Publications

Linear, economical cellular scaling rules apply to primate brains

Herculano-Houzel S, Collins CE, Wong P and Kaas JH, Proc Natl Acad Sci USA (2007)

 

This was the first of many studies performed in colaboration with Jon Kaas and his team at Vanderbilt U., specially Christine Collins and Peiyan Wong. Our goal was to compare the cellular scaling rules that apply to rodent and primate brains. According to the literature, which always tacitly considered that all mammalian brains are made the same way (mixing widly different taxa in the same comparisons) and that, therefore, brain size is an informative parameter in cross-species comparisons, rodents and primates should share the same cellular scaling rules.

 

All brains are not created equal

A quick side-by-side comparison, however, strongly suggests that rodent and primate brains of the same size must not be equivalent. For instance, the behavioral complexity of the agouti is no match for the abilities of the owl monkey - even though the agouti brain is, of all things, slightly larger than the owl monkey's. Worse, still: a capybara's brain is a good match in size to a rhesus monkey's, and yet its repertoire (mostly wading in the water in search of food) is no match for the monkey's.

 

Linear cellular scaling rules

Indeed, in this study we found that while rodent brains increase in size very fast as a power function of their number of neurons, primate brains scale linearly: while a 10 x increase in the number of neurons in the rodent brain is accompanied by a 63 x increase in brain size, a primate brain that has 10 x more neurons than another is also only about 10 x larger than it. Given that neuronal densities do not covary with structure size in primates, it follows that the average neuronal size must remain relatively constant across primates species. Remarkably, we find that, in primates, the non-neuronal/neuronal cell ratio does not vary with brain size.

 

Implications for cognitive abilities

Primate brains, therefore, are built according to more economical cellular scaling rules than rodent brains, which allow the former animals to concentrate larger numbers of neurons than rodents in the same brain volume. In this way, the rhesus monkey, it turns out, has almost 3 x as many neurons as the capybara - despite their equivalent brain sizes. This is an example of how the more economical primate scaling rules might account for the superior cognitive abilities of these species compared with others of similar brain size.

The discrepancies in the scaling rules between rodents and primates show that not all mammalian brains are created equal. Brain size, therefore, should no longer be used indiscriminately as a proxy for number of neurons in a mammal's brain.

 

Implications for the human brain

Surprisingly, we found that, for the species examined here, body size and brain size were linearly related to each other. This linearity can be attributed to the absence of great apes from our sample. Remarkably, if we use the primate cellular scaling rules found here to predict body and brain size for a generic primate with about 100 billion neurons in the brain, we obtain numbers that are strikingly close to human values: a brain of 1.45 kg in a body of 73 kg. Since this strongly implies that the human brain is built according to the same scaling rules that apply to other primates, we next sought to determine the actual cellular composition of the human brain (Azevedo et al., 2009).

 

Original article:

Herculano-Houzel S, Collins CE, Wong P, Kaas H (2007) Cellular scaling rules for primate brains. Proc Natl Acad Sci USA 104, 3562-3567.