Feb 02 2007

Brain Size and Intelligence

One aspect of the Hobbit brain debate I did not discuss, but will now in response to Nathan’s question from yesterday’s post, is the relationship between brain size and intelligence. There are actually some complexities to this question, and I’ll try to break it down.

How do you measure “size?”
The two basic ways to measure brain size is volume and weight. Weight is a better measure than volume, because it better accounts for the density of neurons, but there is a fairly linear relationship between volume and weight, at least within a class of animals. (It is always more meaningful to compare brain size with other closely related animals) For fossil species we cannot directly measure weight, but we can measure volume, which is the reason brain “size” for fossil species is always given in volume (cubic centimeters or ml). For living species it is common to designate brain weight.

Cortical surface area
The best measure of brain computational power, however, is cortical surface area. The cortex is organized into layers, with neurons in the layers mapping to specific brain functions. One way to increase surface area is to increase volume, but another method is to fold the brain surface in on itself, forming hills (gyri) and valleys (sulci). This is indeed what happened evolutionarily in the more encephalized vertebrates, allowing them to pack more computational power into a smaller volume.

In fossil species we can infer the degree of cortical folding from endocasts – the impression that the brain left on the inside of the skull. But this method does not give nearly as much information as examining brain tissue can in living species can give.

Brain organization
But perhaps the ultimate determinant of brain power is not size or even surface area, but the microscopic organization of neurons (the cells that make electrical connections) and glia (support cells that also modulate neuronal function). In other words, the pattern and robustness of connections made between brain cells. Within a species this is likely to be the major cause of differences in intelligence. This also cannot yet be measured. We cannot look at the brain of a genius and tell it apart from that of someone with average or even far below average intelligence. A mathematical genius must have a pattern of hardwiring in the calculating part of their brain that functions better, more efficiently, more robustly, or perhaps with more sophisticated algorithms, than the corresponding brain area of someone who has little aptitude for math. Obviously, education plays a role, but there is no question that people are born with vastly different potentials for specific higher cognitive functions, such as math, music, language, art, etc.

Getting back to the Hobbit – it is now hypothesized by Falk and others that the LB1’s brain was organized better and more efficiently than other hominids, granting it superior intelligence for its brain size. This is inferred from the cortical pattern seen on the 3-dimensional reconstruction of the brain. But we cannot examine the cell structure and organization of the Hobbit’s brain, or that of any fossil species. So this hypothesis is still largely inferential.

Size Matters
Given the complexity described above, there is still a fairly consistent relationship in the animal kingdom between brain size and intelligence. Humans do, after all, have larger brain than our primate cousins. Dolphins, it is often pointed out, have even larger brains, but the increased size of the dolphin brain is due largely to an enlargement of the white matter portion of the brain (the part containing the wires, as apposed to the gray matter which contains the brain cells). It is thought that the increased wiring is responsible for the dolphin’s echolocation ability and functions to process sound into a three-dimensional mental image. So this is subcortical processing that does not make the dolphin overall more intelligent, but rather grants it the specific ability of echolocation.

Absolute brain size is not a good measure of intelligence, because larger animals require larger brains just to physically operate their bodies. So an elephant may be no smarter than a rat, even though their brains are much larger. The brain size to body size ratio is therefore more meaningful as a measure of “encephalization” of a species. However, this ratio is not linear – in other words an animal of the same degree of intelligence that is twice as massive as another animal will not have a brain that is twice as massive. Rather, it will be somewhat less than twice as massive. So biologists have come up with a brain size to body size ratio curve that represents the average for a type of animal. For example, there is a curve for all mammals, which is different that the curve for reptiles or birds. The degree of encephalization (called the encephalization quotient or EQ), therefore, is a measure of how far above or below the appropriate curve a specific species falls. The dolphin EQ, by the way, is smaller than that of Homo sapiens.

This holds for comparisons within a species as well. Bigger people will have bigger brains without being necessarily more intelligent. For this reason men have large brains than women (because men on average are bigger than women) and in the bad old days when sexism was endemic in the world of science this was offered (falsely) as evidence of male intellectual superiority. Pygmy species have smaller brains, but are usually on the curve with their larger cousins, and therefore have the same EQ. The Hobbit (Homo floresiensis) is very short statured. However, their brain size, or EQ, is smaller than would be a pygmy Homo sapiens of the same size. This is what led to the hypothesis that their brain organization may be better.

Although EQ is a meaningful measurement, it isn’t everything in brain evaluation. As stated above for dolphins, their EQ is high for a mammal partly because of their echolocation. Likewise EQ may not tell the whole story in Homo floresiensis. This is why it is more meaningful to make comparisons to close relatives, because brain organization is likely to more similar. Research in this area is moving toward more sophisticated methods for quantifying brain power, reflecting a deepening understanding of brain organization and function and improving tools for measuring it.

One response so far