Behold! 108 animal species (mostly vertebrates (mostly mammals)), plotted on a doubly logarithmic graph by brain and body mass. (Brain mass is represented in calculations as E, body mass as S, both in grams.)
It's tempting to use brain size as a proxy of animal intelligence, but it's also obvious that simply weighting a brain is not enough. Just look at the leftmost ranking in the blue box: alright, cetaceans and elephants as the top, as expected, but does anyone believe sperm whales to be six times smarter than humans? And what is the chimpanzee doing below the walrus, the camel, and the cow? We know parrots and crows can be pretty damn smart, compared to birds and even most mammals - and yet they're much less brainy than ostriches and chickens.
OK, this isn't exactly a riddle for the ages. It's obvious that a larger animal needs a larger brain to perform the same functions. A 120-kg ostrich with a 6-gram brain wouldn't be as smart as a crow with the same E, and 2 grams of brain are much more active in a 250-gram pigeon than in a 56-tons Brachiosaurus
So we only have to look at the relative
brain size! We divide E by S, and... huh. Now the top of the list (middle column in the blue box) is dominated by tiny birds. The brain of a humble sparrow takes up 8% of its body weight - the equivalent for a 70 kg human would be 3 kg of brain! Homo erectus
ranks barely above the guinea pig, the gorilla is just above a slug, and the chimpanzee is below an ant and a goldfish! And of course this time elephants and whales rank pathetically low.
Why does this happen? Well, it turns out that while a larger brain is required for a larger body, it doesn't quite scale up linearly. The specter of the square-cube law
haunts everything. The scale of body structures and processes is not directly proportional to the overall mass; for example, an animal metabolic rate is only proportional
to the 3/4th power of its mass, so that if one animalis 10,000 heavier than another, everything else being equal, it only needs 1000 times as much food. The legs of an elephant are obviously larger than those of an ant, not only in absolute shape, but in proportions, too - stout pillars instead of thin wires. The eyes of a sperm whale are larger than those of a dormouse, but they're proportionally much smaller in its face. This is called allometry
And here comes the encephalization quotient
(EQ). When you start studying the allometry of the brain
(see for example Jerison, 1973 in the sources below), you find out that it grows with the 2/3th power of the body mass just, presumably, to perform the same functions. Thus, to quantify the braininess of an animal, you do not divide E by S, but E by S^(2/3). And then we divide it by 0.12, so that the typical value for mammals is about 1. The result is the formula you can see in the top left corner of the graph.
Now we're getting somewhere! As you can see yourself, the data points form a belt that runs through the graph in a slope corresponding to our formula. Primates and cetaceans rule the upper part of the new ranking (rightmost column in the blue box), followed by elephants, crows, and parrots. As they damn well should. Most mammals are indeed clustered around EQ = 1, as are birds, while other vertebrates and invertebrates are found below. (Sorry, Brachiosaurus
: you're terrible either way.)
Now look at the diagonal lines crossing the graph. The red ones represent constant brain/body ratios. For example, the red line that starts in the lower left corner of the graph shows all the points where E/S = 1/100, or where the brain mass is exactly 1% of the body mass. (The highest red line shows were E/S = 1, or brain mass is exactly equal to total body mass, so obviously the region above is impossible to fill.) The blue lines mark constant EQs: for example, the EQ = 1 line cutting throughthe bulk of birds and mammals. If you go from the sparrow to the parrot, you can see you're going farther above the lower blue line (EQ rises), but away from the upper red line (E/S falls).
A few observations:1.
There's a rather narrow belt of EQs, say between 0.02 and 0.1, that seems to be typical for the animal kingdom in general; arthropods, mollusks, ectotherm vertebrates, all lie right on it. Mammals and birds, on the other hand, are overwhelmingly located on a parallel but higher line, at EQ 0.2-1.5 or so, with only a few specialized groups rising above. The large brain in birds and mammals has probably evolved separately, as both stem-mammals (e.g. Thrinaxodon
) and non-avian dinosaurs (e.g. Allosaurus
) score pretty low. Whatever common factor they have (probably endothermy - i.e. the ability to keep inner temperature constant - allowing the brain to specialize its functions for a particular temperature, rather than having to make do in different conditions?), it multiplied their brain size but left the allometric relation relatively unchanged.
