In their 2003 Trends in Neurosciences article, Hutsler & Galuske refer to the well-known history of hemispheric asymmetry research as too focused on large-scale morphological differences, at the expense of microanatomical and connectivity differences. An understanding of these more detailed structural differences might translate into a more detailed understanding of hemispheric differences in computation and function.
Hutsler & Galuske identify three levels of structural analysis in cerebral cortex: the microcolumn, the macrocolumn, and the functional column. Microcolumns contain vertically-aligned pyramidal neurons, about 20-50 microns wide, and are sometimes considered the “smallest processing unit in the cerebral cortex.” Anatomical macrocolumns can be defined by similarities in patterns of long-range afferent connectivity, and are usually 200-700 microns wide. Finally, functional macrocolumns are similar in terms of their receptive field properties – the stimuli to which neurons in a particular region fire.
Differences between the right and left hemispheres have been observed at all three levels of analysis. For example, Hutsler & Galuske note that the distance between anatomical macrocolumns is larger in the left hemisphere than the right. A compensatory increase in basiliar dendritic length can be observed the left hemisphere, thus generally allowing similar intercolumnar connectivity between left and right hemispheres. Hutsler & Galuske also note that there are more large pyramidal cells in left- than right-hemispheric auditory regions, as well as larger layer-III pyramidal neurons in Broca’s area in the left-hemisphere. Similarly, regions of dense local interconnectivity are spaced more widely in the left-hemispheric Wernicke’s area than the homologous region on the right.
How do these structural differences relate to functional differences? The authors suggest that less dense left-hemispheric intercolumnar connectivity might result in more “local” processing, particularly in areas associated with language, where the increase in dendritic length is not as pronounced as in other left hemispheric regions. Hutsler & Galuske also conclude that the left hemisphere may have a more “refined processing architecture” because any given left-hemispheric macrocolumn is likely to contain fewer microcolumns than a similarly-sized macrocolumn on the right.
Although evidence is scarce on hemispheric asymmetry in long-range connectivity (largely due to the technical difficulty of acquiring such evidence), Hutsler & Galuske do present facts that are highly suggestive of large-scale connectivity differences. First, inter-regional connectivity is important for shaping cortical folds, which are known to be asymmetric between left and right. Second, since the number of “support” (e.g., glial) cells is similar between left and right, increased inter-regional connectivity may compensate for the left-hemispheres’ increased column spacing. Finally, myelination is greater in the left hemisphere, also suggestive of increased long-range connectivity. In my opinion, these findings would seem to belie the conclusion that the left hemisphere is more “local” than the right (one might easily interpret increased long-range connectivitiy in the left hemisphere as more “global” ).
In their conclusion, Hutsler & Galuske suggest that although hemispheric asymmetry appears to be present at birth (and thus may be genetically pre-sepcificed), evidence from children with hemispherectomies demonstrates that the processes leading to specialization are not set in stone. As with many other things, asymmetries are a product of both internal and external factors, and remain plastic throughout development.