't even learned to talk yet!

These are just two examples of a huge body of post-Piagetian research that undermined Piaget's stage theory of cognitive development. Young children simply had greater cognitive skills than Piaget's theory allowed them. These empirical findings, in turn, led developmentalists to propose alternative theories of cognitive development.

Link to an interview with Annette Karmiloff-Smith.

Similar findings have been obtained in other areas, such as logical thinking.  Piaget argued that formal logic was a relatively late cognitive accomplishment, but some psychologists have found evidence of logical thinking even in very young infants.  For example, Visual behaviors, such as a shift in one’s gaze or a prolonged stare, can be diagnostic of internal thoughts. Cesana-Arlotti et al. ("Precursors of Logical Thinking in Preverbal Human Infants", Science, 03/16/2018) asked whether infants (aged 1 to 1-1/2 years) can entertain a disjunctive syllogism of the form Either A or B is true; A is false; Therefore B must be true.  When the infants were presented with scenes depicting the syllogism, they looked longer, expressing surprise, when B proved to be false. 

Novices and Experts

One post-Piagetian approach construes development as the development of cognitive skills. According to this view, the infant starts out as a novice in all domains of problem-solving, and acquires expertise through learning - through experience and practice. However, in this view expertise is not just a quantitative difference of "knowing more" than one did before. Rather, the argument is that experts represent problems differently than novices do:

One model for the development of expertise is the difference between novice and expert chess players. Both kinds of players know the rules, but experts represent the game differently, and play differently, than novices do.

Of course, all of this is not very far from the theory of development as learning. Recall the definition of learning as "a relatively permanent change in behavior that results from experience"; in the same way, expertise develops with experience and practice. However, there are at least two important differences between the theory of expertise and the theory of learning.

These differences between the theory of expertise and the theory of learning reflect the lasting contribution of Piaget to developmental theory -- despite the fact that his theory appears to be wrong in many salient details.

Metacognition

Theories of expertise seem to imply some degree of continuity over the course of development, but studies of expertise reveal one big difference between younger and older children: Put bluntly, older children know what they're doing, while younger children don't. Older children are not simply more expert than younger ones: they're also more reflective, more deliberate, and more strategic in their thought and action. In other words, what older children possess, and younger children lack, is metacognition.

Metacognition was defined by John Flavell (1979) as, literally,cognition about cognition -- or, put another way, our ability to monitor and control our own cognitive processes. Metacognition is one's "knowledge about cognition and cognitive phenomena", including:

Flavell has argued that there are several different aspects of cognitive monitoring, including:

His general idea is that with development, children make more conscious, deliberate use of their mental faculties.

The Theory of Mind

More broadly, we might say that as they develop children come into possession of a theory of mind -- a term coined by Premack and Woodruff (1978; see also Premack, 1988) based on their comparative study of cognition in humans and chimpanzees. Put briefly, the theory of mind is the ability to impute mental states to oneself and to others. It includes:

The development of a theory of mind is commonly indexed by what is known as the false belief task. This task, typically, involves an experimenter, a child, and a puppet. The puppet hides a ball in an oatmeal container. After the puppet is put away in a cupboard, the experimenter and the child together switch the ball from the oatmeal container to a box. Then the puppet is brought out of the cupboard, and the child is asked where the puppet will look for the ball.

An early study by Wellman et al. found that children younger than 40 months typically failed the false-belief test, while children older than 50 months typically passed it. somewhere between the ages of 4 and 5, children get a theory of mind -- they understand that our minds are our own, and that different people will have different percepts, memories, knowledge and beliefs.

Of course, the moment researchers established a landmark like this, other researchers began to try to push the envelope back -- to determine whether even younger children have a theory of mind, too, if only we tested them the right way. One important feature of the standard false-belief test is that it is verbal -- the experimenter asks questions, and the child has to answer. So what would happen if we concocted a nonverbal version of the false-belief task?

To make a long story short, Onishi and Baillargeon (2005) tested 15-month-old infants on a nonverbal version of the false-belief task. The experiment itself is a thing of wondrous beauty, they devised a totally nonverbal version of the FB task, relying on the general finding that even infants look longer at events that violate their expectations.

So even very young children have some rudimentary theory of mind, even if they can't express it verbally.

I know what you're thinking: if we have a nonverbal test of the theory of mind, maybe we can use it to see if non-human animals have a theory of mind, too. In fact we can, and Call and Tomasello did, and they found that chimpanzees utterly failed the test. But that's a study for another course (I discuss it in my courses on "Scientific Approaches to Consciousness" and "Social Cognition" -- so if you're really interested, go to those websites and click on "Development").

The "Theory" Theory

In some respects, cognitive development is the development of social cognition -- the ability to think about oneself and other people. The acquisition of a theory of mind is a qualitative change, like the shift from one of Piaget's stages to another. But it is a qualitative shift that takes place against a continuous acquisition of knowledge.

But it turns out that the child isn't just developing a theory of self and others. The child is developing a theory of the whole world -- of physics and biology as well as psychology -- and testing it out in much the same way that a scientist would (the metaphor of child as scientist is another legacy of Piaget's theory). This is the essential proposition of what has come to be known as the "theory theory" of development -- that the developing child is engaged in a continuous process of proposing, testing, revising, and rejecting theories of how the world works -- including theories of how minds work, as part of the world.

The "theory theory" of cognitive development takes seriously Piaget's notion that the child is operating as a naive scientist, actively exploring the world and experimenting with it -- formulating hypotheses ('I wonder if it works this way"), gathering evidence ("Let's see what happens if I do this"), and revising hypotheses based on the outcome ("OK, that didn't work, maybe it works this way, instead"). In this way, the child develops theories of the world -- abstract, coherent systems of knowledge that enable him or her to predict or control events, and also to interpret and explain events.

Like the view of development as the acquisition of expertise, the "theory theory" emphasizes continuities rather than changes in cognitive development. The child is constantly experimenting, right from birth. Put another way, the child is constantly learning -- learning to predict and control the world around him. Learning, in fact, is central to theory-formation.

But while some conditioning theorists viewed the learning organism as a tabula rasa, or blank slate --what the 18th-century French philosopher called a "perfect idiot" -- which is written on by experience, that's not the view of the "theory theorists". Instead, they believe that the child comes into the world with an innate theoretical capacity -- a rudimentary ability to form, test, and revise its understanding of the world. So, like Piaget, they believe that the child comes into the world already prepared with a rudimentary cognitive schema. This position is known as starting-state nativism, and it holds that in some nontrivial sense the child comes into the world with "substantive innate theories" of various domains -- of physics, biology, a theory of the mind, and, possibly, a theory of society as well. The child develops as it puts these innate theories into action, testing them in the real world, and revising them when the test results prove to be surprising.  Just as in science, what starts out as a very narrow, primitive theory of some domain becomes progressively more expansive, refined, and robust -- a theory that actually predicts, and explains, what goes on in the world outside and inside the child's mind.

Much of what we know about what infants know comes from studies using the "looking" paradigms of the sort employed to investigate the infant's concept of number and understanding of other minds.  Infants tend to look longer at novel, unexpected events -- as if they're surprised by them, and are trying to figure out what's going on.  So, by tracking what infant look at, we can figure out what they already expect -- what they already know.  Other studies observe the infants' actions -- what they reach for, what they crawl to, what they imitate. 

Using these sorts of paradigms, we now understand that even infants have some rudimentary understanding of basic physical principles:

They haven't been taught these things, and they seem to know them before they've had any opportunity to learn them.  In fact, it seems that this innate knowledge about the physical and social world makes it possible for them to learn new things.

Probability Learning

According to the "theory theory", infants then build on this innate theoretical knowledge to develop a more refined understanding of themselves and the world around them -- all without benefit of language or much deliberate teaching on the part of their parents.

As discussed in the lectures on Language, infants quickly begin dividing up the sound world to which they are exposed.  They learn to expect which syllables are likely to follow other syllables, and which musical tones are likely to follow other tones (e.g., Saffran, Aslin, & Newport, 1996).  Of course, as discussed in the lectures on Learning, nonhuman animals do much the same thing -- picking up on conditional probabilities in the course of classical and instrumental conditioning.  So it's not a particular surprise that human infants can do this, too.  But the point is that infants already "know" something about probabilities.

They already now something about sampling, too.  UCB's Fei Xu (2008) showed 8-month-olds a container full of colored ping-pong balls -- e.g., 80% white, and 20% red.  The experimenter then dipped into the container, and removed five balls.  The infants showed sins of surprise if the color distribution of her selection departed markedly from the color distribution in the container as a whole -- showing that they already had some idea that a sample should resemble its parent population. 

Given these rudimentary conceptions of probability, infants can begin acquiring knowledge about their world -- what goes with what, and what causes what to happen. 

And this can take a while.  In phylogenetic terms, those species with the most intelligent and flexible adults also have the longest periods of infancy. 

For more on this, see "How Babies Think" by UCB's own Allison Gopnik (Scientific American, July 2010), from which these examples are drawn. 
See also Gopnik's books intended for a popular audience, including

• The Scientist in the Crib: Minds, Brains, and How Children Learn (1999; reviewed by Jerome Bruner, who was Gopnik's mentor in graduate school, in "Tot Thought", New York Review of Books, 03/09/2000);  and 
• The Philosophical Baby: What Children's Minds Tell Us About Truth, Love, and the Meaning of Life (2009).

Ethics and Morality

What an innate theory might look like is illustrated by research concerning infants' conceptions of ethics and morality. Long before they've gone to Sunday School, and long before they've received any serious rewards or punishments, infants apparently have some rudimentary sense of good and bad, right and wrong. Early on, of course, they have some of the prerequisites for making moral judgments.  For example, they understand the difference between animate and inanimate objects -- a logical prerequisite to understanding the concept of agency.

An experiment by Kuhlmeier, Wynn, and Bloom (2003) shows that even infants have a rudimentary concept of agency.  They watched an animated film of a ball climbed a hill: another figure, a triangle, pushed the ball up, while another figure, a square, kept him down.  Later, they saw a test film in which the ball approached either the square or the triangle on level ground.  Five-month-old infants didn't discriminate between the two films, but 12-month-old infants did.  Apparently, the older infants expected the triangle to approach the "good" object that "helped" it up the hill, but not the "bad" object that hindered its progress.

In another series of experiments, Felix Warneken and his colleagues arranged accidents, like a dropping a pen.  Infants as young as 18 months of age will actually help the experimenter -- but not if the experimenter intentionally throws the pen on the floor. 

The age of the infants in these studies, and the behavior of the chimpanzees, suggests to some theorists that children are born with an innate, if rudimentary, sense of right and wrong -- in innate moral knowledge that is further amplified through learning.  Others deny this, but claim that infants are very fast learners in this domain.  Perhaps humans and other primates are "prepared" to learn about right and wrong, even if they're not actually born with this knowledge.

The Pendulum of Developmental Theory

In some sense, theories of development cycle back and forth between continuity and change, and between qualitative and quantitative changes. Developmental psychologists aren't just on a swinging pendulum, however. Every turn in the cycle represents an increasingly sophisticated shift in our understanding of the nature of mental development -- and, for that matter, in our understanding of the mind.