(At one point, I dared estimate E and S for the flatworm Dugesia japonica
; extrapolating from here
, and assuming the flatworm to be a 0.7*0.2*0.05 mm block of jelly with the density of water, I put E = 0.0000008 g and S = 0.000007 g, which of course doesn't it to the graph, but gives E/S = 1:9 and EQ = 0.018, in the correct order of magnitude at least.)2.
Some EQs are genuinely surprising: I would have expected pigs (which are below cows!) to score much higher. The disparity between the two elephant species is strange, as is that between us and Neanderthals (though that's probably to due to issues with body weight: see below). On the other hand, pinnipeds do pretty well for themselves, with a sea lion and a walrus rivalling primates. The manta ray and the hammerhead shark score around 0.3, bizarrely high compared to other fish. And what the heck is the sun bear doing in the middle of apes??? Also, given the trend of discoveries about dinosaurs being much more active and complexly behaving than once believed, it feels downright unfair for them to score so low. Even Stenonychosaurus
(once known as Troodon
) is only slightly above the chicken.3.
We can attempt a few general statements about the high-EQ species. Omnivores like crows and primates, and also the goat and the raccoon, score the highest, followed by carnivores like pinnipeds and cetaceans. Ground hornbills, like primates, are omnivores that search food from many different sources. Herbivores tend to be lower than both, with the conspicuous exception of parrots (which eat high-calory fruit and nuts) and elephants (which are just weird all around). Carnivorans as a bloc are solidly above ungulates. Note that this mostly applies to active hunters: passive predators like frogs and spiders not as much (but is that just an effect of endotherm passive predators being rare in general?). Primates and crows are also enthusiastic tool users, and parrots and raccoons are at least able to crudely manipulate objects (pinnipeds and cetaceans, on the other hand...). Most of these species - apes and elephants and dolphins - are also strongly social; parrots and raccoons, not as much. (Maybe octopodes would score higher if they didn't die before meeting their own offspring.) In sum, all these features seem strongly correlated with high EQ and intelligence, but they all have obvious exceptions. Perhaps our secret was having all three at the same time.4.
At risk of being accused of teleology, I'll say there are clear trends of EQ increasing over time. In this graph I focused on living species, with only a few extinct token ones; but Jerison's book, in source i, has a pretty extensive list of brain data for extinct mammals, and it seems clear that Holocene mammals lie on a line above Oligocene mammals, which are above Palaeocene mammals, which are above stem-mammals (also see the graphs in source s). Jerison's examples of extinct fish (e.g. the placoderm Macropetalichthys
) never rise above 0.1, which both cartilaginous and bony fish surpass today. Same with dinosaurs: as I mentioned above, the highest-scoring non-avian dinosaurs are right besides the lowest-scoring birds.
(In the case you're wondering, brain size for extinct species is mostly estimated from endocasts, solid casts of the internal cavity of the skull.)
Before we get too excited, though, here's a few important caveats.
First, and most obvious, the very concept of "brain" is not universal in animals. EQ was invented for vertebrates in general, and mammals in particular; most animal phyla don't even have brain-analogue structures. Arthropods and cephalopods have clusters of nerve tissue that can be called "brains", true; but their nervous systems tend to be much less centralized than that of vertebrates. The nerve cells that control the motion of our legs is still part of our brain; in a grasshopper, their equivalents would be located in ganglia in the thorax. Two thirds
of the neurons of an octopus are in its arms; only a minority of nervous activity occurs in the "brain". (It would be so interesting to find out how a creature such as an octopus think; would it see itself as a group of mostly independent beings that happen to be joined together?) Another clear hurdle is extinct species, where we largely have to guess what fraction of the endocast volume was filled by the brain. What counts as the body is also not trivial. If I became fatter, my body weight would increase but my brain weight wouldn't; so my EQ would drop. But would that change anything about my cognition?
reminded me that bird brains have a much higher neuron density than mammal brains, so they might actually deserve a higher ranking. Olkowicz et al. 2015
suggests that a bird brain has about 2.3 times as many neurons as a mammal brain of equal size. If we thus count bird brains as 2.3 their actual mass, the New Caledonian crow has E = 14 g, S = 280 g, E/S = 1:19, EQ = 2.9 (!!!); and the African grey parrot has E = 21 g, S = 410 g, E/S = 1:19, EQ = 3.2 (!!!!), ranking both between the chimpanzee and the porpoise. While this fits with studies of their intelligence, I'm not sure how sound this methodology is.
