The evolution of human intelligence

Social Competition May Be Reason For Bigger Brain:

"Our findings suggest brain size increases the most in areas with larger populations and this almost certainly increased the intensity of social competition," said David Geary, Curator's Professor and Thomas Jefferson Professor of Psychosocial Sciences in the MU College of Arts and Science. "When humans had to compete for necessities and social status, which allowed better access to these necessities, bigger brains provided an advantage."

The researchers also found some credibility to the climate-change hypothesis, which assumes that global climate change and migrations away from the equator resulted in humans becoming better at coping with climate change. But the importance of coping with climate was much smaller than the importance of coping with other people.

At the link above it says that the paper is in press in Human Nature. Actually I think they must be looking at an old press release, because the description of the paper matches perfectly this paper published in January, Testing Climatic, Ecological, and Social Competition Models:

Hypotheses regarding the selective pressures driving the threefold increase in the size of the hominid brain since Homo habilis include climatic conditions, ecological demands, and social competition. We provide a multivariate analysis that enables the simultaneous assessment of variables representing each of these potential selective forces. Data were collated for latitude, prevalence of harmful parasites, mean annual temperature, and variation in annual temperature for the location of 175 hominid crania dating from 1.9 million to 10 thousand years ago. We also included a proxy for population density and two indexes of paleoclimatic variability for the time at which each cranium was discovered. Results revealed independent contributions of population density, variation in paleoclimate, and temperature variation to the prediction of change in hominid cranial capacity (CC). Although the effects of paleoclimatic variability and temperature variation provide support for climatic hypotheses, the proxy for population density predicted more unique variance in CC than all other variables. The pattern suggests multiple pressures drove hominid brain evolution and that the core selective force was social competition.

Table 1 has most of the important stats:

i-78cef76569eb8c93bf647230d8404cb4-geary.jpg

The text is clearer:

Population density predicts substantially more unique variance in CC (41% of the total variance in CC was unique to the population density variables) than parasite load (1%), temperature variation (7%), and δ18O variation (9%) and yields much larger unstandardized regression weights than the other variables. A 1 SD change in population density is associated with several times more variance in CC than a 1 SD change in parasite prevalence or paleoclimatic variation. The quadratic components of δ18O variation and temperature variation predict significant unique variance in CC (Bonferroni adjusted alpha=0.007), but the linear components do not, indicating inverted U-shaped relationships between δ18O variation changes (independently of population density, temperature variation, or parasite load changes) and CC changes, and between temperature variation changes (independently of population density, δ18O, or parasite load changes) and CC changes. Latitude does not predict unique variance in CC when added to this model (t166 =1.52, p>0.10).

CC = "cranial capacity." Their sample consisted of ~150 skulls from a range of fossils, so it isn't as if they had the original brains on hand. δ18O is an isotope that's used as a proxy for paleoclimate variation. Figure 2 shows the effect of the independent variables graphically:

i-b628cf476612cb424256fe05e4c553de-geary2.jpg

The presumption here is that large skulls mean there must be larger brains within those skulls. As I've noted before not only is our large brain relatively weird for a mammal our size, but they're serviced by ~20% of our metabolism, as opposed to 10% for a a typical mammal. Our nearest extant relatives, the chimpanzees are closer to 10% than 20. Since from what I have read and seen we're the most intelligent mammals as well it seems likely that our cognitive abilities had some adaptive benefit to justify the proximate anatomical changes whereby we have large brains, but small guts. As mentioned elsewhere the undersized digestive system means that humans had to procure more high quality food. That procurement was presumably made easy by our genius at social organization enabled by the faculties of our mind (language, gesture, etc.).

As noted in this paper one of the peculiarities of the human line is that our cranial capacities have been increasing steadily for about ~2 million years, until they maximized 100-50,000 years ago. In fact, the human lineage with the largest cranial capacity were Neandertals. The steady secular increase suggests a persistent selective pressure, an arms race.* Many paleoathropologists suggest that that pressure were human social networks, a view elucidated in Robin Dunbar's book Grooming, Gossip and the Evolution of Language. Dunbar shows that the average size of a social group in a species has a direct correlation with cranial capacity. Small brained monkeys have relatively simple social structures. Chimpanzees have larger social structures, and larger brains. And humans of course.... Dunbar's data set includes many more species, so the trend line is a bit more robust that I'm putting down in words.

Unfortunately we don't have fossils of social groups, so the theory must go based on inferences. The interesting point is that cranial capacity kept increasing even when tool technology was extremely stagnant, and human artistic creativity seems to be a feature of the last several hundred thousand years, being generous with interpretations of finds. The lack of material evidence leads many anthropologists to conclude that there must have been selective pressures which were not preserved; our societies & cultures & language.