Cognitive Aging Erikson was clear that social development continued throughout the lifecycle: roughly half of his "Eight Ages of Man" occur after, in Piaget's view, the child acquires formal operations -- and certainly after children acquire a theory of mind. The pimplication of Piaget's theory, however, is that cognitive development stops with the acquisition of formal operations.  After that, intelligence assumes a kind of steady state. Or, worse, it's often said that the story after early adolescence is one of steady cognitive decline (never mind the effects of age-related dementias such as Alzheimer's disease).  But it turns out that even that story is complicated.  Recall, for example, from the lectures on Thinking, that -- at least in Cattell's view -- fluid intelligence may decline with age, but crystallized intelligence stays steady, or increases, as the individual acquires new knowledge.  Similarly, recalling (sorry!) the lectures on Memory, procedural knowledge may decline, especially if we no longer practice a particular skill (I used to be a pretty good French horn player, but that was more than 50 years ago, and I don't think I could manage more than the C-major scale now).   One function that seems to consistently decline with age is processing speed, as measured by reaction times on various cognitive tasks; but even in this case there are complications, depending on what's being processed.  And, in any event, this would just disadvantage older individuals on "speed" as opposed to "power" tasks.  The elderly may not win Jeopardy!, with its signalling devices, but they will do just fine at Trivial Pursuit. A review of performance on various standardized tests of intelligence and memory (such as the WAIS) by Joshua Hartshorne and Lura Germine (Psychological Science, 2015) shows a clear decline in performance with age.  What's even more interesting, however, is that the peak of performance varies, depending on the function being tested.  Of course, as with all studies such as this, there is considerable variability around each of these data points, meaning that there are even some very old individuals who continue to perform reasonably well.  One important principle in cognitive aging, articulated most forcefully by Nancy Denney at the University of Wisconsin (Developmental Psychology, 1984), is "use it or lose it".  That is, individuals who continue to practice a cognitive skill show less decline in that skill than those who do not.  So, if you can, keep playing that French horn. A related point is that it is possible for middle-aged and elderly individuals to acquire new skills.  A social trend toward lifelong learning is documented in several recent books, such as Beginners: The Joy and Transformative Power of Lifelong Learning by Tom Vanderbilt (reviewed in "The Joys of Approaching Life san Amateur" by Cal Newport, New York Times, 01/31/2021), or Late Bloomers: The Hidden Strengths of Learnng and Succeeding at Your Own Pace by Rich Karalgaard (both reviewed in "Starting Fresh" by Margaret Talbot, New Yorker, 01/18/2021).  A spectacular example is that of Nell Irvin Painter who, after a distinguished career at Princeton as a historian of the 19th-century American South, decided to become a painter, completing both BFA and MFA programs in retirement.  It's not that Painter turned to a talent or interest that she already possessed, but had set aside during her academic career.  She just decided that she was going to become a real painter, not a dilettante like Winston Churchill or George W. Bush '43, good as these amateurs were and are (see her memoir, Old in Art School: A Memoir of Starting Over). Of course, it's may be easier to regain a skill that had once been laid down.  That's what savings in relearning is all about.  From time to time, I think of going on the market for a used French horn.  But I don't think my family would like it. Whether it's learning something for the first time or relearning something you used to know how to do, Rachel Wu of UC Riverside and her colleagues have identified a number of factors that promote success (Human Development, 2016): Open-minded input-driven learning, which relies on new observations rather than prior knowledge. Individualized scaffolding, such that the steps in skill-acquisition are arranged in increasing order of difficulty. A growth mindset (the term comes from Carol Dweck, as discussed in the lectures on Motivation) that abilities are not fixed and innate, but can improve with practice and effort; A serious commitment to learning. A forgiving environment that supports the learning activity, even through initial failure; A practice of learning several skills simultaneously, so what what you learn in one domain may help you learn in another. Wu et al. note that these six factors are present when children learn, and decline in the environments in which adults learn.  The implication, is that lifelong learning requires not just what the Buddhists would call a "beginner's mind" (in which habitual modes of thought disappear, making the world seem new and unfamiliar), but also a "beginner's environment".

The Phylogenetic View of Development

One important perspective on the development of mind is provided by evolution. The phylogenetic point of view on development traces the evolution of mind in the human species as a whole, often by comparing mental processes in subjects of different species. This is the field known as comparative psychology. It is an interesting challenge to develop tests of perception, memory, learning, categorization, problem-solving, and even language that can be reasonably applied to nonhuman animals, and comparative psychologists often exercise great ingenuity in their work.

Another expression of the phylogenetic point of view is evolutionary psychology, an offshoot of the sociobiology proposed by E.O. Wilson in his book by that title. Sociobiology assumed that patterns of social behavior evolved in the service of adaptation. Put another way, large segments of social behavior are instinctual in nature. Similarly, evolutionary psychology assumes that mental functions also evolved to serve adaptive purposes. Put another way, these modes of experience, thought, and action are also instinctual in nature -- they are part of our innate biological endowment, a product of evolution.

Kinds of Selection

Most discussions of biological evolution focus on Darwin's first principle of natural selection -- that those traits, resulting from natural, random variation, that enhance the organism's ability to survive in a particular environment, and thus pass its genes to its offspring. In the case of Darwin's finches,

Darwin's concept of natural selection was deliberately modeled on the "artificial" selection by which farmers and ranchers have "improved" their livestock since time immemorial. Thus, consider Carl Sagan's example (in Cosmos: A Personal Voyage) of the samurai crab (Heikea japonica). The Japanese "Tale of the Heike" tells the story of Heike warriors in a naval battle in the 12th century, who committed suicide by jumping overboard rather than face defeat. Later, fishermen working in the vicinity of the battle caught some crabs whose shells resembled human faces. Believing that these crabs were the reincarnations of the drowned warriors, they returned them to the sea. So, the face-like appearance of the shell became an adaptive trait, which was passed on to subsequent generations.

Detail from illustration: The ghost of Taira Tomomori and heikegani with faces of fallen soldiers,ukiyo-e print by Utagawa Kuniyoshi (1797-1861).

Photograph of a demon-faced crab, found in the waters surrounding Japan (Smithsonian Magazine).

But it turns out that natural selection is not the only possible form of selection. There are certain traits that do not necessarily promote the survival of the individual organism, but which are adaptive in other ways, and thus also likely to be passed from one generation to another.

In sexual selection, certain traits enhance the likelihood that the organism possessing them will be able to mate -- even if that trait is maladaptive in other respects. The classic example are strength and size (in males), elaborate plumage in birds (again, males), and various kinds of courtship displays. the idea actually was originated by Darwin's grandfather, Erasmus Darwin, but Charles Darwin elaborated on the concept in The Descent of Man and Selection in Relation to Sex (1871). Sexual selection is theoretically important, because it makes clear that the basis for evolution is not merely "survival of the fittest", as the stereotype goes, but rather reproductive success. Fitness doesn't mean mere survival. It means the ability to pass on one's genes. The classic example of a trait subject to sexual selection, one noted by Darwin himself, is the male peacock's display of tail-feathers. Another, also noted by Darwin, is the bower bird of Australia and New Guinea, which adorns its nest with all sorts of colorful objects, including fruits and flowers but also found objects like pieces of glass, in an attempt to attract a mate. Richard Milner, in the Encyclopedia of Evolution: Humanity's Search for Its Origins (1990) characterizes sexual selection as "survival of the flamboyant".

A further principle of group selection is often invoked to show how certain group behaviors, like cooperation, might be adaptive. The classic examples are found in social insects: beehives often function as if they were a single organism. Individuals contribute to the reproductive success of the group, and so pass on certain genes that are common to their group -- even if they (and their immediate family members) do not reproduce. But it's not the group's survival that's at issue. Rather, the selection favors genes that the group members have in common. Milner (1990) calls group selection the "survival of the social unit". Evolutionary psychologists often invoke group selection as the biological basis of religion.

In kin selection, organisms pass on traits that do not necessarily promote their own. This concept has its origins in the work of William Hamilton, who was interested in problems of altruism. For example, why do worker ants and bees, themselves sterile, sacrifice themselves to serve their queen? Hamilton's answer is that such activities incidentally increase the chances that one's own genes will be passed on. Even if the individual dies, the survival of his genetic relatives insures that the family's genes are passed on. Consider that, by virtue of sexual reproduction, each individual organism shares about 50% of its genes with first-degree relatives. We share an average of 50% of our genes with our parents, siblings, and children; an average of 25% of our genes with our first cousins, and an average of 12.5% of our genes with our second cousins. If an organism has a trait, like working instead of reproducing, that increases the likelihood that its (many) relatives will reproduce successfully, then it thereby increases the likelihood that it's own genes will be passed on as well. Instead of the individual's reproductive fitness, kin selection enhances inclusive fitness. J.B.S. Haldane, a British geneticist, once remarked that he would be willing to sacrifice himself if he could be assured that his genes would live on: when pressed further, he announced his willingness to die for "two brothers, four uncles, or eight cousins" (as quoted by Milner, 1990; see also Lehrer, 2012). Kin selection is the basis of Richard Dawkins's theory of "the selfish gene" -- that it's not species that are struggling to reproduce themselves, nor individuals, nor group members -- but, rather, genes themselves.

According to William Hamilton, a British mathematical biologist who undertook to express Haldane's theory precisely, rB >C, meaning that a gene for altruism could evolve if the benefit of the genetically controlled behavior exceeded its cost, provided that genetic relatedness was taken into account. This formula became the basis of inclusive fitness theory which gets its name because the reproductive value of the trait includes the individual's close genetic relatives as well as the individual itself. Hamilton's theory, in turn, was popularized by E.O. Wilson, the Harvard entomologist who, in such treatises as Sociobiology: The New Synthesis (1975) and On Human Nature (1979), was among the first to apply evolutionary theory to human social behavior -- a foreshadowing of today's evolutionary psychology.

Inclusive fitness theory, in turn, has been challenged by other theorists, including now Wilson himself, who have proposed a principle of group fitness instead. According to this theory, the reproductive value of a trait includes other members of the individual's social group (think: tribe), regardless of how genetically related the group members are. In some sense, group fitness reverses inclusive fitness. In inclusive fitness, the effect of the gene determines he composition of the group, because it favors the survival of genetic relatives. But in group fitness, the group comes first, and the gene favors the survival of group members.

In fact, both principles, kin selection and group selection, operate, but at different levels: we have genes for both selfishness (which favors the survival of the individual and his genetic relatives) and cooperation (which favors the survival of unrelated group members), which tend to oppose each other. As Wilson (2007) has put it: "Selfishness beats altruism within groups. Altruistic groups beat selfish groups. Everything else is commentary."