There's also the fact that most of these measurements are based on a few specimens, or even just one, which do not have to be representative of their species. Often, as in the gorilla and the walrus, brain and body size change enormously between males and females, and young organisms tend to have much higher EQs than adults. The sea lion from source o is a 1-year old female weighing 34.5 kg, much less than adult of its species. Regarding the discrepancy between the two elephants, source i notes that the Indian elephant may have been young. The gap between modern humans and Neandertals could be explained by data selection; the sources f and i both give the average weight of H. sapiens
at 53.5 kg, much lower than Wikipedia's 68 kg, but I followed their numbers. (If you're interested, E = 1.35 kg and S = 68 kg gives E/S = 1:50 and EQ = 6.8.)
The sun bear from source t is especially problematic. The authors apparently received only the brain and could not examine the animal, which was reported as weighing 82 kg at death; but they note this is an astonishingly high weight for that species. Source i gives an even higher brain weight (390 vs. 300 g) for a more typical body weight of 45 kg, so the authors of t base their calculations on the latter number, noting that the bear was raised in captivity, and speculating it may have been overfed. If S = 82 kg, E/S = 1:86, and EQ = 1.3.3.
And then, of course, not all of the volume of a brain is given to what we'd call "intelligence". In at least some cases, most of a brain is used to process a large amount of raw sensorial data: this is definitely the case for the elephant-nose fish, and I would bet for the hammerhead shark (the former's "nose" and the latter's "hammer" essentially being sensorial antennae).4.
While EQ measuring seems pretty solid at middle scales, with the caveats above, I suspect it might break down altogether at the extremes. A brain can only get so small before it stops functioning as a brain altogether, which could make the EQ of insects and spiders artificially high. More importantly, when a brain reaches a certain size, expanding it might be useless even as the body grows. Perhaps the brain of a sauropod already does anything the brain of an ectotherm (???) could possibly do, and a larger one wouldn't help even in coordinating a larger body (since the speed of nerve impulses is actually pretty slow
, such a being would benefit more by decentralized nerve control). This could explain why both ectotherms (with large dinosaurs) and endotherms (with whales) seem to reach an EQ "roof" above which the body grows without corresponding brain expansion.
Thoughts?Sources for brain and body data:a.
Jerison HJ (2004). Dinosaur Brains, in Encyclopedia of Neuroscience
(3rd ed.), Elsevier Science (link
) (related data in source i): Allosaurus, Brachiosaurus
(body mass from source v), Iguanodon, Triceratops, Tyrannosaurusb.
Ari C (2011). Encephalization and Brain Organization of Mobulid Rays (Myliobatiformes, Elasmobranchii) with Ecological Perspectives, The Open Anatomy Journal
, 6(3):1-13 (link
): Manta, Rhincodon, Sphyrnac.
“Brain and Body Size” (2000), Serendip Studio, Bryn Mawr College (link
Mlikovsky J (2003). Brain size and foramen magnum area in crows and allies (Aves: Corvidae), Acta Societatis Zoologicae Bohemicae
, 67:203-211 (link
): Corvus, Picae.
Linzey DW (2012), Vertebrate Biology
(2nd ed.), Johns Hopkins University Press (link
): Gallus, Lacerta, Perca, Triturus, Troglodytes, Varanus, Viperaf.
McHenry HM (2009). Human Evolution, in Evolution: The First Four Billion Years
, Harvard University Press (link
): Australopithecus, H. erectus, H. neanderthalensisg.
Nilsson GE (1996). Brain and body oxygen requirements of Gnathonemus petersii
, a fish with an exceptionally large brain. Journal of Experimental Biology
, 199:603-607 (link
): Carassius, Gnathonemush.
