How, and how fast, did the human brain evolve?

November 24, 2010 • 2:29 pm

by Greg Mayer

While in Colombia last week, Jerry directed my attention to a paper by Roy Britten (abstract only free) in that week’s issue of the Proceedings of the National Academy of Sciences. Britten is a venerable figure in evolutionary molecular genetics, one of the pioneers of DNA-DNA hybridization who helped elucidate the structure of the genome long before sequencing was possible. The paper was indeed interesting. This post is a bit longer and more data-and-analysis-laden than usual, but I think the paper merits discussion.

Britten summarizes his latest paper’s conclusions succinctly:

The aim of this paper is an explanation for the high speed of evolution of the human lineage, which has been exceptional compared with other animals. The high speed of evolution of human lineage brain size is recognized by comparison of fossil brain sizes (1, 2). Evolution of the lineage leading to humans during the last several million years was striking. … A major source of variation [for brain evolution] has been the insertion of transposable elements (TEs).

He goes on to note that besides rapid brain evolution, humans have many TE insertions. For him

This is an extraordinary correlation. Human evolution has been rapid, particularly brain evolution in the last several million years. It is the only species known to make such rapid evolutionary progress. Now it is shown that human is the only species studied to have so many TE insertions. Recognition of this correlation leads to the concept that Alu insertions underlie rapid human evolution.

Importantly, he states up front that:

We believe the brain evolution was due to natural selection and genomic variation.

He is thus not seconding Colin Blakemore’s unwarranted claim that brain size is a neutral character, conferring neither selective advantage nor disadvantage, that must therefore have evolved via genetic drift. Britten is definitely not saying this, and is thus not open to the criticisms of Blakemore made by Jerry (here and here) and John Hawks. Britten thinks a big brain is advantageous.

So, how well supported are the claims he does make? I’d first note that one data point does not a correlation make, especially not an extraordinary one. TE’s, by increasing mutation rates, can certainly increase evolutionary rates, but all sorts of other singular characteristics apply to the Homo lineage. In addition to having many TE’s, they were savannah-dwelling, nearly hairless, bipedal, etc. Which of these correlates with rapid brain growth is the important one? I don’t know, but a more forceful argument than simple occurrence in the same lineage would be needed to establish which is the most likely causal factor.

And what about the factual points? Much of the paper is devoted to establishing the prevalence of TE’s in the human lineage, and I, as at least a first approximation, would yield to Britten’s expertise on this point. What about the rate of brain size evolution? (To be fair, Britten takes his cue here from the literature; the high rate is a premise, not a conclusion.) G.G. Simpson, one of the founders of modern evolutionary biology, spent a major part of his career documenting the variability of evolutionary rates. He showed that there is great variability of evolutionary rates between lineages, among characters within lineages, and within lineages at different times. The following figure is based on an original in Simpson’s 1953 Major Features of Evolution. It shows that the rate of evolution in lungfish was high about 300 million years ago, but not so much at other times (i.e. variation within lineages at different times).

Rate of lungfish evolution, from Mark Ridley's Evolution, 2004.

Historically, claims of human exceptionality have tended to become less exceptional when examined more closely. Huxley’s debate with Owen over the brain is perhaps the classic example: contra Owen, Huxley “showed that the brains of apes and humans were fundamentally similar in every anatomical detail.” Knowing this, I decided to check on this important premise of Britten’s paper. Is the speed of evolution of the human brain “striking”, and “exceptional compared with other animals”? In a word, no: over the last several million years, human brain size has evolved at rates which are typical of paleontologically measured evolutionary rates.

Here’s a graph by John Hawks (using the same or similar data as Lee and Wolpoff, 2003) showing the pattern of change in cranial capacity (on the vertical axis, in cubic centimeters) over the last 2 million years or so.