For the record, there's at least one other form of selection, known as balancing selection.  Consider the distribution of the "Big Five" personality traits discussed in the lectures on "Personality and Social Interaction".  We know that each of these traits is, to some degree, under genetic control.  And we can assume, as well, that there is an optimal distribution for each of these traits -- let's say, for purposes of argument, a fair amount of extraversion, agreeableness, and openness to experience; lots of conscientiousness; and not too much neuroticism.  How come natural selection hasn't operated to give us all the optimum levels of these traits, just as it's given every normal human the capacity for language?   How come there are any neurotics, or psychopaths, at all?  Why don't we all just get along?  According to the principle of balancing selection, different combinations of traits are optimum for different environments, both physical and social.  If everyone were low on neuroticism, it would be adaptive for "neurotic" people to be aware of, and responsive to, threats that non-neurotic people are oblivious to.  If everyone were high on openness to experience, it would be adaptive for "closed" people not to go jumping out of planes without a parachute.  So, over time, evolution establishes a sort of equilibrium.  And because the genetic contribution to neuroticism or openness isn't a single gene, but rather a large number of genes, as more-neurotic and less-neurotic people mate and produce offspring, you tend to get something that looks like a normal distribution.

Genes, Culture, and Altruism Evolutionary psychologists get twisted up in knots trying to explain altruism: How could it possibly be adaptive for an organism to sacrifice itself for others?  How could such a tendency be built into the genes?  It's to solve this problem that ideas like kin selection and group fitness were put forth.  But these principles have an ad-hoc quality to them.  In scientific terms, they're not very parsimonious, and appear (to me) to have been proposed in acts of desperation to make sure that there is a genetic, biological, Darwinian explanation for everything.  The first thing to ask is: how often does true altruism occur?  How often do individuals sacrifice themselves for others who are not their genetic offspring?  I bet a statistically valid survey would answer: not very often.   One place where it does occur is on the battlefield, where soldiers often (but  not always -- which is not a criticism) often sacrifice themselves to save others.  Consider, to take a nonrandom example, Humayun Khan, son of Khizer and Ghazala Khan, who famously reprimanded soon-to-be-President Donald Trump for his attitudes toward Islam and Muslims.  In 2004, while serving as an  Army captain in Iraq, he ordered his company to stay sheltered while he confronted a suspected car-bomber alone.  His suspicions proved valid, and the car bomb went off: Khan is now buried at Arlington National Cemetery, but the rest of his company lived.  This happens with some frequency in the military and first-responders at home (consider the firefighters who went up the Twin Towers on 9/11, when everyone else was coming down.  These examples suggest that altruism isn't built into the genes.  It's built into the culture -- more precisely, the particular sub-culture that is the military, or the fire department, or similar organization. So altruism does occur, and it has to be explained.  One way of studying altruism has been through the Prisoner's Dilemma, a game which I discussed briefly in the lectures on "Personality and Social Interaction".  In the game, two players take the roles of criminal suspects, A and B, who face a prison sentence for committing a crime (Poundstone, 1992).  The prosecutor offers each of them the following deal: If A and B both confess, saving the prosecution the expense of a trial, they will each be sentenced to 2 years in prison. If A confesses and implicates B, A will go free while B will serve 3 years. If B confesses and implicates implicates A, B will go free while A will serve 3 years. If both stay silent, each will serve 1 year in prison on a lesser charge. Here's a depiction of the payoff matrix in the classic Prisoner's Dilemma:  Prisoner A Prisoner B Cooperate Defect Cooperate A -- 1 Year B -- 1 Year A -- 3 Years B -- 0 Year Defect A -- 0 Year B -- 3 Years A -- 2 Years B -- 2 Years The two prisoners cannot communicate with each other, so each has a choice of whether to cooperate by remaining silent or to defect (compete) by confessing.  Obviously, defection is the rational choice for the one who defects; but cooperation is the altruistic outcome, because is minimizes the loss to the other person as well as to oneself.  But cooperation requires each prisoner to trust that the other will not defect.   There are two basic versions of PD.  In the standard PD, there is just one round of the game.  In iterated PD, there are several rounds, and subjects' behavior changes over time.  On the first round, as in the standard game, many, perhaps most, subjects compete by defecting.  On the second round, an initially cooperative subject may retaliate by competing in turn.  In this "tit for tat" strategy, each player mirrors the other's behavior.  Eventually, however, both players come to understand that they can maximize their joint outcomes by cooperating.  In addition to this process, known as direct reciprocity, there are other processes involved in the emergence of cooperative behavior (for details, see "Five Rules for the Evolution of Cooperation" by Martin A. Nowack, Science, 2006; also "Why We Help" by Nowak, Scientific American, 07/2012): Spatial selection: neighbors tend to cooperate with each other, and the network of cooperators gradually expands. Kin selection, as discussed earlier. Indirect reciprocity, in which people help others who have already established a reputation for helping. Group selection, also as discussed earlier, in which individuals help others for the "greater good". Nowack argues that evidence for all these mechanisms can be found in many different species of animals, suggesting that they have in common some basis in Darwinian evolution by natural selection.  But, as he also points out, some of these mechanisms aren't carried on the genes.  In particular, direct and indirect reciprocity depend on experience, and the sharing of information (e.g., about reputation) via language.  This is especially the case for direct reciprocity, which depends on players' ability to learn from experience; and indirect reciprocity, which depends on players' ability to learn from others (i.e., social learning).  The mechanism for Darwinian evolution by natural selection is based on genetic variability in innate traits.  But we also know that there's no inheritance of acquired characteristics.  So, helping behavior isn't all encoded in the genes.  It's also encoded in knowledge, beliefs, and attitudes -- it's encoded in the mind.

Our Phylogenetic Heritage

Humans have a particular place in the phylogenetic scheme of things. The earliest analyses of the human place in nature were based on morphological similarity -- the similarity in form and structure between humans and other animals. From this perspective, humans are warm-blooded vertebrates, primates related to the hominoid apes: orangutans, chimpanzees, gorillas, and gibbons. One popular morphological analysis holds that we are most closely related to orangutans.

Mammals, Primates, Hominoids, Hominins, Hominids, Humans

The fossil evidence suggests a gradual divergence among the hominoids. Life on earth began about 3 billion ya, during the Precambrian era of geologic time, in a probiotic soup of organic molecules (i.e., molecules containing the element carbon). For the next 2 billion years, only very simple life forms -- bacteria and algae -- existed. About 500 million ya, during the Paleozoic era,complex invertebrates began to evolve. Then, especially in the Mesozoic era, about 250 million ya, came the vertebrate species -- fish, then amphibians, then reptiles. Finally, in the cretaceous period of the Mesozoic era, beginning about 145 million years, ago, and especially in the tertiary period of the Cenozoic era, beginning about 65 million ya, birds, and mammals.

This is what our earliest mammal ancestor, the morganucodon (Morganucodon oehlieri, to be exact) looked like, as depicted in a bronze sculpture at the Smithsonian Museum of Natural History in Washington, D.C.

Among mammals,primates are a relatively recent development. All primates have a set of morphological features in common, that tend to distinguish them from other mammals:

The earliest primates, emerging more than 60 million ya, were tree-dwelling ancestors of present-day tree-shrews and lemurs. About 40 million ya, the "higher primates" -- or, more correctly, the anthropoid primates -- began to emerge. These came in two groups, evolving in distinct areas of the globe: the "New World" monkeys, first appearing in North America, but then colonizing Central and South America; and the "Old World" monkeys and great apes, in Eurasia, diversifying into Africa.

Hominoids and Hominids

The great apes -- present-day gibbons, gorillas, chimpanzees, and orangutans -- are also known as hominid primates, and they share ancestors in common with modern-day humans.

The fossil evidence further indicates that the gibbons split off from the rest of the primates about 25 million ya. Then, about 19 million ya there was a big split, dividing chimps and gorillas from humans and orangutans. Finally, about 18 million ya, hominins -- the ancestors of modern humans -- split from orangutans. Thus, by the morphological and fossil evidence, humans are most closely related to orangutans.

For many years this was the standard view in the field. More recently, however, this view has been revised by genetic evidence. The human genetic endowment consists of 23 pairs of chromosomes, which are contained in the nucleus of each cell in the human body. (There is one exception to this rule: the sperm cells of men, and the egg cells in women, contain only one randomly selected member of each pair. When the egg is fertilized by the sperm, the chromosomes in one cell are matched with their counterparts in the other, yielding a one-celled embryo that contains all 23 of the required pairs, one chromosome in each pair contributed by each parent.

Each chromosome consists of thousands of genes. These are the basic units of heredity, and affect the organism's physical features, and the course of its development. The genes themselves are composed of DNA (deoxyribonucleic acid), a chain of four chemical bases (adenine, guanine, thymine, and cytosine). Every gene is located at a particular place on a specific chromosome. According to the theory of evolution, closely related species have closely similar sequences of bases. Thus, examining the similarity in DNA molecules between modern humans and modern nonhuman hominoids indicates the evolutionary relations among these species. And with knowledge of the rate of DNA change, we can determine how early, or how recently, these species diverged.

It turns out, somewhat paradoxically, that morphological change is not necessarily related to genetic change. Comparisons of the structure of DNA in the blood (see material on genetics, below) indicate that humans are most closely related to chimpanzees -- in fact, we have about 98.5% of our genetic material in common what that species. Further, the genetic evidence indicates that the split between chimpanzees and hominids occurred only about 6 to 8 million ya.

So, our evolutionary history is encoded in our genes, and our genes tell us that we are most closely related to chimpanzees. At the same time, those 1.5% of genes that are not held in common by the two species can make quite a difference. Our characteristically hominid features include:

The last two features,vocal apparatus and brain mass, are the most characteristically human. And, of course, the brain provides the biological substrate of cognition -- it supports our learning, thinking, and problem-solving. It also contains specific cortical structures specialized for language, permitting symbolic representation of objects and events, and flexible, creative communication with other humans. Thus, the most important legacy of evolution is the human mind -- a particularly powerful cognitive apparatus coupled with a capacity for linguistic representation and communication. This feature, the human mind, sharply divides the dullest human from the smartest chimpanzee -- it is the difference that 1% makes.

The special features of the human brain doesn't mean that other animals don't have minds. For example, pigeons have a high capacity for abstracting concepts. And chimpanzees and dolphins have some limited linguistic abilities (symbolic representation, some degree of flexibility and creativity) -- though no capacity for speech because of the different configuration of their vocal tracts. The mental abilities of nonhuman animals are interesting in their own right, and we can learn a great about ourselves from studying other species.

But this does mean that the human mind is something quite special, and that we should focus on its development in the life of the individual -- that is, move from the phylogenetic perspective on development to the ontogenetic perspective.

The Evolution of Hominins

The precise manner in which modern humans evolved from ancient hominids is not known, and there is considerable controversy over this matter among physical anthropologists. To make life even more interesting, every so often a new discovery will shake up the field of paleontology. Still, the basic outlines are known. The account that follows is based on my understanding of the theory of Donald Johanson, discoverer of the skeleton of Lucy, one of our earliest ancestors (the rival theory is by Louis and Mary Leakey). According to Johanson's view, the earliest hominids split off from the ancestors of African apes (gibbons, gorillas, orangutans, and chimpanzees) about 6-8 million ya.

About 4.4 million ya, in the midst of the Ice Age, the genus Australopithecus ("Southern ape") emerged. Australopiths have human-like teeth, but in other respects they resembled terrestrial apes: short body, long arms, small brain. Perhaps their most important physical feature was that they walked upright on two feet, thus freeing their arms and hands to make and use tools -- something which began about 2.5 million ya, and coincided with a period of increasing brain size.

