The evolutionary origin of mammals is marked among other things by the elaboration of a six-layered neo-cortex and a general expansion of the brain. In addition, visual projections that in reptiles project mainly to the midbrain shift in mammals to the cerebral cortex (cerebral hemispheres) via the lateral geniculate nucleus of the thalamus (Aboitiz, 1992, 1995). In reptiles and birds, sophisticated topographic maps of the visual field, so-matosensory surface, and the auditory space tend to be located in the optic tectum of the midbrain (equivalent to the superior colliculus in mammals; see Stein and Meredith, 1993), while in the cerebral hemispheres the regions equivalent to the mammalian sensory cortices have poorly defined maps of the respective sensory surfaces (Ulinski, 1990; Aboitiz, 1992, 1995). In the optic tectum there is a commissure (the tectal commissure)
that plays the role of fusing the two visual hemifields in reptiles. We suggest that, as a consequence of the relative reduction of the optic tectum and the associated displacement of the topographic sensory information to the cerebral cortex, selective pressure developed in early mammals for an efficient commissural system in the cerebral hemispheres that could perform midline fusion particularly in the visual system. Monotremes and marsupials bear a six-layered neocortex with topographic sensory maps but have no corpus callosum. Instead, they have a well-developed anterior commissure (and sometimes an associated fasciculus aberrans) containing fibers that connect frontal, somatosensory, auditory, and visual cortices (Granger, Masterton, and Glendenning, 1985; Bradshaw and Rogers, 1993). However, this is not a "smart" route to follow, since visual areas located in the occipital lobe have to send fibers traveling all the way forward to the anterior commissure and then return to the contralateral occipital lobe to perform midline fusion. (There are also fibers crossing through the hippo-campal commissure, but they seem to be relatively few.) A more efficient design is to send the axons via a shortcut. As we mentioned above, embryological evidence indicates that this was the solution attained by placental mammals: Commissural fibers from visual cortex may have started crossing through the hippocampal commissure, which turned out to be of advantage in terms of visual processing. Eventually, a fiber tract differentiated from the hippocampal commissure and developed as the corpus callosum. Why did monotremes and marsupials not find this simple solution? Perhaps they did not need it. The superior colliculus (optic tectum) is relatively well developed in primitive mammals, and thus midline fusion may still be performed in the midbrain to a large extent, even though the cerebral cortex is already processing topographic relations. In other words, the shift of topographic sensory information from the midbrain to the mammalian cerebral cortex may have been a gradual process in which midline fusion was among the latest functions to be fully transferred to the cerebral hemispheres. Marsupials and monotremes thus represent intermediate stages in this process.
Midline fusion is still an important callosal function in modern mammals. Preliminary findings in our laboratory (Olivares, Michalland, and Aboitiz, 1995) indicate that frontally looking species such as carnivores and primates tend to have a relatively larger posterior third of the corpus callosum than do laterally looking species such as ungulates and rodents. Although it is not certain to what extent these fibers connect homologous areas, the findings are consistent with a higher proportion of posterior (i.e., visually related) callosal fibers in frontally looking species than in laterally looking ones, since the former tend to develop depth vision and therefore the task of midline fusion may be of more importance than in other species. Additionally, it is of interest that among placental mammals there is a good correlation between brain size and total callosal area (with an exponent approximating 0.75; see Olivares et al., 1995), although cetaceans have a smaller callosum than many other mammals (Glezer et al., 1988). The visual system of cetaceans is not as developed as that of other large-brained mammals, and perhaps tasks such as midline fusion are not of high relevance. Furthermore, the inferior and superior colliculi are relatively large in these species, and it is possible that many commisural interactions associated with echolocation are achieved in the mesencephalon.
The above scenario proposes that the main adaptive value of the corpus callosum relates to the process of midline fusion in sensorimotor regions of the neocortex. However, we have seen that callosal fibers connecting these areas are relatively few, most callosal projections belonging to higher-order or association cortices. Together with developing midline tasks, the corpus cal-losum probably became involved in coordination of more subtle processing between the two hemispheres. It may have worked as a monitoring system to check for different stages of parallel processing in the two hemispheres and as a way to transfer sensory and motor information to the other hemisphere, especially in relation to learning and plasticity. More generally, there has been a tendency in mammals to develop corticocortical connections between different regions as the number of cortical areas increases, and since a straightforward path between the hemispheres was available through the corpus callosum, interhemispheric connections became no exception to this trend. Thus, we claim that the main evolutionary advantage of having a corpus callosum relies, at least initially, more on cortical midline fusion than on interhemispheric communication between higher-order areas, although the latter benefited from the origin of the corpus callosum. In later stages of eu-therian (placental mammal) evolution, the callosum may have permitted appropriate synchrony of processes in the two hemispheres and, at the same time, the transfer of information from one side to the other (which might not occur to the same degree in animals whose hemispheres are disconnected), and this may have enhanced the processing capacity of the brain.
Was this article helpful?