There are a number of ways of measuring evolutionary rates of morphological features such as cranial capacity. One useful measure is the haldane, developed by Phil Gingerich (1993) of the University of Michigan Museum of Paleontology, based on a suggestion made in 1949 by the original most interesting man in the world, the great geneticist-physiologist-soldier-pacifist-communist-Hindu-atheist-patriot-expatriate J.B.S. Haldane, one of the founders of modern evolutionary theory. The haldane is the change, on a logarithmic scale, of the feature in question in units of the standard deviation (a measure of how variable the feature is), per generation.

Using data from three papers on modern human cranial capacity, I found the average to be 1345 cubic centimeters (cc), calculated as the unweighted average of males and females from the measured populations from Korea, Turkey, and Nigeria (a total sample of 1151). 1.8 million years ago, the cranial capacity of the Homo lineage was 702 cc, calculated as the average of skulls of that age from Perning, Kenya (one each, given by Lee and Wolpoff, 2003), and Dmanisi (three skulls, given by Gabunia et al., 2000, for two of them, and the median of two estimates by Lee, 2005, for the third).

The logarithmic standard deviation is well approximated by the coefficient of variation (CV: the untransformed standard deviation divided by the mean; Lewontin, 1966). The unweighted average CV for the modern humans was .0955, and for the five early Homo it was .1149; averaging, we get .1052.

So, the amount by which cranial capacity has changed on the log scale is ln(1345)-ln(702)=.6502; dividing this by the estimated logarithmic standard deviation, .1052, gives 6.181. In the last 1.8 million years, our cranial capacity has increased about 6 standard deviations. We don’t know generation time for early Homo, but we do know it for modern humans (about 25-30 years) and chimps (19-24 years; Matsumura and Forster, 2008). Using 25 years as an estimate for the whole lineage, we get 72000 generations in the last 1.8 million years, giving an evolutionary rate for brain size of .00008584 (or 10^-4.066) haldanes.

Is this a high rate or a low rate? Neither– it’s absolutely typical for evolutionary rates measured over this generational time scale. Gingerich (2001) compiled a large data set on rates of evolution, measured in haldanes, over a wide variety of time scales. He states:

Macroevolutionary studies yield rates on the order of 10^-2–10^-6 haldanes calculated over intervals of geological time ranging from 10^2–10^6 generations.

If we look more precisely, at about 72000 (10^4.86) generations, we find measured rates of about 10^-3.5 to 10^-6.5. So the rate of human brain evolution is above the median, but nothing “exceptional”. Could this unexceptional result be due to the particular initial time (1.8 mya) selected? What if we looked further back in time? I redid the analysis using an average chimpanzee cranial capacity of 383 cc (McKee et al., 2005), and a divergence time of 6 million years. Using the same logarithmic standard deviation and generation time, we get  [ln(1345)-ln(383)]/.1052 = 11.940 standard deviations, nearly twice the 6 standard deviation change in the last 1.8 million years. Dividing by the 6 million years/25 years per generation = 240000 generations gives .00004975 (or 10^-4.3) haldanes. Again, above the median, but nothing exceptional.

So, there’s nothing much remarkable about the speed of human brain size evolution. If the TE’s (or hairlessness or bipedalism or whatever) of the Homo lineage had an effect on human evolution, it was not expressed as an unparalleled increase in the rate of evolution of cranial capacity.


Acer, N., M. Usanmaz, U. Tugay, and T. Ertekin. 2007. Estimation of cranial capacity in 17-26 years old university students. Int. J. Morphol. 25:65-70. pdf

Britten, R.J. 2010. Transposable element insertions have strongly affected human evolution. Proceedings of the National Academy of Science 107:19945-19948.

Gabunia, L., et al. 2000. Earliest Pleistocene hominid cranial remains from Dmanisi, Republic of Georgia: taxonomy, geological setting, and age. Science 288:1019–1025.

Gingerich, P. D. 1993. Quantification and comparison of evolutionary rates.  American Journal of Science 293A: 453-478. pdf

Gingerich, P. D. 2001. Rates of evolution on the time scale of the evolutionary process. Genetica 112-113: 127-144. pdf

Haldane, J.B.S. 1949. Suggestions as to quantitative measurement of rates of evolution. Evolution 3:51-56.

Hwang, I.-L., et al. 1995. Study on the adult Korean cranial capacity. Journal of Korean Medical Science 10:239-242. pdf

Matsumura, S. and P. Forster. 2008. Generation time and effective population size in Polar Eskimos. Proc. R. Soc. B (2008) 275:1501–1508. pdf

McKee, J.K.,  F.E. Poirier, and W.S. McGraw. 2005. Understanding Human Evolution. Pearson, Upper Saddle River, New Jersey.

Lee, S.-H. 2005. Is variation in the cranial capacity of the Dmanisi sample too high to be from a single species? American Journal of Physical Anthropology 127:263–266. pdf

Lee, S.-H. and M.H. Wolpoff. 2003. The pattern of evolution in Pleistocene human brain size. Paleobiology 29:186-196. pdf

Lewontin, R.C. 1966. On the measurement of relative variability. Systematic Zoology 15:141-142.

Odokuma, E.I., P.S. Igbigbi, F.C. Akpuaka, and U. Esigbenu. 2010. Craniometric patterns of three Nigerian ethnic groups. African Journal of Biotechnology  9:1510-1513. pdf

Simpson, G.G. 1953. Major Features of Evolution. Columbia University Press, New York.

21 thoughts on “How, and how fast, did the human brain evolve?

  1. I find the entire paper to be totally bizarre. It’s tough to find evidence of transposable elements in low-coverage genomes; it’s no surprise that humans appear to have a larger number than our relatives, even if they’re neutral, because of this ascertainment bias. When we have high-coverage genomes based on long-reads, then we will start to have some evidence about transposable element rates.

    Plus, the paper is full of “in conversation” and other such references. It seems like it wasn’t edited.

  2. Is this a peek into how the peer review process works? If so…I’d think that the reviewer’s notes might be at least as informative and useful as the original paper’s, and should probably be published together.



  3. Fascinating.

    Will you present this as a response to the paper? Seems as if most of the heavy lifting is done. Just coordinate against other evolutionary changes in apes (including humans), evolutionary changes in a non-mammalian species or two, and you’re publishable.

    The write-up of Haldane was also interesting. His is a name I had heard often, but hadn’t really contextualized until now.

    And you’ve now got me scurrying down google-lane for articles on TEs as well.


  4. I personally can’t fathom that PNAS continues to allow track I (see here for a criticism), and I basically think contributed papers are to be bypassed.

    However, while it may be so that human cranial capacity os not overwhelmingly fast, the idea that TEs are in part responsible for the still very clear increase is intriguing. Many much worse ideas have been espoused through track I, imo.

  5. I call your attention to Nick Matzke’s posts titled “Fun with Hominim Cranial Capacity Datasets” Parts 1 and 2 on Panda’s Thumb here and here. He uses the De Miguel and M. Henneberg 2001 data set of measurements which included a 29-page Appendix listing every known published measurement of a hominin skull older than 10,000 years to illustrate a similar point:

    The authors conclude that their study supports previous studies indicating that the change in cranial capacity can be explained by “a typical Darwinian mechanism” of exponential increase in a quantitative trait under selection. (Creationists, take note again: “exponential increase” does not mean “sudden” – think of compound interest.)


  6. Why do we assume that the cranial capacity is everything? Is it possible to redo this using other metrics, such as number of neurons or the area of the brain (say between Chimpanzees and modern humans)?

    1. Good point. My question is one of neuronal “content”; i.e., it’s not size alone that matters but “information stored”.

      In my little quasi-scientific mind I imagine the rapid evolution of the mind–how the brain is engaged–as a significant contributor, especially after the onset of language.

      Am currently reading Damasio’s Self Comes to Mind and find it fascinating.

      1. If evolution is a pointer here, it is not the stored information amount or specific details what matters in building and using brains, it is the usage; use it or loose it, as in the genome.

        So the question is perhaps why we use more neurons than similar massed hominids, say chimps.

