Does the primate brain follow the eye?

3 minute read

In the same issue of Journal of Human Evolution that we find the snake paper, there is an article by Christopher Kirk proposing that neocortex sizes in different primate lineages are explained by vision requirements. In my earlier comment on the possibility that vision drove encephalization, I wrote this:

The interesting relations must be inside the brain -- for instance, are larger neocortex sizes specifically adaptations to group living? Or are they involved with interpreting and acting upon visual signs? Visual signs would include those associated with predators, foods, and other individuals. It's not obvious to me that visual cortex allometry is going to give you a good test of these relations, since adaptation would require interpretation and action upon visual signals, and not merely sensing them.

Kirk has found that species with larger optic foramina also have larger brains. Optic foramen area is argued to track the number of retinal ganglion cells as an indicator of total visual information input. So the conclusion is that species with more visual inputs also have larger brains:

Evolutionary increases in the amount of visual input to the brain have led to reciprocal increases in encephalization among primates. These effects were probably due in part to the hypertrophy of cortical areas devoted to visual tasks in accord with the principle of proper mass. Sequential increases in visual input may partly explain the relatively high encephalization both of primates relative to non-primate mammals and of anthropoids relative to other primates.

He also finds this relationship to hold in carnivores, another group where visual processing may be highly linked to fitness.

And one of the kinks for the hypothesis -- the fact that Aegyptopithecus probably had a similar visual input level as extant anthropoids but had a relatively smaller brain -- emerges as a reason for including the entire neocortex in the hypothesis and not just the primary visual areas:

However, this grade shift between extant anthropoids and strepsirrhines cannot be entirely the result of differences in visual input per se because some fossil anthropoids that presumably had all-cone foveae (e.g., Aegyptopithecus and Simonsius) were less encephalized than most living anthropoids (Radinsky, 1977, Jerison, 1979, Martin, 1990 and Bush et al., 2004). As noted by Bush and colleagues (2004), data for Simonsius suggest that high visual acuity evolved before significant brain expansion occurred in the anthropoid lineage, although high acuity may have subsequently favored increased encephalization in the context of visually-directed social or ecological tasks (cf. Barton, 1998 and Barton, 2000). Nonetheless, endocast morphology led Radinsky (1975) to suggest that V1 in Aegyptopithecus was "expanded as in modern anthropoids and larger than would be expected in a prosimian endocast of comparable size" (p. 660). If these interpretations (Radinsky, 1975 and Bush et al., 2004) are correct, then the fossil record of anthropoid brain evolution may provide evidence for two processes by which increased visual input has influenced relative brain size. First, cortical visual areas may have expanded early in anthropoid evolution coincident with the evolution of increased visual input and high acuity vision. Such early expansion could account for putative increase in the size of V1 in Aegyptopithecus (Radinsky, 1974 and Radinsky, 1975) and the fact that Aegyptopithecus was more encephalized than most other Eocene-Oligocene primates (Radinsky, 1977, Jerison, 1979 and Martin, 1990). Second, selection for increased visual processing in the context of various social or ecological tasks (e.g., interpreting facial expressions of conspecifics, creating mental maps of a home range, etc.) could have led to further increases in relative brain size among anthropoids (Barton, 1996, Barton, 1998, Barton, 2000 and Bush et al., 2004). Over time, both mechanisms could have contributed to the high degree of encephalization that distinguishes living anthropoids from other primates.

That's an important point -- visual "processing" is not merely a cognitive subtask performed by posterior cortical regions; it involves the integration of visual perception with sign recognition and higher cognition. In other words, perceiving a richer visual world lays open the possibility of adapting to that increased information by finding new ways of discriminating signs. Richer visual information allows the construction of more detailed mental maps. More ability to recognize other conspecifics allows more finely nuanced social decision-making and behavior. Being better able to distinguish the action signs of other individuals makes it more possible to incorporate those actions into future behaviors, so that vision supports social learning.


Kirk EC. 2006. Visual influences on primate encephalization. J Hum Evol 51:76-90. DOI link