The earliest known example of Australopithecus is, in fact, Lucy -- discovered in the Afar Triangle of northern Ethiopia, and thus named Australopithecus afarensis ("Selam", an infant A. afarensis also dubbed "Lucy's baby", although the fossil was about 100,000 years older than was also discovered nearby). A. afarensis lived from about 4 million to about 2.5 million ya. Another ancestor,Australopithecus africanus, lived from 3 to 1 million ya: specimens have been found in eastern and southern Africa; the most famous sites are in Kenya and Tanzania. Yet another ancestor,Australopithecus robustus, bigger and more powerful than the others of its kind (hence the name), lived from 2.5 to 1.5 million ya: specimens have been found in southern Africa. A fourth ancestor,Australopithecus boisei (named for Charles Boise, a benefactor of many fossil hunts) lived from 2.5 to 1 million ya.

Another hominid line,Paranthropus, lived from about 2.8 million to about 1.4 million ya.There's also a Paranthropus boise, with the same namesake.  There is evidence that P. boise lived in East Africa  alongside early early Homo species, which is how its got its name: para from the Greek for "beside" and anthropus, of course, for the Greek for "human being"  . 

For a summary of what we know about P. boisei, see "Meet Your Exotic, Extinct Close Relative" by Bernard Wood and Alexis Williams, American Scientist, 11-12/2020.

Actually, according to Johanson, neither Australopithecus nor Paranthropus is not a direct ancestor of humans. That honor belongs to another hominid entirely:Homo habilis ("handy Man", the initial capital indicating that it refers to both males and females of the species), discovered in eastern Africa by the Leakeys. H. habilis had a much bigger brain than any Australopith. It made and used tools, while the Australopiths probably did not. It lived as a community, building shelters and surrounding its camps with fences or windbreaks. H. habilis emerged in eastern, southeastern, and southern Africa about 2 million ya, and lived alongside several genera of Australopiths for about 500 thousand years. Homo habilis apparently gave direct rise to another ancestor,Homo erectus ("upright Man"), which lived from about 1.6 million ya to about 200 thousand ya. H. erectus has been found everywhere in the Old World, including Europe, Asia ("Peking Man"), and Southeast Asia ("Java Man"). It had an even bigger brain, a better toolkit and building materials, hunted, and used fire.

About 300 thousand ya, yet a new subspecies,Homo sapiens ("Wise Man"), emerged. To put it bluntly, this is us. The archaic form of H. sapiens, popularly known as Neanderthal Man (from fossils found in the Neander Valley near Dusseldorf, Germany), controlled fire and made clothes; thus they were the first hominids to be able to survive in cold climates (naked humans cannot survive outside the tropics). They cared for the sick and buried their dead. They produced art. But they didn't last. They competed unsuccessfully with another subspecies,Homo sapiens sapiens ("very wise Man", I guess), popularly known as Cro-Magnon Man (from fossils found at Cro-Magnon, in southwest France). Neanderthals went extinct approximately 30,000 ya. Recently, Neanderthal Man has been renamed h. neanderthalensis, and modern man, simply,h. sapiens.  The illustration at left shows a reconstruction of a Neanderthal skeleton (foreground), with a skeleton of Homo sapiens in the background (from the American Museum of Natural History).

Neanderthals get a bad rap, because they were replaced by modern humans.  Then again, they lasted for about 350,000 years, and never once came even close to blowing themselves up with nuclear weapons, or threatening the planet with global warming.  And, for the record, the first modern humans to migrate out of Africa didn't last too long, either.  A more positive appreciation of Neanderthals is found in Kindred: Neanderthal Life, Love, Death, and Art (2020) by Rebecca Wragg Sykes.  Reviewing the book in Science (10/30/2020), Emma Pomeroy, an anthropologist at Cambridge University, wrote that "Wragg Sykes evaluates the available evidence on Neanderthals with empathy and even-handedness, revealing the group to be less 'them' and more 'us'".  (For another, more extensive review, see "Why Did They Vanish?" by Tm Flannery, New York Review of Books, 05/13/2021.).

The ancestors of Neanderthals emigrated from Africa roughly 500-250 thousand ya, turned north and west, and ended up in Europe.  Another archaic form of homo sapiens, known as the Denisovans, turned east, and ended up in Asia (the first Denisovan bones were discovered in Denisova Cave, in the Altay Mountains of southern Siberia).  Unfortunately, the only fossils definitively known to be Denisovan consisted of a single finger and a couple of teeth from a single individual ("Denisova Girl")found in Siberia, and a jawbone found in China.  However, Carmel, Gokhman, and their colleagues were able to use advanced DNA analyses (don't ask) to generate a guess at what Denisova Girl might have looked like.  The image, published in Cell (2019), turns out to be consistent with the independently discovered jawbone -- but the researchers caution that this is only a reconstruction of a single individual who may or may not be representative of her species as a whole.


About 50,000 ya, the ancestors of modern humans also left Africa, encountered and interbred with both Neanderthals and Denisovans (depending on which turn they took), competed with them, and became dominant, after which the Neanderthal line died out.  As a result of this cross-breeding, however, modern humans still carry some Neanderthal genetic material -- amounting to as much as 5% of the human genome.

Here's one version of the story, based mostly on DNA evidence (Akey et al., Science, 2016):

1. The first encounter with Neanderthals occurred soon after modern humans left Africa, somewhere in West Asia, leaving traces of Neanderthal DNA among the ancestors of modern Europeans, East Asians, and Melanesia.  (At this point, the ancestors of modern Melanesians split off from the Europeans and Asians.)
2. A second encounter with Neanderthals, somewhere in the Middle East resulted in further interbreeding with the ancestors of modern Europeans and East Asians, and South Asians.
3. An encounter with Denisovans resulted in interbreeding with the ancestors of modern Melanesians.
4. And a third encounter involved only Neanderthals and the ancestors of modern East Asians.
5. It was long thought that the ancestors of Neanderthals migrated out of Africa, turned left or right, but never turned around.  In 2020, researchers from the Max Planck Institute for the Science of Human History, found that some Neanderthal DNA can be found in the genomes of modern Africans, providing evidence of 'backmigration" from Eurasia into Africa.
6. Recent DNA modeling by Alan Rogers, a population geneticist at the University of Utah, suggests that there may have been interbreeding between Neanderthals, and Denisovans and "archaic", "ghost" lineages, such as h. erectus, who also migrated out of Africa (Science, 02/21/2020, from which the image at the right is taken).  Some of this interbreeding may also have involved the ancestors of modern humans.  If so, we don't just carry a little Neanderthal DNA with us -- the "ghosts" in the human family tree go back way further than that.
7. Actually, recent DNA sequencing studies (e.g., Petr et al., Science, 09/25/2020) indicate that there may have been two encounters between Neanderthals and h. sapiens.  The first migration of h. sapiens out of Africa and into Europe may have occurred as long as 200-300,000 years ago.  These first modern humans in Europe died out, but not before transferring some of their genes (ahem) to Neanderthals who were already living there.  Then there was a second migration about 40-80,000 years ago, after which the Neanderthals went extinct, and the modern humans thrived (this is the standard story).
   

There may have been other, isolated encounters, some of which did not result in the passing of DNA to descendants. But the four interminglings documented in the DNA evidence means that the ancestors of modern Melanesians got only one "pulse" of Neanderthal DNA -- most of their archaic DNA comes from Denisovans.  Europeans and South Asians got two "pulses" of Neanderthal DNA, while the ancestors of modern East Asians got three.  And of course, the ancestors of modern Africans, who never migrated out of Africa and so never encountered Neanderthals or Denisovans, got none at all.

This whole story is told by Svante Paabo, the Swedish biologist who first sequenced the Neanderthal genome, in Neanderthal Man: In search of Lost Genomes (2014), a book which has been compared with James Watson's The Double Helix (1968).

For a shorter version of the story, see "Neanderthal Minds" by Kate Wong, Scientific American, 02/2015.

Neanderthals and Cro-Magnons were closely related in genetic terms. In 2006, two different teams of researchers reported the first steps in reconstructing the Neanderthal genome, based on samples from preserved bone tissue: preliminary results indicate that the two genomes are about 99.5% identical.  So, that's 1.5% genetic difference between us and chimpanzees, and 0.5% difference between us and Neanderthals. It was also determined that Neanderthals had the FOXP2 gene that is considered critical (though probably not sufficient) for speech, if not language, suggesting that they may have had the ability to use language as well.  However, it's not really possible to make inferences about the mental and behavioral capabilities of Neanderthals from knowledge of their genomes.  For that, we have to rely on the physical evidence they left behind to be studied by archeologists

For a summary of what we know about Neanderthal psychology, see How to Think Like a Neanderthal by Thomas Wynn and Frederick L. Coolidge (2011).

The traditional theory is that modern humans invaded Neanderthal territory in Europe and eliminated them -- and that similar scenarios played out between modern humans and other bands of "archaic" humans.  However, the close similarity in DNA, and other evidence,now suggests that modern humans actually interbred with Neanderthals, and perhaps other archaic forms with whom they shared close genetic resemblance.  (See "Human Hybrids" by Michael F. Hammer, Scientific American May 2013).

These are the broad outlines.  It's clear where we came from, but the precise path is unclear, and it's certainly not a straight line from Pan troglodytes to Homo sapiens. 

For a recent overview of the complexities of human evolution, see "Shattered: New Fossil Discoveries Complicate the Already Devilish Task of Identifying Our Most Ancient Progenitors" by Katherine Harmon, Scientific American, February 2013.

Whether the Neanderthals were killed by Cro-Magnon, or interbred with them, the line died out about 30 thousand ya. H. sapiens sapiens thrived, through the Old (Paleolithic), Middle (Mesolithic) and New (Neolithic) Stone Age.

Why the Cro-Magnons thrived, while the Neanderthals did not, will likely remain an anthropological mystery. The most likely hypothesis is that the Cro-Magnons had a capacity for language, and for symbolic thought, that gave them an advantage (as indeed it would). So far as we can tell from the scant evidence, Neanderthal culture was pretty primitive even by the standards of Early Man -- much more so than Cro-Magnon culture. But it's all pretty speculative -- pretty close to reasoning backward from the fact that we have a sophisticated capacity for language and symbolic thought.

About the only place we didn't get by walking was Antarctica, and now we're there, too, with permanent settlements since 1959.

Pushing Back Our Origins

The story of human evolution from Australopithecus through Paranthropus to Homo is fairly well accepted, though there are disagreements about the details. However, as noted earlier, occasionally a major discovery will occur that calls for major revisions in the story. The major consequence of these new discoveries is to push the origins of hominids back a little (or, in some cases,a lot) further back in time, or to add a branch to the main lines of human descent. (Image by Viktor Deak from "The Human Pedigree" by Kate Wong, Scientific American 01/2009).

For example, until recently it was generally thought that the earliest hominids, known as Australopithecus, dated back some 4.5 million ya, to an area in eastern Africa. But in just two short weeks in 2002, different teams of paleontologists reported discoveries that pushed the emergence of hominids back a lot in time and over a bit in space.