Crile G, Quiring DP (1940). A record of the body weight and certain organ and gland weights of 3690 animals. Ohio Journal of Science
, 40(5):219-259 (link
): Alligator, Amazilia, Blaberus, Bos
(Jersey breed), Bradypus, Bubo, Bucorvus
(male only), Canis
(greyhound), Capra, Castor, Cavia, Clarias, Columba, Desmodus, Didelphis, Felis, Giraffa
(male only), Gymnothorax, Hippopotamus, Larus
(female only), Lemur, Limax, Lithobates, Loxodonta, Melanoplus, Odobenus, Pan, Passer, Phrynosoma, Rattus, Salmo, Sorex, Struthio, Sus
(female only), Testudo, Thunnus
(averages from many males and females unless otherwise noted)i.
Jerison HJ (1973), Evolution of the Brain and Intelligence
, Academic Press (synapsids
): Ateles, Barylambda, Camelus, Cebus, Chlorocebus, Elephantulus, Elephas, Eryops, Gorilla, H. sapiens, Hyaenodon, Latimeria, Lystrosaurus, Macaca, Macropetalichthys, Megalonyx, Ornithorhynchus, Papio, Phenacodus, Pongo, Procyon, Rhamphorhynchus, Talpa, Thrinaxodon, Triconodon, Ursusj.
Marino L (2009). Brain size evolution, in Encyclopedia of Marine Mammals
, 149-152 (link
): Balaenoptera, Basilosaurus, Megaptera, Orcinus, Phocoena, Physeter, Trichechus, Tursiopsk.
Salas CA, Yopak KE, Lisney TJ, Potter IC, Collin SP (2017). The Central Nervous System of Jawless Vertebrates: Encephalization in Lampreys and Hagfishes. Brain, Behavior and Evolution
, 89(1), 33–47 (link
): Lampetra, Myxinel.
Seid MA, Castillo A, Wcislo WT (2011). The Allometry of Brain Miniaturization in Ants. Brain, Behavior and Evolution
, 77(1), 5–13 (link
): Camponotus, Paraponeram.
Iwaniuk AN, Dean KM, Nelson JE (2004). Interspecific Allometry of the Brain and Brain Regions in Parrots (Psittaciformes): Comparisons with Other Birds and Primates. Brain, Behavior and Evolution
, 65(1), 40–59 (link
Mares S, Ash L, Gronenberg W (2005). Brain Allometry in Bumblebee and Honey Bee Workers. Brain, Behavior and Evolution
, 66(1), 50–61 (link
Montie EW, Pussini N, Schneider GE, Battey TWK, Dennison S, Barakos J, Gulland F (2009). Neuroanatomy and Volumes of Brain Structures of a Live California Sea Lion (Zalophus californianus
) From Magnetic Resonance Images. The Anatomical Record
, 292:1524-1547 (link
(female, 1 year old)p.
Lu JS, Yue F, Liu X, Chen T, Zhuo M (2016). Characterization of the anterior cingulate cortex in adult tree shrew. Molecular Pain
, 12:1-11 (link
Packard A (1972). Cephalopods and fish: the Limits of Convergence. Biological Reviews
, 47(2):241-307 (link
): Loligo, Octopus, Sepiar.
"Dichotimistic" (2007), John McCrone (link
Hopson JA (1977). Relative Brain Size and Behavior in Archosaurian Reptiles. Annual Review of Ecology and Systematics
, 8(1), 429–448 (link
): Archaeopteryx, Stenonychosaurust.
Kamtya T, Pirlot P (2009). The brain of the Malayan bear (Helarctos malayanus
), Journal of Zoological Systematics and Evolutionary Research
, 26(3), 225–235 (link
(male, but see "caveat" 2 above)u.
Hill DE (2010). Jumping Spider feet (Araneae: Salticidae). Peckhamia
, 85(1):1-48 (link
): see belowv.
