The ten-million-year explosion: Paleocognitive reconstructions of domain-general cognitive ability (G) in extinct primates

Abstract

The correlation between primate “Big G” scores and brain volume in 68 extant species was employed to estimate probable G values for an additional 68 extinct and 1 extant species with endocranial volume data, employing phylogenetic bracketing. Three different methods were used to generate bracketed estimates, which all showed high convergence. The average of these G estimates (for the extinct primates) coupled with the values from the extant species were found to correlate strongly with neurocognitive measures of both extant and extinct primate taxa, specifically Transfer Index scores (an indicator of cognitive flexibility) and the neuroanatomical covariance ratio (a measure of neural integration). Ancestral character reconstruction incorporating G values was made possible with a phylogenetic tree containing data on the relationships among extant and extinct primates. Negative correlations were found between G and branch length, indicating that higher-G species do not persist as long as lower-G ones, consistent with the presence of the grey-ceiling effect (brain mass negatively predicts maximum population growth rate, and therefore a heightened vulnerability to extinction). Cladogenesis rates were also positively associated with G. Both associations were robust to models that controlled for false positive rates. Comparative models revealed that G evolved in extinct and extant primates in a punctuated pattern. The biggest increase in G occurred after the split between the members of the tribes Hominini and Gorillini 10 million years ago. Hence at the macroevolutionary scale, there can be said to have been a “ten-million-year explosion” in primate G leading up to modern humans.