Sahelanthropus tchadensis.The first of these, and the most surprising, was the discovery by Dr. Michel Brunet of the University of Poitiers, in France, of the "Chad skull" in Chad (hence its name), in the region of the southern Sahara desert known as the Sahel (actually, the skull was discovered by Ahounta Djimdoumalbaye, an undergraduate at Chad's University of N'Djamena, who was a member of Brunet's research group). This skull apparently belongs to a new hominid species known as Sahelanthropus tchadensis ("Sahel man of Chad") -- or Toumai ("Hope of Life") in Goran, the language of the people in the area where it was discovered. Toumai lived as long as 6 or 7 million ya, about 1 million years earlier than the earliest hominid previously known, and more than 3 million years before the famous Australopithecus known as "Lucy" (see below). Moreover, previous hominid skulls were found in the eastern and southern portions of Africa.

Before Toumai, the earliest known hominid,Orrorin tugenensis, dating from about 6 million ya, was discovered in Kenya.

The fact that the Toumai skull was found in western Africa suggests that our ape and hominid ancestors were more widely dispersed on the continent than previously believed. The Toumai skull mixes features of apes, such as a small brain-case, with features characteristics of hominids, such as a flat rather than protruding face. For this reason, it has been suggested that it may be an ancestor of both chimpanzees and hominids, living before the two lines diverged. Another theory is that it represents another primate branch that has simply gone extinct.

The "Georgia Skull".The other discovery, known as the "Georgia skull" because it was found in Georgia, a country that was formerly part of the Soviet Union, in an area near Tbilisi between the Black and Caspian Seas, is about 1.75 million years old and belongs to one or the other of three hominid species already known --Homo habilis,H. ergaster, or H. erectus. But it and similar skulls were found with stone tools, such as choppers and scrapers, that seem much more primitive than those typically associated with H. erectus, so some theorists , So, unlike the Toumai skull, it doesn't suggest any sort of "missing link". Rather, the surprise here is that the skull was found outside of Africa. Previously, it was believed that the Homo species evolved in eastern Africa, and that H. erectus moved out of Africa sometime more recently than 1.6 million ya. But here's a homo skull well outside Africa, and more than 1.6 million years old. So, the migration out of Africa might have begun earlier than we previously thought.

Homo floresiensis. In October 2004, a team of Australian anthropologists discovered a "downsized" or "hobbit" version of homo erectus, with adults standing only 3-31/2 feet high, in a cave on Flores, an island in the Indonesian archipelago near Bali. Individual members of this species, named homo floresiensis, have been dated between 95,000 and 13,000 ya -- meaning that they overlapped with h. sapiens, which arrived in Australia about 40,000 ya. Stone tools also found have been attributed to h. floresiensis rather than later Neanderthals -- which is interesting, because these humans had brains hardly larger than those of adult chimpanzees. So, it's not brain size that's related to intelligence, as brain structure -- including, perhaps, the sulci and gyri that feature so prominently in human brains, and that permit humans to pack a large amount of cortex into a relatively small space. The precise placement of h. floresiensis in the hominid family tree remains a matter of controversy. In some respects, they appear more primitive than either h. erectus or h. sapiens -- yet they made and used tools. Some anthropologists speculated that they might be genetically deformed h. sapiens dwarfs. As of 2009, the consensus was that they were, indeed, a distinct species of hominids, which split off from the main hominid line 1-2 million ya, more closely related to h. habilis than to h. sapiens..

Ardipithecus ramidus. In 2009, paleontologists from UC Berkeley and their colleagues from Ethiopia discovered a new fossil hominid,ardipithecus ramidus (nicknamed "Ardi") that is a million years older than Lucy, dating to about 4.4 million years before the present.Ardipithecus was discovered in the Awash River valley of Ethiopia, not very far from where Lucy had been found -- but in an area that was, at the time, forest rather than savanna. What's really interesting about Ardi is that it walked upright (if not particularly gracefully, with short legs and lacking an arched foot), but still had the capacity -- like long arms and opposing thumbs on the feet as well as the hands -- to move about easily in trees. It had previously been thought that bipedalism had emerged on the Savannah, as an adaptation useful for living in an area that had no trees. But Ardi suggests that bipedalism first emerged in an arboreal area. Apparently, by the time of Lucy, a million years later, bipedalism had advanced to the point that the new species,Australopithecus, could walk rapidly over long distances.

As of its discovery, Ardi was the oldest known hominid skeleton. It is different enough from chimpanzees to suggest that the common ancestor of chimpanzees and humans was about 6-7 million years old.

Australopithecus sediba. In 2011, a group of anthropologists at the University of Witwatersrand, in South Africa, led by Lee Berger, announced the discovery of a new pre-human species, named A. sediba, which appeared to comprise a "patchwork" of ape and human features. There is a hand that is able to grip tools as well as hold on to tree branches, and a foot that can support upright walking as well as climbing trees. The pelvis was human-like, able to accommodate big-brained fetuses; but the skull itself was small -- suggesting that the pelvis evolved, not to accommodate big-brained fetuses, but rather as a byproduct of bipedalism. The researchers dated the skeleton to about 1.977 million ya, and suggested that it is now the oldest "missing link" -- not between ape and man, exactly, but between the most recent known Australopith species,A. africanus, and the earliest known human species,H. habilis. This conclusion is, of course, vigorously disputed by other anthropologists. Perhaps more important, at least so far as the theory of evolution is concerned, is what it suggests about the emergence of complex human characteristics, such as the hand. Rather than evolving microscopically (or, for that matter, rather than being created instantaneously out of whole cloth), complex features like the hand might have evolved in a sort of modular fashion. There might be several different configurations of thumb, wrist, finger length, etc., all occurring more or less randomly, until one particular combination -- short, opposing thumb, long fingers, and a particular twist of the wrist, perhaps -- happened to appear, proved to be particularly adaptive, and so was passed down to subsequent generations while other combinations, perhaps less adaptive, simply (and literally) died out. [For more on A. sediba, see "First of Our Kind" by Kate Wong,Scientific American, 04/2012.]

Lucy's Grandfather(?)  In 2019, a group of Ethiopian researchers published a reconstruction of a skull of A. anamensis, the direct predecessor to "Lucy".  Previously, anamensis was known only from teeth and jaws.  But in 2016, Ali Bereino, an Ethiopian goat herder who had a side interest in fossil hunting, uncovered a complete anamensis skull while preparing a shelter for his animals.  The fossil, known as MRD for Mira Dora, the place where it was found, has been dated to 3.8 million ya.  The current theory is that Lucy's species, A. afarensis, branched off from A. anamensis about 3.7 to 3 million ya.  The researchers also promptly hired Bereino.

The Dmanisi Skulls.  In an interesting twist on the "Georgia skull" described earlier, a set of skulls uncovered at Dmanisi, Georgia, and reported in 2013, cast doubts on the traditional classification of early hominids into separate species -- Homo erectus, Homo ergaster, Homo habilis, and Homo rudolfensis.  The traditional story, remember, is that only H. erectus flourished, eventually evolving into H. sapiens.  The distinctions among the various species are based largely on features of the skull.  The Dmanisi skulls, which include the "Georgia skull"" are all dated to roughly 1.8 million ya, but they show amazing variability -- as much as the variability found in skulls from Africa, which led to the naming of several different species.  The implication is that there may have been just one early species of Homo, namely erectus, characterized by a high degree of within-species variability with respect to skull morphology.  Rather than there having been several different Homo species, all competing, as it were, for survival, there was just one -- H. erectus, which eventually evolved into H. sapiens.  Another implication is that, rather than waiting, H. erectus began to move out of Africa almost as soon as it appeared on the East African savanna.

In 2015, Lee Berger, an American anthropologist teaching at the University of Witwatersrand, in South Africa, reported the discovery of hominid skeletons in the Rising Star Cave near Pretoriain South Africa (the bones were accidentally discovered by spelunkers; Berger was also the discoverer of A. sediba, so he's one lucky paleontologist!).  Analysis strongly suggests that these bones come from an entirely new hominin species, Homo naledi ("naledi" means "star" in a local indigenous language).  Based on the geology of the cave and naledi's primitive anatomy, Berger has suggested that the skeletons date from more than 2.5-2.8 million ya -- that is, close to the dawn of H. sapiens.  As such, naledi might be close to the "missing link" between australopiths and homo.  A photo of a substantial portion of a naledi skeleton was published in Scientific American, 08/2017 ("Our Cousin Neo", by Kate Wong).  See also "Return to the Cave of Bones" by Lee Berger and John Hawks, National Geographic, 07/2023.)

Link to a NOVA/National Geographic documentary on the discovery of H. naledi, broadcast in September 2015.

All Aboard for Marrakech?  Excavation at a mining site near Marrakech, Morocco, uncovered human and animal bones and stone tools, originally dated to 40,000 BP and identified as Neanderthal remains.  But more recent investigation (Hublin et al., Nature, 2017) has reclassified the human remains as H. sapiens and dated the site to 350,000-280,000 BPE -- which is pretty surprising, given that standard theory dates the emergence of H sapiens to East Africa "only" 60,000 ya!  So if correct, the Morocco H. sapiens discovery would push back the origins of our particular species back by about 100,000 years.  They also indicate that some H. sapiens, instead of simply migrating out of East Africa, populated the entire African continent as well.  On the other hand, there are enough morphological differences between the Moroccan specimens and H. Sapiens that they might represent a distinct, yet to be named, H. antecessor.

From Fossils to DNA.  Most of the story of human evolution is told through fossils like Lucy or the Dmanisi Skulls, but it's now possible to answer some questions about our origins from analyses of DNA extracted from fossilized remains.  For example, Matthias Meyer, Juan-Luis Arsuag, and their colleagues (2013) sequenced the DNA from a 'Neanderthal" skeleton found at a site in Spain known as Sima de los Huesos, dated to 400,000 before the present era (BPE).  Surprisingly, they found that it closely resembled DNA extracted from a "Denisovan" skeleton found in Siberia, and dated to 80,000 BPE.  Based on the fossil evidence alone, it has long been assumed that Neanderthals and Deniisovans represented distinct species of Homo, which left Africa about 300,000 BPE.  The Neanderthals turned left into Europe, while the Denisovans turned right into Asia. Later, about 200,000 BPE, H. sapiens migrated out of Africa, and replaced both the Neanderthals and the Denisovans.  Possibly, the Spanish fossils represent yet another Homo subspecies, yet to be named.  Or, the Denisovans migrated to Europe as well as Siberia, and the two interbred.  Or something. 

A series of papers published in Nature in 2016 may have settled the matter.  Comparing samples of DNA collected from all over the world, including many populations of indigenous people, three separate teams of researchers concluded that all non-Africans are descended from a single group of H. sapiens that emigrated from Africa 50-80,000 ya.  This doesn't mean that there weren't earlier waves of migration, such as the Neanderthals and Denisovans.  It's just that the descendants of any earlier migrations didn't last.  So far as modern Europeans, Asians, Australian aborigines, and Native Americans, it seems that we're all descended from that last migration out of Africa,

Ian Tattersall, an anthropologist at the American Museum of Natural History, has traced recent changes in our understanding of human evolution with two excellent geneological trees -- the one on the left, drawn in 1993, and the one on the right, drawn in 2011 (from "Human Evolution in Perspective" by I. Tattersall, Natural History, 06/2015).  Note that h. naledi isn't there -- it's that new a discovery.