Benson RBJ, Campione NE, Carrano MT, Mannion PD, Sullivan C, Upchurch P, Evans DC (2014). Rates of Dinosaur Body Mass Evolution Indicate 170 Million Years of Sustained Ecological Innovation on the Avian Stem Lineage, PLOS
(body mass only)
(* source r, whose reliability I'm not quite confident of, gives the size of Portia
's brain as 60% of that of a honeybee, which I took from source n. I estimated the overall body weight by interpolating those of related spiders Phidippus
, as given in source u, for Portia
's size.)Sources for animal silhouettesThese ones were taken from the website PhyloPic, which credited them to the following authors: Becky Barnes: Sorex; Timothy J. Bartley: Perca (with NOAA Great Lakes Environmental Research Laboratory); Dmitry Bogdanov: Rhamphorhynchus; Andrew Butko: Passer; Anthony Caravaggi: Troglodytes; Dori & Nevit Dilmen: Columba; Rebecca Groom: Larus; Scott Hartman: Allosaurus, Archaeopteryx, Brachiosaurus; Jaime Headden: Iguanodon; Tracy A. Heath: Helarctos; Robert Bruce Horsfall: Hyaenodon; Chris huh: Balaenoptera, Megaptera, Orcinus, Phocoena, Tursiops; Stuart Humphries: Thunnus; Maija Karala: Gnathonemus, Latimeria, Rattus; T. Michael Keesey: Myxine (with A. H. Baldwin), Corvus (with Bc999), Gorilla (with Colin M. L. Burnett), Eryops (with Dmitry Bogdanov), Loligo (with Hans Hillewaert), Pan (with Tony Hisgett), Rhincodon (with Scarlet23), Loxodonta (with J. A. Venter, H. H. T. Prins, D. A. Balfour, R. Slotow), Basilosaurus, Elephas, H. erectus, H. neanderthalensis, H. sapiens, Pica, Stenonychosaurus; Birgit Lang: Talpa; Lukasiniho: Bubo, Struthio; Mattia Menchetti: Apis; Gareth Monger: Pongo; Gustav Muetzel: Lacerta; Gordon E. Robertson: Bucorvus; Noah Schlottman: Physeter; Roberto Diaz Sibaja: Desmodus, Lemur; Vince Smith: Trichechus; David Tana: Triceratops; Steven Traver: Bos, Camelus, Capra, Felis, Gallus, Giraffa, Hippopotamus, Lystrosaurus, Odobenus, Procyon, Sus, Triturus, Varanus; Sarah Werning: Bradypus, Cebus, Didelphis, Macropus, Ornithorhynchus; Emily Willoughby: Tyrannosaurus; Yan Wong: Ateles; (uncredited): Castor, Macaca, Paraponera, Sphyrna
These ones I made from the following Wikimedia files: commons.wikimedia.org/wiki/Fil… (Alligator, commons.wikimedia.org/wiki/Fil… (Amazilia), commons.wikimedia.org/wiki/Fil… (Australopithecus), commons.wikimedia.org/wiki/Fil… (Barylambda), commons.wikimedia.org/wiki/Fil… (Blaberus), commons.wikimedia.org/wiki/Fil… (Camponotus), commons.wikimedia.org/wiki/Fil… (Carassius), commons.wikimedia.org/wiki/Fil… (Chlorocebus), commons.wikimedia.org/wiki/Fil… (Clarias), commons.wikimedia.org/wiki/Fil… (Elephantulus), commons.wikimedia.org/wiki/Fil… (Gymnothorax), commons.wikimedia.org/wiki/Fil… (Lampetra), commons.wikimedia.org/wiki/Fil… (Limax), commons.wikimedia.org/wiki/Fil… (Lithobates), commons.wikimedia.org/wiki/Fil… (Macropetalichthys*), commons.wikimedia.org/wiki/Fil… (Manta), commons.wikimedia.org/wiki/Fil… (Megalonyx), commons.wikimedia.org/wiki/Fil… (Melanoplus), commons.wikimedia.org/wiki/Fil… (Octopus), commons.wikimedia.org/wiki/Fil… (Papio), commons.wikimedia.org/wiki/Fil… (Phenacodus), commons.wikimedia.org/wiki/Fil… (Portia), commons.wikimedia.org/wiki/Fil… (Salmo), commons.wikimedia.org/wiki/Fil… (Sepia), commons.wikimedia.org/wiki/Fil… (Testudo), commons.wikimedia.org/wiki/Fil… (Thrinaxodon), commons.wikimedia.org/wiki/Fil… (Triconodon**), commons.wikimedia.org/wiki/Fil… (Tupaia), commons.wikimedia.org/wiki/Fil… (Ursus), commons.wikimedia.org/wiki/Fil… (Vipera), commons.wikimedia.org/wiki/Fil… (Zalophus)
(* actually the related Quasipetalichthys) (** actually the related Yanoconodon)