News and Views

Evolution: Eating away at the social brain

Primates, especially humans, have large brains and this is thought to reflect our level of cognitive complexity or ‘intelligence’. Could this all be down to what we eat?

Humans exist in a complex social world, far more complex than any other primate, as the late American singer–songwriter Lou Reed describes in just two lines of his song ‘New York Telephone Conversation’: “Did you see what she did to him, did you hear what they said. Just a New York conversation, rattling in my head.” Humans have developed such extraordinary cognitive complexity that we even have the intelligence to write songs about the familiar situation where two people are talking about what numerous other people have said and done, using a device specifically invented to communicate with people who are far away.

There is a long-standing notion in evolutionary biology that social complexity is linked to cognitive complexity. In the mid-1990s, drawing inspiration from earlier work on the subject, this idea was solidified and has become known as the social brain hypothesis (SBH)1,2. This controversial hypothesis asserts that social complexity is the predominant evolutionary force driving intelligence. Its supporters argue that it can explain the purported elevated level of cognitive complexity observed in primates and that the exceptional evolutionary pressures associated with human sociality eventually resulted in our unique intelligence and astonishingly large brains. Writing in Nature Ecology & Evolution, DeCasien et al.3 test SBH using the largest dataset ever brought to bear on this question, analysing more than three times the number of primate species than have previously been used. After accounting for species body size, they find no support for SBH.

Instead they find that what a species eats predicts its brain size — specifically that species whose diet is predominantly made up of fruit have, on average, larger brains than those that specialize on eating leaves (Fig. 1). The idea that the diet of a species is important is by no means a new one and is likely to be associated with spatial and/or temporal memory demands associated with foraging. Diet is an important component of what is often referred to as the ecological or foraging hypothesis4 — an alternative explanation for the evolution of cognitive complexity — which researchers have found evidence for time and time again5,6. But the study by DeCasien et al. is novel in that it simultaneously tests between social and ecological hypotheses across a large number of primate species.

Figure 1: Rhesus macaque (Macaca mulatta), a primarily frugivorous Old World primate (top) and mantled howler monkey (Alouatta palliata), a New World folivore (bottom).
Figure 1

Photo credits: Mauritius images GmbH / Alamy Stock Photo (top), and Diana Rebman / Alamy Stock Photo (bottom).

At first glance, this study would seem to put an end to SBH. However, I doubt that this will be the last word on the matter. DeCasien et al. should be applauded for the construction of their dataset and their methodology, which accounts for phylogenetic uncertainty and interspecific variation in brain size. I feel confident that their study will refocus and reinvigorate research seeking to explain cognitive complexity in primates and other mammals. But many questions remain.

Like many before them, this study assumes that whole brain size is a good proxy for cognitive complexity. While our intuition might be satisfied with this assumption, evidence shows that the brain is a complex compartmentalized organ, and that these compartments evolve in a mosaic fashion7 with natural selection acting to change some parts while others remain unaltered. So then, does the evolutionary signature of the social brain reside in one of these compartments? The main contender here is likely to be the neocortex. This region of the brain is considered key in facilitating complex cognitive tasks and has been described as the “crowning achievement of evolution and the biological substrate of human mental prowess”8. The problem is that we don't have enough data on primate neocortex size to test the idea that it is related to social complexity on anything like the same scale as the whole brain — this is something DeCasien et al. recognize. With this in mind, can our last goodbye to the social brain really come until we rectify this situation?

DeCasien et al. employ social group size as their measure of social complexity, which, although widely used, is by no means universally accepted as a good proxy. Firstly, it is very variable within a species. For example, a recent study shows that chimpanzees (Pan troglodytes) live in groups that range in size from 21 to 187 individuals9. In addition to its lability some criticize the use of social group size owing to the fact that that it does not inform about interactions within the group. It has been noted recently that an objective and universal measure of social complexity is lacking — but the number of distinct relationships an individual of a group has is likely to be important10.

So where does this leave us in our quest to understand the evolutionary forces driving cognitive complexity or intelligence? DeCasien et al. may have delivered a blow to SBH that has it reeling, and if future work irons out some of the remaining methodological creases it may be down and out. Then we will be left in the extraordinary position of trying to explain primate cognition without sociality. But surely diet cannot be the whole story.


  1. 1.

    & in Machiavellian Intelligence II: Extensions and Evaluations (eds Whiten, A. & Byrne, R. W.) Ch. 9, 240–263 (1997).

  2. 2.

    Evol. Anthropol. 6, 178–190 (1998).

  3. 3.

    , & Nat. Ecol. Evol. 1, 0112 (2017).

  4. 4.

    Evol. Anthropol. 15, 224–236 (2006).

  5. 5.

    & J. Zool. 190, 309–323 (1980).

  6. 6.

    & Science 244, 1589–1594 (1989).

  7. 7.

    & Nature 405, 1055–1058 (2000).

  8. 8.

    Nat. Rev. Neurosci. 10, 724–735 (2009).

  9. 9.

    , , & Int. J. Primatol. 35, 990–1003 (2014).

  10. 10.

    & Anim. Behav. 103, 203–209 (2015).

Download references

Author information


  1. Chris Venditti is in the School of Biological Sciences, University of Reading, Reading RG6 6AS, UK.

    • Chris Venditti


  1. Search for Chris Venditti in:

Competing interests

The author declares no competing financial interests.

Corresponding author

Correspondence to Chris Venditti.