Breaking News! Just when you think everyone's reached consensus: along comes an anthropologist who tries to upend it all -- or most of it.  In 2021, Madeleine Bohme, an anthropologist at the University of Tubingen, intended to upend most, if not all, of the "Out of Africa" consensus (in Ancient Bones: Unearthing the Astonishing New Story of How We Became Human, co-authored with Rudiger Braun and Florian Breier; reviewed by Tim Flannery in "Out of Savannastan", New York Review of Books, 11/04/2021).  Briefly, that consensus has three elements: The hominem lineage, which ends with modern h. sapiens, split from chimpanzees between 7 and 13 million years ago. Our specific genus, homo, arose in Africa about 2.3 million years ago. Our specific (sorry) species, h. sapiens, arose in Africa about 300,000 years ago. First, a jawbone found in Greece during World War I (German soldiers were digging a bunker), and determined to be more than 7 million years old, was determined to belong to the family Hominiae.  That means that the oldest ancestor of modern humans lived in Greece, not Ethiopia.  Second, another paleoanthropologist spotted set of fossilized human-like footprints in some rocks by a beach in Crete (he was vacationing with his girlfriend) was dated to 6 million years ago.  That challenges Point #1. Second, Bohne also cites Homo wushanensis, a group of fossils found in China in 1991, which have been dated to about 2.5 million years ago -- making them older than the oldest Homo habilis fossils found in Africa.  That challenges Point #2 -- except that the researcher who originally dated the fossils now believes that he was mistaken. Bohne, for her part, believess that our genus, Homo, evolved in a once-grassy woodland known to professionals as "Savannastan", which encompassed parts of Africa, Asia, and Europe about 2.6 million years ago  -- hence the Georgia skull, hence the Cretan footprints, hence the Greek jawbone.  In any event, Point #3, that Homo sapiens arose in Africa, remains unchallenged.  So far.  Stay tuned.

For an excellent survey of human evolution, see Masters of the Planet, and the Strange Case of the Rickety Cossack: and Other Cautionary Tales from Human Evolution, both by Ian Tattersall (2012).

"Races"

Still, as I noted earlier, the main outlines of the story of human evolution remain intact.Here's the story as of Fall 2009, as depicted in the Science paper announcing Ardi.

As H. sapiens sapiens proliferated around the world, adaptation to different climatic zones produced the physical differences associated with the different "races" of humans (much as h. floresiensis developed small stature in an. For example, pale skin is an advantage to those living in cool, cloudy lands because it facilitates the absorption of ultraviolet radiation, promoting the manufacture of vitamin D, a substance necessary for healthy bone growth. For people who live in the tropics, skin darkened by melanin protects against peeling and blistering, cancers caused by constant exposure to these same ultraviolet rays. In the tropics, a tall, slim physique radiates surplus heat, and keeps the body cool. In the cold climates of the far north, a short, squat body with high fat levels serves an insulating function. In areas with unreliable or inadequate supplies of food, short stature is adaptive, as are fatty buttocks. Improved diets increase stature and shrink teeth, muscles, and bones.

Much ink (and blood) has been spilled over the biological reality of various racial distinctions. Perhaps the best perspective on this issue is presented in an essay, "Human Equality is a Contingent Fact of History", by Stephen Jay Gould, a paleontologist at Harvard, who contributed a regular column, "This View of Life", to Natural History magazine.  The essay, written while Gould was giving a series of lectures on racism in South Africa (whose black majority was then suffering under the white-imposed apartheid regime) was published in the November 1984, issue of the magazine, and contains Gould's reflection on the evolution of our notions of race and racial differences.  In his article, Gould argues that "Human equality is a contingent fact of history" (italics original).  By this he means that it could have happened that different, and unequal, races evolved through human history; it just didn't happen that way.  Initially, Western thought was thoroughly imbued with the idea that the different races had separate origins (polygeny), and that racial inequality was somehow a "natural" consequence of this separate creation.  But with the gradual acceptance of Darwin's ideas about The Origin of Species, suggesting evolutionary links between humans and other animals, Western views of race have themselves evolved.

• Early- to mid-19th century anthropologist debated between two geneological theories of racial origins.  According to a monogenic theory, all human lineages had their origins with Adam and Eve, but some races degenerated from this Edenic state.  According to a polygenic theory, Adam and Even were the ancestors only of whites, and the other, "colored" races were the product of separate creation.  When Darwin came along, the argument shifted: all races had a primeval ancestor in common, but the various races began quickly began to diverge.  This latter theory, of common origins but ancient separation, was promulgated as recently as the 1960s by Carleton Coons (Origin of Races, 1962), who argued that "caucasoids" and "mongoloids" were more advanced than "australoids", congoids", and "capeoids" because they evolved further to adapt to the more challenging environments of the northern climate.  The modern geneological view is that the modern races are all "recent, poorly differentiated sub-populations of... homo sapiens, products of at most tens or hundreds of thousands of years, and marked by remarkably small genetic variations".
• Early anthropologists also explained racial differences in terms of geographical theories.  Darwin had speculated that h. sapiens had its origins in Africa.  But, in an echo of the revised geneological view, others argued for Asian origins, and that the "tropical" races reflected a subsequent migration into the "less challenging" zones of Africa, Australasia, and the Americas -- and, therefore, a less advanced form of human.  When the evidence for African origins became uncontrovertible, the argument shifted to a more psychological one: that human consciousness and intelligence evolved in Asia.  As Coon wrote: "Of Africa was the cradle of mankind, it was only an indifferent kindergarten.  Europe and Asia were our principal schools".  The modern geographical view is that both h. erectus and h. sapiens arose in Africa and then spread outward. 
  

In addition to debunking the geneological and geographical arguments for racial inequality, Gould also presented what he called "positive" arguments for the equality of the races.

• First: technically, biologists recognize only one division within a species -- that of subspecies, which inhabit a particular geographical region and display recognizable traits that make them distinctive.  But he points out that subspecies function only as "categories of convenience" representing geographic variation.  But designation of a subspecies makes sense only when the groups are geographically distinct.  To be blunt about it, the human races aren't geographically distinct.  We intermingle and interbreed, resulting in "fluidity and gradation", not distinct subcategories, so there's not really much point in naming the races as if they really constituted anything like subspecies.
• Obvious "racial" differences may exist, such as skin color, but these are of relatively recent origin -- to recent to provide the foundation for any sort of inequality among the races.
• Empirically, there just aren't enough genetic differences between the races to permit them to be meaningful subcategories.  As Gould wrote in 1984, "Intense studies for more than a decade have detected not a single "race gene" -- that is, a genes present in all members of one group and none of another.  Frequencies vary, often considerably, among groups, but al human races are much of a muchness".  Echoing Lewontin's summary of within-group and between-group differences, Gould goes on to note that "Variation among individuals within any race is so great that we encounter very little new variation by adding another race to the sample.  In other words, the great preponderance of human variation occurs within groups, not in the differences between them....  The recent origin of races... squares well with the minor genetic differences now measured.  Human groups do vary strikingly in a few highly visible characters (skin color, hair form) -- and this may fool us into thinking that overall differences must be great.  But we now know that our usual metaphor of superficiality -- skin deep -- is literally accurate.
  

More recent research by Marc Feldman and associates, based at Stanford University, suggests that there are, after all, small genetic differences among the races. Surveying the DNA sequences of 1000 people sampled from each of 52 populations, he found that DNA differences between the groups fell into five clusters, or groups, roughly corresponding to their continents of origin: Africa, Eurasia (including Europe, the Middle East, and South Asia), East Asia, Oceania (including Australia), and North and South America. These, of course, correspond to the five geographical "races" of folk taxonomy: Negroid, Indo-European, Mongolian, Pacific Islander, and (American) Indian. 

For more details of the genetic differences between the races, see A Troublesome Inheritance by Nicholas Wade (2014).  But pay attention only to the first half of the book, which sets out the genomic differences among the geographical races.  There are some, but you should also bear in mind that these racial differences are minuscule when compared to the entire human genome, with its 22,500-some genes.  As Richard Lewontin, a prominent population geneticist, has pointed out, the differences between racial or ethnic groups are far smaller than the similarities.  Put another way -- and this is a general rule: the differences between groups are smaller than those within groups.  In the second half, Wade tries to argue that racial differences in social outcomes -- such as the "achievement gap" between Asian-, European-, and African-Americans -- are due to these differences -- and here, his science is on very shaky ground indeed.  Earthquake-sized shaky.  Magnitude 8 shaky.  Not least because the differences in social outcomes within the races are far greater than the differences between them.  Moreover, Wade offers no evidence that any of the genetic differences between the races, small as they are (and they are tiny), have anything to do with the behavioral differences that he is concerned with (or much of anything else, except skin color and lactose tolerance). 

For a counterargument, that race is essentially a social construction, see Fatal Invention: How Science, Politics, and Big Business Re-Create Race in the 21st Century by Dorothy Roberts (2014).

A more balanced view is offered by Adam Rutherford, a science journalist with a PhD in genetics, in A Brief History of Everyone Who Ever Lived (2017).  Rutherford argues that genetic studies can tell us a lot about where we came from -- it turns out, for example, that the British people generally known as "Celts" (Welsh, Irish, Scots, etc.) are not genetically related: "Celt" really is a social construction!  Neanderthals had the FOXP2 gene that is crucial for speech.  And it's now been demonstrated convincingly that every person now living on Earth, no mater what their "race", is descended from a very small group of H. sapiens who lived just a few thousand ya.  Outward appearances to the contrary notwithstanding, our genomes don't divide us into the usual racial categories.

See also Black and White, a special issue of National Geographic (04/2018) devoted to the biological and sociocultural aspects of race.  The two girls on the cover are Marcia and Millie Biggs, fraternal twin daughters of Michael Biggs, a native of Jamaica, and Amanda Wanklin, an Englishwoman (you've seen them before, in my discussion of behavior genetics).  One of the articles in the special issue, "A Color Wheel of Humanity" by Nina Strochlic,  illustrates the amazing variety of human skin tones by selections from Humanae, a project by portrait photographer Angelica Dass, who matched the skin tones her subjects to the standard color palette maintained by Pantone (discussed in the lectures on Sensation).  

The important thing to remember is that despite superficial physical differences (and even more superficial differences in biochemistry), all "races" of H. sapiens sapiens, Black, White, Mongoloid, or whatever, reflect differences in the same species of animal. Moreover, there are major individual differences among members of any single "race", and there are people with superficially similar features -- Africans and Australian Aborigines, for example, who are not members of the same "race". The bottom line is that while "race" has some degree of biological reality, it is largely a mythical concept that is better discarded. All of us have a single common ancestor. In fact, there is a theory, based on analyses of mitochondrial DNA (which is passed only from female to female) that the entire race of modern humans -- black, white, and yellow -- are the descendants of a single female who lived in East Africa 200 thousand ya. This theory has been called into question, but the essential point remains: we are all very closely related to each other, so we might as well treat each other with respect and affection.

The best guess is that any two randomly selected individuals are more than 99% identical in their genes -- if not 99.9% identical, then maybe 99.5% identical. And remember, there is about 98% similarity between the human genome and that of chimpanzees, our closest primate relatives. So, there may be some minor genetic differences between people of different continental ancestries (African, Eurasian, and East Asian), which is a nice way of saying different "races", but "race" is still largely a social construct -- a way of classifying people that has no serious biological justification.

