Date Published: May 29, 2011
Publisher: Hindawi Publishing Corporation
Author(s): Seth D. Dobson, Chet C. Sherwood.
Facial motor nucleus volume coevolves with both social group size and primary visual cortex volume in catarrhine primates as part of a specialized neuroethological system for communication using facial expressions. Here, we examine whether facial nucleus volume also coevolves with functionally unrelated brainstem motor nuclei (trigeminal motor and hypoglossal) due to developmental constraints. Using phylogenetically informed multiple regression analyses of previously published brain component data, we demonstrate that facial nucleus volume is not correlated with the volume of other motor nuclei after controlling for medulla volume. Our results show that brainstem motor nuclei can evolve independently of other developmentally linked structures in association with specific behavioral ecological conditions. This finding provides additional support for the mosaic view of brain evolution.
Two competing models of brain evolution have dominated the neuroscience literature over the past 15 years. The first posits that the interspecific scaling of vertebrate brain components is explained mostly by a conserved pattern of neurogenesis, such that structures that develop later tend to be relatively large [1, 2]. This is supported by the fact that later developing structures exhibit larger allometric exponents when scaled against overall brain size . Supporters of the developmental correlation model argue that brain structure evolves due primarily to selection on overall brain size, as opposed to the specialization of particular areas for specific functions . Thus, individual brain structures vary in size according to general scaling principles that constrain adaptive evolution, thereby limiting the impact of behavioral ecological conditions on brain structure.
Brain component volumes for 14 group-living, nonhuman catarrhine species were taken from previously published sources [13, 18, 19]. Group size data were taken from an unpublished dataset available on C. Nunn’s website  (http://www.people.fas.harvard.edu/~nunn/index.html). We examined trait correlations using multiple regression analyses. Two sets of analyses were carried out: (i) we examined the volume of the trigeminal motor nucleus and hypoglossal nucleus in relation to facial nucleus volume after controlling for medulla size  and (ii) we examined the relative volume of the trigeminal motor nucleus and hypoglossal nucleus in relation to group size. Autocorrelation, which can occur when the independent variable represents a large part of the dependent variable, is not a serious issue for our analyses because each nucleus comprises less than 0.5% of the volume of the total medulla . All data were log-transformed (natural) prior to analysis.
Figure 1 demonstrates the strong degree of covariation between brainstem orofacial motor nuclei in catarrhines prior to size correction. However, the multiple regression results in Table 1 indicate that this pattern of covariation disappears after controlling for medulla size, that is, neither trigeminal motor nucleus volume nor hypoglossal nucleus volume is a significant predictor of facial nucleus volume independent of medulla volume. Moreover, social group size is not positively correlated with either trigeminal motor nucleus volume or hypoglossal nucleus volume after size correction (Table 1). The results of the tree block analyses are identical with the consensus tree results because the degree of phylogenetic uncertainty in our sample is negligible. Thus, the hypothesis that catarrhine brainstem motor nuclei evolve in coordination with each other due to a shared developmental basis is not supported by our results. Instead, it appears that relative facial motor nucleus size evolves independently of the rest of the medulla and in association with social group size.
Previous research has shown that correlated evolution may occur within structurally interconnected neural systems [5, 6]. Our findings are unique in demonstrating that mosaic brain evolution can also involve coordinated changes in the volume of brain components that are not structurally linked by direct axonal pathways, but that participate in a common adaptive complex . These results also provide further support for the idea that neural specializations in mammals are not restricted to executive brain functions. Brainstem structures can also undergo adaptive specialization in response to the motor and/or sensory demands of specific behavioral ecological conditions [33–37].