Humans do have a wide range of cognitive competencies which are striking and exceptional. Of course we can do algebra, but the reality is that we're not very good at it. Rather, what humans are good at is modeling social relations implicitly. As an analogy, if someone throws down 4 marbles you recognize 4 marbles without having to count. If someone throws down 40 marbles you haven to count to get a precise number. Counting is a cultural tool applying our general intelligence. It's slow and clumsy, but it scales. In contrast our innate ability to recognize low number counts is fast and does not depend on culture (most humans lose accuracy in terms of gestalt number recognition beyond 6). Socially autistics are like those who have to count for even 4 marbles (there are individuals with brain damage who in fact have lost innate numeracy, though their cognitive function may otherwise be typical. See The Number Sense). Compared to us most animals who we reckon to be rather bright are absolutely incompetent. Chimpanzees show some glimmers of social fluency, but lack deep abilities engage in modeling nested layers of intentionality (e.g., "I know you know I know you know I know....").

The inference evolutionary anthropologists make from our relatively universal high level of social intelligence is that strong selective pressures have fixed a high minimal level of competency. The same applies to language. There is variance in social and linguistic fluency, but there is a clear difference between those who lack competency and so are defined as having a pathology. If you can't learn language or social relations you aren't just at a tail of the natural distribution of the trait, you're lacking a universal human faculty. But while those who lack language are evidently impaired, the tendency to naturally projected our own cognitive default states to others leaves the autistic or social challenged with less sympathy. Though the ability to learn language is innate, our expression exhibits a great deal of conscious awareness. On the other hand social intelligence operates under the hood to such an extent that I believe most humans are not consciously aware of the complex cognitive operations which they are engaging in as a matter of course. Naturally when you encounter individuals who lack normal cognitive capacities in this domain one might impute upon them rude or disrespectful intent, because disrespect would be the only reason that a normal person would behave in such a manner.

Other explanations for encephalization have been offered. The neurobiologist William Calvin has suggested that manipulation of tools required greater visuo-spatial and coordination capabilities. But in general it seems to me that the social explanation has power because in terms of richness even hunter-gatherers lived in a very "thick" world socially and culturally, though they might be relatively lacking in material goods. Robin Dunbar suggests that the ability to hold distinct individuals in one's mind goes no further than 200. He uses this result to show how it maps onto the fissiporous tendencies of sects like the Hutterites. But of course human societies are much more complex today, and have been for thousands of years, than bands of 200. Additionally, the rise of task specialization and shifts that accompanied agriculture entails diversity of interaction across these 200 and added complexity of the resultant networks.

The anthropologists have offered some intriguing theories and data. Now it's time for the neuroscientists to start digging into the details of how the human mind operates, and if all that expensive brain tissue is being used the way we think it is being used. Is there is some biophysical reason why humans can't model more than 200 individuals?

Cite: Hominid Brain Evolution Testing Climatic, Ecological, and Social Competition Models, Drew H. Bailey & David C. Geary, Hum Nat (2009) 20:67-79, DOI 10.1007/s12110-008-9054-0

* The fossil data being what it is I have read that there are actually periodic "bursts" as opposed to a gradual rise. I don't think we have granularity to establish either, but for this theory I don't think that that matters.

H/T Kambiz.

More like this

Your BS detector should be set higher. How exactly did they estimate "population density" for the past 2 million years?

We assume a close relationship between population density and migration patternsâ
this is apparent in the fossil record (see Results)âand thus created a population
density proxy using the number of individuals living in surrounding areas during the
hominidâs ancestral history. The population density score for cranium j is represented
as the number of crania dated from the same date or before j, outside a radius of 30°
(in any direction) of Al Mashriq, Ethiopia.

In other words, the spatial extent of humanity is their measure of "population density." Of course, spatial extent is autocorrelated with time. As is cranial capacity. QED

D'oh!

All their analysis is showing is that climate doesn't explain human colonization.

cranial capacity and intelligence - i think this is a weak proxy for intelligence. between species, yeah, you have differences that correlate w/ intelligence, but you also have species with much larger brains than us. within our species, the cranial capacity -> intelligence idea is fraught with problems as well. i don't think that it's possible to start making reliable evolutionary conjectures with any of this. eventually we'll have a neuroscientific understanding of intelligence, and maybe we can look precisely at that to see how its evolved.

population density and intelligence - so, they're correlated. what is the evidence that the former is causing the latter?

encephalization and intelligence - also correlated, but i'm yet to see evidence for a causal arrow. a multivariate genetic analysis could possibly answer whether the two share genes, but i think we'd still be at the point where we're unable to produce reliable evolutionary theories from it.

Given that cranial capacity and brain size isn't a perfect correlation (the recent Sibyrhynchus fossil comes to mind for example) and given that brain size isn't always correlated with intelligence this seems questionable.

Since only one extant species seems to have evolved to the high levels of intelligence that humans have, it really is very hard to do much more than speculate about what caused human intelligence.

One hypothesis that I like a little bit (and unfortunately forget who proposed it) is that large brains developed specifically for the ability to throw objects effectively. From a math perspective, this is a good idea because the calculations that humans do when they throw or catch irregular objects are really quite complicated.

Also, an interesting note: Some birds and apes have larger counting numbers than most humans. There was a study recently showing that the average chimp counting number was around 8.

The throwing objects hypothesis isn't too strong. Most of the really complex calculations done for sensorymotor tasks occur in the Cerebellum. Compared to other species, it actually takes up a much smaller percentage of brain matter. It is in the neocortex where human brains are substantially larger, and it is here where social interaction is processed.

By Stefan Krueger (not verified) on 05 Jul 2009 #permalink