The Beginning (and Perhaps the End) of History

Since the Ice Age ended, about 10 thousand ya, H. sapiens sapiens has continued to thrive. During the New (Neolithic) Stone Age, hunters and foragers became farmers and herders who cultivated plants and domesticated animals.

Neolithic people built villages and towns that gave rise to cities, the development of economies that were not devoted solely to the production of food, and the rise of hierarchical social structures. Ceramics were developed to store food. Reliance on stone tools gave way to bronze and copper -- first by hammering metal, later by forging and casting it. The earliest wooden plows date from 6 thousand ya; the first wheels for transportation, 5.5 thousand ya; the first sailing ships, and the first writing, 5 thousand ya.

Not all of these developments occurred simultaneously in every geographical area; and in some areas, some developments did not occur at all.

When writing begins, where it begins, history begins too, and science and culture begin to develop and proliferate extremely rapidly. At this point, about 5000 ya, biological evolution essentially ends: in genetic terms, and speaking metaphorically, we are the same species as Adam and Eve (created, or so Bishop Usher calculated, on the night before October 23, 4004 BCE). The development of tools, clothes, medicine, and social structure mean that we are protected -- or, more correctly,we protect ourselves -- against the biological pressures that formerly killed those who were weak or stupid. The evolution of a new species requires biological isolation, and migration and inbreeding, within and between "racial" groups, effectively precludes that. It also requires a hostile environment, which insures the survival only of the fittest. So, biologically speaking, we are pretty much at the end of our line. Now the only threats to our existence come from ourselves -- overpopulation, ecological disaster, and nuclear holocaust.

To our knowledge, few other species have lasted even three million years before their inevitable extinction. But unlike nonhuman animals, we know what the threats to our existence are, we understand that they are largely of our own making, and we have the intelligence and technology to do something about them. We can save ourselves from extinction, but only if we think, and try. That's where human intelligence comes in. The ultimate gift of evolution, the human mind, has been and will remain the key to our survival as a species.

Actually, Though, It Ain't Over 'Till It's Over

It's common to think that the biological evolution of the human species has pretty much reached its end point. That's pretty much the point of view taken here: once people start changing the environment, the environment has less chance to change them through natural selection. And, more or less, that's also the point of view taken by evolutionary psychologists (see below), who argue that patterns of human thought and behavior that evolved in the late Pleistocene era have remained pretty much unchanged up to the present.

These are good arguments, but they're apparently not quite true. Obviously, there's still opportunity for natural selection to operate on the human genome. In fact, in 2006, Jonathan Pritchard and his colleagues identified a number of segments of the human genome that have been subject to change via natural selection as recently as the last 5-10,000 years -- roughly since the beginning of agriculture (for the details, see A Troublesome Inheritance [2014] by Nicholas Wade).

In 2011, a study of parish records from the Canadian Isle aux Coudres, which date back to 1799, revealed a steep decline in the age at which women had their first child. Other data confirms a decline in age at first reproduction, and an increased age at menopause -- both of which have been plausibly attributed to natural selection (although, frankly, it seems to me that both changes could just as easily have occurred as a result of improvements in nutrition and other aspects of health). If the evolutionary biologists are right, evolutionary change in these traits has been both very recent and very rapid.

So, evolution continues at the genetic level, at the level of body morphology, mind, and behavior, if not the mind as well. The fact that human evolution continued even after we moved out of the EEA, and into other environments that made other demands on survival, has led some evolutionary psychologists to espouse a concept of "fast evolution" -- essentially, that evolution can proceed at a much faster pace than previously believed -- possibly in response to much more recent environmental changes. And to focus on the Holocene epoch rather than the Pleistocene -- i.e., the most recent 12,000 years, since the beginnings of agriculture.

OK, point well taken, but let's not go overboard.

How Did We Get to Be Human? Evolution did not draw a straight line from early hominins to modern humans. At one point, we shared the planet with a number of near-relatives. Note: On 11/19/2018, the New York Times commemorated 40 years of its "Science Times" section by looking at "11 Things We'd Really Like to Know -- And A Few We'd Rather Not Discuss".  One of the essays, by Carl Zimmer, traces progress in understanding human origins -- and our deepening understanding of our complex evolutionary history. I spoke recently to a scientist who was writing up a summary of what we know about human evolution. He should have had a head start, having written a similar article five ya. But when he looked at what he had written then, he realized that little of it was relevant. “I can’t use much of any of it,” he told me. As a journalist, I can sympathize. In recent years, scientists have offered a flood of insights into how we became human. Fairly often, the new evidence doesn’t square with what we thought we knew. Instead, many of these findings demand that researchers ask new questions about the human past, and envision a more complex prehistory. When Science Times debuted 40 ya, scientists knew far less about how our ancestors branched off from other apes and evolved into new species, known as hominins. Back then, the oldest known hominin fossil was a diminutive, small-brained female unearthed in Ethiopia named Lucy. Her species, now known as Australopithecus afarensis, existed from about 3.85 million ya to about 2.95 million ya. Lucy and her kin still had apelike features, like long arms and curved hands. They could walk on the ground, but inefficiently. Running was out of the question. Hominin evolution appeared to have taken a relatively direct path from her to modern humans. The earliest known members of our genus, Homo, were taller and had long legs for walking and running, as well as much larger brains. Eventually, early Homo gave rise to our own exceptional species, Homo sapiens. Now, it’s clear that Lucy’s species wasn’t the beginning of our evolution; it was a branch that sprouted midway along the trunk of our family tree. Researchers have found fossils of hominins dating back over six million years. Those vestiges — a leg bone here, a crushed skull there — hint at even more apelike ancestors. But even the earliest known hominins were like us in one important regard. They appear to have been able to walk on the ground, at least for short distances. Paleoanthropologists have uncovered a wealth of new fossils from all points on the spectrum of hominin evolution. Some clearly belonged to known species, such as Australopithecus afarensis. Some were so distinct that they deserved a new designation. But others have fallen somewhere in between. Often they look like mosaics of other species, carrying remarkable combinations of traits. Some of these mosaics may have been the result of interbreeding between species. But it may be, too, that hominins independently evolved many traits many times, along separate lines of evolution. All this mixing and experimentation produced as many as 30 different sorts of hominins — that we know of. And one kind did not simply succeed another through history: For millions of years, several sorts of hominins coexisted. Indeed, our own species shared this planet with near-relatives until just recently. In 2017, researchers found the oldest known fossils of our species in Morocco, bones dating back about 300,000 years. At that time, Neanderthals also existed. They continued to live across Europe and Asia until 40,000 ya. At that time, too, Homo erectus, one of the oldest members of our genus, still clung to existence in what is now Indonesia. The species did not go extinct until at least 143,000 ya. Homo erectus and Neanderthals are hardly new to paleoanthropologists. Neanderthals came to light in 1851, and Homo erectus fossils were discovered in the 1890s. But still other hominins, recent research has shown, shared the planet with our own species. In 2015, researchers unearthed 250,000-year-old fossils in a South African cave. Known as Homo naledi, this new species had a Lucy-sized brain, but it was also a complex structure in ways that resembled our own. The wrist and other hand bones of Homo naledi were human-like, while its long, curved fingers seemed more like an ape’s. While Homo naledi thrived in Africa, another mysterious species could be found on an island now called Flores, in Indonesia. Known as Homo floresiensis, these hominins stood only three feet high and had brains even smaller than that of Homo naledi. The species may have arrived on Flores as early as 700,000 ya, and these hominins endured until at least 60,000 ya. Homo floresiensis appears to have made stone tools, perhaps to hunt and butcher the dwarf elephants that once lived on the island. Paleoanthropologists today are no longer limited to just examining the size and shape of fossils. Over the past 20 years, geneticists have learned how to extract DNA from bones dating back tens of thousands of years. In one remarkable discovery in Siberia, researchers examining a nondescript pinkie bone discovered the genome of a separate line of hominins, now known as Denisovans. As it turns out, we have had the planet to ourselves only in the past 40,000 years — a small fraction of Homo sapiens’ existence. Perhaps we out-competed other species. Maybe they just had bad luck in evolution’s lottery. But in one way, we are still living with them. Both Neanderthals and Denisovans interbred with our ancestors some 60,000 ya, and billions of people today carry their DNA. Still mosaics, after all this time.

Evolutionary Psychology

Evolution doesn't just leave its mark on body morphology, giving fish scales and birds feathers, and humans opposable thumbs. It also leaves its mark on behavior, as seen in the "instincts", or fixed action patterns, discussed in the lectures on learning. In the 1970s, the evolutionary biologist E.O. Wilson coined the term sociobiology to represent the idea that patterns of social behavior evolved under the pressure of natural selection, just as physical traits did. In other words, a number of human social behaviors are instinctive, part of our innate behavioral endowment. We've seen instincts before, in the context of innate stimulus-response connections (remember reflex, taxis, instinct?). In the last chapter of his book, Wilson argued that instinctual social behavior might not be restricted to nonhuman animals like ants (the species Wilson studied) or the species studies by ethologists like Tinbergen, Lorenz, and von Frisch, but might extend to humans as well.

Taking a leaf from Wilson's book, some psychologists -- led by Leda Cosmides, John Tooby, and David Buss, among others -- have argued that mental traits evolved in the same way: that human beings, no less than other animals, have evolved specific patterns of thought, feeling, and desire through natural selection (in fact, Buss's first book was entitled The Evolution of Desire). The reason some of our behaviors and thought processes seem maladaptive, or at least inappropriate, in today's world is that they evolved to foster adaptation to a particular environment, known as the environment of evolutionary adaptedness (also known as the "environment of early adaptation", in either case abbreviated EEA) -- roughly the African savanna of the Pleistocene epoch (modern Ethiopia, Kenya, and Tanzania), where homo sapiens first emerged about 300,000 ya -- and have changed little since then.

Although these assertions are debatable, to say the least, the literature on instincts makes it clear that evolution shapes behavior as well as body morphology. Many species possess innate behavior patterns that were shaped by evolution, permitting them to adapt to a particular environmental niche. Given the basic principle of the continuity of species, it is a mistake to think that humans are entirely immune from such influences -- although humans have other characteristics that largely free us from evolutionary constraints.Since the emergence of humans, the cultural environment has changed a great deal, but there has not been enough time for biological evolution to produce new, more adaptive traits.

Certainly there are good reasons for believing that the uniquely human capacity for language is a product of evolution. So, arguably, are the kinds of mechanisms envisioned by Gibson's idea of direct perception. But there are reasons for thinking that the theory of biological evolution is not the answer to psychology's problems. This is because there are at lest four characteristics of mind that appear to distinguish humans from all other animals.  As outlined by Marc Hauser (himself a distinguished evolutionary psychologist) in "Origin of the Mind" (Scientific American, 09/2009), these are:

1. Generative computation: through recursive and combinatorial thinking, humans are able to "create a virtually limitless variety of words, concepts, and things".  Hauser, Noam Chomsky, and W. Tecumseh Fitch have argued that recursion is the key to human linguistic ability (Science, 2002).  See also "The Uniqueness of Human Recursive Thinking", by Michael C. Corballis, American Scientist, 05-06/2007).
2. Promiscuous Combination of Ideas: intermingling knowledge across different domains "thereby generating new laws, social relationships and technologies".
3. Mental Symbols: representing both real and imagined experiences, which can be expressed to others through language.
4. Abstract Thought: the ability to deal with objects and events that we cannot physically sense.
   

