Fun with hominin brain size as a percentage of body mass

Posted 9 October 2006 by Nick Matzke

Several people have suggested that I factor out body size to produce a chart just showing the relative increase in brain size over time. This is not as simple to do as it sounds, because most of the fossil skulls are not found with bodies, and vice versa [1]. So even if I had the paper with the body size data (De Miguel and Henneberg (1999). "Variation in hominid body size estimates: Do we know how big our ancestors were?" Perspectives in Human Biology, 4(1), pp. 65-80), one could not just do a regression. So we have to improvise. I started with the regression lines of log (cranial capacity) and log (body mass) vs. time in Figure 1 of Henneberg and Miguel 2004. Using a highly scientific method (drawing straight horizontal and vertical lines on the chart in Powerpoint), I measured four points along each line and reconstructed the slope and intercept of each regression. One can then figure out the relationship between the two lines and how to estimate a body size figure based on brain size. Then I just calculated a body size for each cranial capacity measurement, and re-did the chart as a chart of brain mass as a percentage of (extrapolated) body mass: [2] Hominin brain mass as a percentage of body mass

Hominin brain mass as a percentage of body mass

The slope is of course somewhat flatter in this chart, which shows the effect of controlling for the increase in body size. Absolute brain mass approximately tripled in human evolution, but brain mass as a percentage of body mass appears to have doubled in human evolution. This should be about what everyone expected anyway, but it is nice to see it on the chart. Yes, I am aware there are various issues with this quick-and-dirty body size extrapolation, but this is just an attempt to give people an approximate idea of what the data look like with body size approximately factored out, which is what many requested. For real analysis, of course, go to the paleoanthropology literature. (Suggestions for improvement, data sources, etc. are of course welcome.) Notes: 1. Specifically, Henneberg and de Miguel (2004) say on p. 25: "In the entire literature there are only 45 specimens of individual hominins for whom both CC and body weight estimates are available (Henneberg 1998). Amongst those are only four pertaining to hominins dated before 1.5 ma." 2. According to various references, cranial capacity = brain mass * 1.14.

22 Comments

John Wilkins · 9 October 2006

I have seen some charts using encephalisation quotients that show how modern Homo is off the primate norm, as well as primates being off the mammal norm. Perhaps you can find these somewhere for comparison? Unfortunately I failed to keep the refs.

Steviepinhead · 9 October 2006

Cool, Nick.

Recognizing the limitations of the data, it's still interesting that you get a much cleaner straight line through the mass of these points--and less of the "three lines separated by jumps" effect...

J. G. Cox · 9 October 2006

Not that maximal rigor is needed to establish your basic point here, but nowadays such allometric comparisons are typically done by regressing the variable that you are interested in (brain size) on the confounding variable (body size) and then examining the residuals instead of looking purely at proportions. Brain size and intelligence is an interesting conundrum. There is a fairly strong positive relationship between the two among species, but not within species. The lack of intraspecific relationship was probably just an artifact of low statistical power, but who knows? My guess is that most intraspecific differences in intelligence (broadly defined) are simply organizational, but that once a good set of brain assembly rules are evolved, then larger brains that can exploit those rules better become adaptive. Brains get larger until some fitness peak, then stay mostly the same size until an even better set of assembly rules establishes itself in the species, then brains start getting bigger again, rinse, repeat. Admittedly, that scenario is totally speculative in an area in which I have done no research.

I have seen some charts using encephalization quotients that show how modern Homo is off the primate norm, as well as primates being off the mammal norm. Perhaps you can find these somewhere for comparison? Unfortunately I failed to keep the refs.

Not my field, but I do know that encephalization quotients can do some funny stuff. A couple of graduate students in my department recently dredged some up for use in a comparative phylogenetic analysis, and found some brilliant Siberian hamster species that popped out between chimps and humans (somewhat closer to humans, actually); that's a hamster I'd like to meet.

Nick (Matzke) · 9 October 2006

Hi John,

Here is an example of what you are thinking of:
http://brainmuseum.org/evolution/paleo/brnBodWt.html

If someone could find a database of vertebrate brain/body masses we could do such a plot and throw in the hominins for fun.

Patrick Caldon · 10 October 2006

If there was some way of highlighting the hominids from which the extrapolation was performed that would be useful. Maybe fill in the box on the full-data point and use an outline box for the extrapolated data.

J. G. Cox · 10 October 2006

I actually might just have access to just such a database, which I believe contains over 80 mammals (including that champion hamster) and 40+ birds. However, given that the paper for which it was constructed will not be published until December, I might not be able to get permission to distribute it. The authors would be happy to share, but it is likely that some journal copyright issues would come into play.

J. G. Cox · 10 October 2006

A reduced data set has been published and is available as a supplement online. The URL is below. You should be able to yank body and brain masses (not sure about volumes) from it quite easily. I had mis-remembered; the genius rodent was a lemming, not a hamster.

http://sapphire.indstate.edu/~jlesku/Lesku_et_al_2006b_appendix.pdf

Steviepinhead · 10 October 2006

that's a hamster I'd like to meet.

Which reminds me: Spread the word: End A War! Save A Gerbil!

Henry J · 10 October 2006

All this fuss about gerbils - what about hamsters, guinea pigs, ferrets, squirrels, capaburros, etc.? :)

Henry

(Btw, the blogs spellcheck doesn't have "capaburros" listed, unless I misspelled it.)

Andrea Bottaro · 10 October 2006

Mmmm... that critter is supposed to have a 9-gram brain in a 47-gram body. I wonder whether there was a typo somewhere, especially because the reference is second-hand: "Karmanova et al, 1979, cited in Elgar et al, 1990". From the pictures I found (e.g here), it doesn't look like a particularly giant-headed freak.

Steviepinhead · 10 October 2006

Capybara.

Of the various cute little (and, in the case of capybara, not so little) rodents that you list, gerbils are found in the Mideast--in fact, G. mesopotamiae, or Harrison's Gerbil (and, since the "A little irony" thread, which see, also known as "Noah's Gerbil") has a range confined to the immediate area of the conflict of controversy.

And that's why the fuss.

Henry J · 10 October 2006

The spell checker doesn't accept capybara, either.

Nick (Matzke) · 10 October 2006

If there was some way of highlighting the hominids from which the extrapolation was performed that would be useful. Maybe fill in the box on the full-data point and use an outline box for the extrapolated data.

As I attempted to make clear in the post, no specific hominid specimens were used for the extrapolation. I just used the two regression lines in this Figure right here to work out the generic "average" relationship. I actually think this is probably pretty accurate overall, but it only describes the average change in brain size/body size ratio over time -- so species-specific deviations from the general trend will be suppressed. This is probably only a noticable issue for the robust austropithecines, but to really assess this would require Real Research (which has probably already been done based on my picking through the literature, but what I am looking for is basic data for graphics).

Steviepinhead · 10 October 2006

Henry, I believe you about the spellchecker, already! :=>

I'm just saying that "capybara" is how that particular critter's name is spelled, whether one checker or another is or isn't with the non-everyday-rodent program.

And, while we're at it--and as long as Nick is letting us get away with it--here's Even More Capybara Fun:
http://www.rebsig.com/capybara/.

afarensis, FCD · 10 October 2006

Nick,
Along the lines of papers Wilkins was asking about you might check into:

Rilling, James Human and NonHuman Primate Brains: Are They Allometrically Scaled Versions of the Same Design - Evolutionary Anthropology 2006, 15:65-77

Finlay and Darlington 1995 Linked Regularities in the Development and Evolution of Mammalian Brains. Science 268:1578-1584

Barton and Harvey 2000 Mosaic Evolution of the Brain Structure in Mammals. Nature 405:1055-1058

Winter and Oxnard 2001 Evolutionary Radiations and Convergences in the Structural Organization of Mammalian Brains. Nature 409:710-714

Sorry I don't have links to the articles...

Henry J · 10 October 2006

Cute not-so-little guys, aren't they? :)
About the size of medium sized dogs, or thereabouts.

Stephen Erickson · 11 October 2006

Nick, for all the work you have put into these figures, are you considering submitting a review/"meta-analysis" article somewhere?

John · 11 October 2006

You might want to add the chimp and the gorilla--this would make it a nice, concise graph to challenge those who claim that there are no transitional fossils.

Nick (Matzke) · 11 October 2006

You might want to add the chimp and the gorilla---this would make it a nice, concise graph to challenge those who claim that there are no transitional fossils.

That would be fun, we could put modern apes on there. The oldest known fossil chimp is only 500,000 years old I think, just published a year or two ago. Forests are considered poor fossilization environments due to acid soil I think. That does raise the issue of how to add new data to the dataset in an organized, collaborative way -- perhaps a wiki or something.

Stephen Erickson · 11 October 2006

Brain size as percentage of body mass has dropped in the U.S. over the past half century . . . .

Michael Suttkus, II · 11 October 2006

My personal brain mass vs. Body mass has dropped over the past twenty years. Have I gotten stupider? :-)

Darth Robo · 12 October 2006

An interesting thing on TV the other day explained one possible explanation for human brain size. Many apes have a strong anchor point at the top of their cranium which helped give them a strong jaw I think. As humans, this is something we lost over time which allowed our heads (and brains) to become larger. I found this amusing, since creationists are always talking about "where does the new genetic 'information' come from?", when ironically, we might have actually lost it instead.