        To play the joker here, that could as well be the result of some inadvertent inefficiency as that we really produce much more or intricate function. Off hand, I would day that the former is _much_ more likely than the later, all it takes is some enzyme system going wrong to mess up a complicated organ/cells. Maybe we are simply “lazy-brains”.

        That would make hunting and agriculture inventions to feed a power hungry “design mistake”; “just so”. 😮

        1. Thanks. As a border-scientist, a technologist, I am terminology challenged. For me, the term evolution usually requires an adjective. In this case I did not distinguish between mind (consciousness) and brain (biology). While I don’t see them as two elements I wonder as to their interrelated development. What does DNA bring to the newborn as “initial conditions” to use an old analog computer term. What evolutionary effects are afforded by vigorous mental activity in a lifetime, especially pre-propagation? Will the current inquiries into neuro-plasticity reveal new insights into DNA evolution?

          Questions, questions… how do they influence our cognitive evolution?

          1. the initial conditions of newborn are neo-nate ignorance and deliberative capability

            deliberative capability is what made humans human – purely physical, machine-that-goes-by-itself, it is our ability for re-ification, for making abstract concrete

            once hominids became deliberatively capable their evolution turned into evolution of their deliberative capability which in turn lead to the rapid increase of brain size

            the advent of deliberative capability in the course of evolution put homo sapiens on top of the food chain and is behind ‘human condition’ that is behind _all_ things human including language capabilities and belief systems.

            humans are able to ask questions that do not make sense simply because they are posable – reflection of deliberative capability at work.

            1. Thanks. As an interested bystander, I need to study this. In Damasio’s book, Self comes to mind, his interest is in whether there is a “self”, which he is convinced there is, what constitutes same. It’s a slow read for me. I’ve been working on it for several days and am up to page 85. Starting at page 80 he describes the differences between “vegetative” children and those born without a cerebral cortex. I’m out of my league here and offer nothing but a reference for any who might be interested. Thanks again for your insight. I’ll print it and put it in my study notes.

          2. You are welcome.

            Maybe the mind-to-brain mapping is outside the purpose of this article. But to reply, I didn’t realize that you were interested in the former.

            As for the biology, I’m sure you can come up with (putative, seeing the article) correlations between human brains and genetics. The problem then is to test them for correlation, and then test them for for causality as per usual (check by varying parameters).

            Personally I take a very dim view on human “exceptionality”. Models have shown that the neocortex may work symbolically at its base (and I have given that reference soo many times :-)), so the brain is implicitly “pre-wired” for explicit symbolic thinking. That it is an easy map is seen by that animals can do it.

            I don’t see what language (say) would contribute, except a facilitation of observing said symbolic thinking. Language may be a specific human culture invention, but I’m sure you can find such specifics in other animals cultures.

            What makes humans “exceptional”? As opposed to some other commenters here, I don’t see it. Success don’t need to mean exception. (And in fact I understand that the main difference to Neanderthals may have been fecundity.)

            OK, I’ve heard of one thing that humans may do much better than other animals: store fat to excess. That may also help success under some conditions (not seen today, at a guess). Yippee us, the fat animals! 🙁

            1. Thanks. Are you familiar with Damasio’s, Self Comes to Mind? If so, what is your opinion and do you have a suggestion for another author?

              My interest is pedestrian. As a matter of reference I am interested in the basic function of the neuron alone and in groups. As a retired computer networking professional I see the brain as a partial mesh network with a connectivity density lower than any other network I know (It’s about two one millionths of one percent of the potential connections of a full mesh network.

              Your insights on “exceptionality” are noted. Thanks.

  7. Have you read Barbara Finlay’s work on this area? She’s written some great articles on brain evolution over the years. You could say that her take on human brain evolution is connected to heterochrony.

  8. So According to this transposable elements theory, when will the human brain evolve again? The graph demonstrates that the TEs in the brain multiply by a lot. Is the next stage of evolution for humans close?

Leave a Reply