Evolutionary psychologists who study mating frequently place a great deal of stress on a particular dimorphism in which older males prefer younger females. This pattern makes evolutionary sense, given that young females (at least, young women, as opposed to young girls) have greater childbearing capacity than older ones (at least, women who have reached menopause). In turn, evolution is evoked to explain a wide variety of mating phenomena, from President Bill Clinton's entanglement with Monica Lewinsky to the common practice of rich men abandoning their first wives (or their second, or their third...) to take younger, ostensibly more attractive "trophy wives" (a term coined by Fortune magazine in the 1980s).

But...

Similar problems attached to other evolutionary explanations of mating behavior, which are commonly explained by some variant of Robert Trivers's (1972) parental investment theory.  According to the theory, men evolved to produce lots of offspring, but women evolved to be more selective about whom they'll mate with, because they have to invest more in caring for their offspring than men do.  Therefore:

It all sounds good, especially when you consider the conditions of child-rearing in hunter-gatherer societies.  On the other hand, there is dispute about the facts to be explained.

For comprehensive analyses of gender differences in mating and other behavior, see:

• Peterson, J.L.,  Hyde, J.S., "A meta-analytic review of research on gender differences in sexuality, 1993-2007" (2010), which concludes that "most gender differences in sexual attitudes and behaviors are small", especially in "nations and ethnic groups with greater gender equity", they also note that "Gender differences decreased with age".
• Conley, T.D., et al. "Women, Men, and the Bedroom: Methodological and Conceptual Insights that Narrow, Reframe, and Eliminate Gender Differences in Sexuality" (2011), whose subtitle says pretty much all there is to say.
• Carpenter, C.J., "Meta-Analyses of Sex Differences in Responses to Sexual Versus Emotional Infidelity:Men and Women Are More Similar than Different" (2012).
• Guildersleeve, et.al., "Do women’s mate preferences change across the ovulatory cycle? A meta-analytic review" (Psychological Bulletin, 2014).
• Wood et al., "Meta-analysis of menstrual cycle effects on women’s mate preferences" (Emotion Review, 2014).
  

So far, evolutionary psychology has gone through three stages:

Evolutionary psychology tends toward the reductive -- that is, it seeks an explanation for everything about mind and behavior in terms of its role in enhancing reproductive fitness. You can see this especially in the debates over homosexuality and altruism.

These folks have even offered an evolutionary explanation for grandmothers. Why should humans have females who, almost alone in the animal kingdom, continue to live beyond their childbearing years? Because they remain available to take care of their children's children -- thus, indirectly, enhancing the reproductive fitness of their children and grandchildren, and, indirectly, passing on their genes. Honestly, I swear this is true, you can't make this stuff up.

I've argued that biological evolution is outpaced by cultural evolution, mediated by social learning, but some evolutionary psychologists have gone so far as to claim that learning itself is a biological adaptation -- that learning itself is a product of natural selection.  In this way, they try to have it all.  But they can't, for the simple reason that, by claiming learning for biological evolution, they've distorted the meaning of both evolution and learning. Yes, there's a sense in which learning occurs via something that looks like natural selection: according to Skinner, for example, behaviors that are reinforced are maintained, while those that are not reinforced disappear.  But this is nothing more than an analogy.  Darwinian theory, and the evolutionary psychology that is based on it, assumes genetic variation, which is operated on by the environment.  But learning doesn't depend on genetic variation.  It depends on cognitive variation -- that is, variation in the thoughts, and behaviors, in the individual organism.  This is not within the scope of the Darwinian paradigm.  Learning is, in fact, more Lamarckian, as it involves the inheritance of acquired characteristics -- but not through genetic inheritance.  The inheritance that occurs in learning is cognitive and cultural inheritance, passed through social learning, not through genetic mechanisms.

Steven Pinker has written (2008): "To understand human nature, first understand the conditions that prevailed during most of human evolution, before the appearance of agriculture, cities, and government". Perhaps. Then again, it was by virtue of human nature that we invented agriculture, cities, and government in the first place. Human nature is not restricted to biological givens: it also extends to sociocultural constructions.

A more scholarly critique of evolutionary psychology has been provided by David J. Buller, a philosopher, who lists "Four Fallacies of Pop Evolutionary Psychology" (Scientific American, 01/2009) -- where "pop evolutionary psychology" (PopEP) "refers to a branch of theoretical psychology that employs evolutionary principles to support claims about human nature for popular consumption".  Here's his list of fallacies:

1. "Analysis of Pleistocene Adaptive Problems Yields Clues to the Mind's Design".  PoPEP-ists generally trace the mind's design to the problem of mate selection, but Buller points out that the paleontological record "is largely silent regarding the social interactions that would have been of principal importance in human psychological evolution".
2. "We Know, or Can Discover, Why distinctively Human Traits Evolved": The comparative method on which evolutionary biology relies involves studying species who share a common ancestor, but which developed different adaptations to deal with different environments -- as in the case of Darwin's finches.  That works well for finches, who come in great variety, but it does work for humans, who diverged from our closest relative, the chimpanzee, about 6 million ya.  the relatives that would allow the comparative method to work, like australopiths and other hominins, just aren't around for us to compare ourselves to.
3. "Our Modern Skulls House a Stone Age Mind".  Some human traits, like our basic emotions, arose long before the Pleistocene era; and environmental change since the Pleistocene has arguably altered human thought patterns as well.
4. "The Psychological Data Provide clear Evidence for Pop EP".  Here Buller criticizes PopEP-ists' failure to consider alternative explanations.  The fact that some human mental or behavioral characteristic fits an evolutionary explanation doesn't matter if there is an alternative explanation that fits better.  Consider, for example, the PopEP claim men are more upset by sexual infidelity, while women are more upset by emotional infidelity.  First, this isn't exactly true.  This finding emerges only from forced-choice questionnaires that require subjects to choose which kind of infidelity upsets them more.  When subjects are asked to rate how much they would be upset by each kind of infidelity, men and women come out about even.  Moreover, any such difference may have nothing to do with the environmental pressures on Stone age hunter-gatherers.  The same sex difference could reflect men's belief that sexual infidelity in women is generally accompanied by emotional infidelity, and women's belief that male sexual infidelity is generally not accompanied by emotional infidelity. 
   

Evolutionary psychologists have been extremely creative in conjuring up plausible accounts of how this or that characteristic of human mental life is adaptive -- or was adaptive in the EEA. But are these anything more than "Just So" stories, a la Rudyard Kipling? To quote Richard Lewontin again ("Not So Natural Selection",New York Review of Books, 05/27/2010):

"The success of evolutionary biology as an explanatory scheme for its proper subject matter has led, in more recent times, to an attempt to transfer that scheme to a variety of other intellectual fields that cry out for systematic explanatory structure....

"One answer has been to transfer the formal elements of variation and natural selection to other aspects of human activity.... We have evolutionary schemes for history, psychology, culture, economics, political structures, and languages. the result has been that the telling of a plausible evolutionary story without any possibility of critical and empirical verification has become an accepted mode of intellectual work even in natural science....

"Even biologists who have made fundamental contributions to our understanding of what the actual genetic changes are in the evolution of species cannot resist the temptation to defend evolution against its known-nothing enemies by appealing to the fact that biologists are always able to provide plausible scenarios for evolution by natural selection. But plausibility is not science. True and sufficient explanations of particular examples of evolution are extremely hard to arrive at because we do not have world enough and time. The cytogeneticist Jakov Krivshenko used to dismiss merely plausible explanations, in a strong Russian accent that lent it greater derisive force, as 'idel specoolations'.

"Even at the expense of having to say 'I don't know how it evolved', most of the time biologists should not engage in idle speculations."

 For another critical view of evolutionary psychology, see "It Ain't necessarily So" by Robert Gottlieb, New Yorker, 09/17/2012.  Here's an extract:

There are plenty of factions in this newish science of the mind.  The most influential... focuses on the challenges our ancestors faced when they were hunter-gatherers on the African savanna in the Pleistocene era..., and it has a snappy slogan: "Our modern skulls house a Stone Age mind."  This mind is regarded as a set of software modules that were written by natural selection and now constitute a universal human nature.  We are, in short, all running apps from Fred Flintstone's not-very-smartphone.  Work out what those apps are -- so the theory goes -- and you will see what the mind was designed to do.

Evolutionary Change and Cultural Change: The Case of Violence

Some inkling of the comparative speed of biological and cultural evolution, with respect to human experience, thought, and action, is afforded by an analysis of historical changes in violence by Steven Pinker -- himself a prominent proponent of evolutionary psychology. In The Better Angels of Our Nature: Why Violence Has Declined (2011), Pinker argues from archival data that the rate of violence among humans has declined radically from the Stone Age until now. For example, criminal records from the 14th century indicate that London's homicide rate was about 55 deaths per 100,000 population, compared to 2 per 100,000 today. And London is far from a unique case. Outside of the United States, for example, capital punishment has virtually vanished from the western world -- and, despite the large number of inmates on death row in states like Texas and California, actual executions are quite rare (OK, maybe not in Texas). Despite the horrors of World Wars I and II, and headline-grabbing terrorist bombings, campus shootings, domestic violence, and gang killings in our inner cities, Pinker asserts that "The decline of violence may be the most significant and least appreciated development in the history of our species".

Pinker's data has been controversial -- the two World Wars were pretty bad; there are persistently high levels of violence in Africa, Asia, and South America; and the homicide rate in American cities in 21st-century Detroit and New Orleans (though not New York City) rivals that of 14th-century London. But if he's right, this radical and rapid decline in violence cannot be accounted for by evolutionary change. After all, as noted earlier, when it comes to mind and behavior there's been no change in the human genome since the time of Adam and Eve. So it has to be a product of cultural change.

Actually, it doesn't have to be a product of cultural change alone. Pinker notes that, in addition to innate tendencies toward competition and violence -- our inner demons -- we also have countervailing innate tendencies toward cooperation and empathy -- our better angels (the phrase comes from Abraham Lincoln's first inaugural address). But Pinker is an evolutionary psychologist, and if our better angels were innately stronger than our inner demons there wouldn't ever have been high levels of violence, and there would have been no historical decline, either -- because there would have been no high level of violence to decline from. So, even from an evolutionary-psychological point of view, the key to the decline in violence has to lie in cultural change. Compared to 50,000 (or 5,000, or even 500) ya, the cultural environment has changed to favor our innate goodness rather than our innate badness. Following the sociologist Norbert Elias, Pinker calls this "the civilizing process".

So what are the elements of the civilizing process. Elias thought that it consisted of the elevation of state power over feudal loyalty, and also the development of commerce. Pinker's view is more monolithic. In addition to four innate better angels (like cooperation and empathy), and five innate inner demons (like competition and violence), he identifies six trends and five historical forces that have fostered the decline of violence. For example: