Both humans and nonhuman primates show changes in cognition, brain morphology, and behavior with age, both as a part of normal aging and as a result of age-related disease. Normal human aging is characterized by declines in cognition, global brain volume, gray matter volume, and functional activation associated with increasing age (Abe et al., 2008; Bigler et al., 2002; Bigler et al., 1997; Bishop et al., 2010; Brickman et al., 2007; Chen et al., 2013; Ge et al., 2002; Good et al., 2001; Hawkins et al., 2015). Most studies on nonhuman aging have utilized non-primate model species such as yeast, nematodes, fruit flies, and mice due mostly to availability, cost, short lifespan and well-documented genetic and/or physiological characteristics. While these models have provided important data and have advanced our understanding of aging, they do not closely model the human aging process. Utilizing nonhuman primate models (particularly those more closely related phylogenetically) is critically important given their genetic similarities to humans, longer lifespans, larger and more complex brains and more sophisticated cognitive abilities compared to other model species. Among nonhuman primates, baboons are particularly well-suited as a model for human aging. First, unlike other monkey species, there is evidence that baboons undergo well-coordinated growth, growth spurts, and an order of growth cessation like humans (Leigh, 2009). Second, the baboon genome is very similar to humans with over 90% homology of coding segments and 85% homology of non-coding regions. For this reason, the baboon has served as a model to study the genetics and epigenetics of several human diseases (Cox et al., 2013; Cox et al., 2006; Robinson et al., 2019; Rogers et al., 2009; Rogers et al., 2000). In addition, both human and baboon lifespan is moderately heritable (0.23-0.26 and 0.23, respectively) and differences in frailty likely causes variation in individual mortality risk within both species (Bronikowski et al., 2002; Martin et al., 2002). Third, baboons have been shown to develop similar age-related diseases and conditions as humans, including osteoporosis and arthritis, menopause, endometriosis, obesity, diabetes, cardiovascular disease, and Alzheimer's-like pathology (Martin et al., 2002; Schultz et al., 2000a; Schultz et al., 2000b; Schultz et al., 2001; Yeung et al., 2016). For example, postmortem studies have shown that baboons develop tau pathology in both neurons and glia and Aβ plaques as they age (Schultz et al., 2000a; Schultz et al., 2002; Schultz et al., 2000b; Schultz et al., 2001). Furthermore, compared to rhesus macaques (a common nonhuman primate model species), baboons have larger brains making them ideal for imaging studies (Black et al., 2009), and the baboon prefrontal cortex is more developed and more similar to the human prefrontal cortex than the rhesus macaque (Brent, 2009; Fridman and Popova, 1988). Due to these characteristics, studies of baboon aging can help bridge the translational gap between other model species and human aging studies.

Like humans, there is evidence of age-related cognitive decline in baboons. In a cross-sectional study of 19 baboons (aged 1–14), Bonté (Bonté et al., 2014) found that executive control (as measured by a transfer index task) declined with age. In another cross-sectional study of 6 baboons, older adults (20-23 years old) performed significantly worse on cognitive tasks (specifically, learning a novel task, precision, and simple discrimination) compared to younger adults (13-16 years old) (Lizarraga et al., 2020). These findings suggest that there are likely neurobiological changes occurring as baboons age. Although baboon neuroanatomy and function is well-studied (e.g. (Amiez et al., 2019; Atkinson et al., 2015; Black et al., 2009; Blaizot et al., 2004; Cain and Wada, 1979; Greer et al., 2002; Love et al., 2016; Marie et al., 2018; McBride et al., 1999; Meguerditchian et al., 2021; Rogers et al., 2007; Westerhausen and Meguerditchian, 2021), and postmortem studies reveal the development of tau pathology as they age (Schultz et al., 2000a; Schultz et al., 2002; Schultz et al., 2000b; Schultz et al., 2001), to date there have been few in vivo neuroimaging studies focused specifically on the aging baboon brain. Franke et al. (Franke et al., 2017) found that prenatal undernutrition (70% of normal daily food intake) had long-term effects on the baboon brain, prematurely aging the young adult female baboon brain by 2.7 years on average. In addition, the authors examined typical brain aging across the baboon lifespan and reported a significant decrease in overall gray matter volume and increase in overall white matter volume with age. More recently, Westerhausen and Meguerditchian (2021) examined corpus callosum morphology across the lifespan and found that the area of the corpus collosum increases slowly with age, particularly within the anterior section.

Here we examine the relationship between age and whole-brain gray matter covariation using source-based morphometry (SBM). SBM is a multivariate approach which incorporates an independent components analysis to create networks of brain regions that covary among gray or white matter voxels across the whole brain (Gupta et al., 2019). Here we examine the contribution of both age and sex to gray matter covariation within the baboon brain. A number of studies in human subjects have reported age-related changes in gray matter covariation networks (DuPre and Spreng, 2017; Evans, 2013; Hafkemeijer et al., 2014; Koini et al., 2018; Liu et al., 2017; Montembeault et al., 2012; Xu et al., 2009b), but there are only a few studies of this type in nonhuman primates. Source-based morphometry-like methods were used to show that older rhesus macaques had reduced gray matter covariation in components including dorsal and ventral prefrontal cortex, superior temporal sulcus, and sylvian fissure (Alexander et al., 2008). Previous studies in chimpanzees have also reported age-related changes in gray matter covariation, with decreased gray matter covariation reported for the superior frontal, supplementary motor, and anterior temporal cortex as chimpanzees age (Hopkins, W. D. et al., 2019). In addition, they reported increased gray matter covariation in the prefrontal and premotor cortices and part of the cerebellum as chimpanzees age (Hopkins, W. D. et al., 2019). A more recent study of age-related changes in chimpanzee gray matter (using voxel-based morphometry) found significant reductions in the prefrontal cortex, middle temporal cortex, superior frontal cortex, insula, superior temporal cortex, and entorhinal cortex with increased age (Mulholland et al., 2021). In the current study, we use source-based morphometry to examine the relationship between age and gray matter covariation in a baboon sample. We hypothesized that significant age effects would be found in one or more of the SBM components, specifically those comprised of regions known to be influenced by age in humans and other non-human primates, such as the middle temporal lobe, prefrontal cortex, and the cerebellum.

In addition, we examined sex differences in gray matter covariation. Studies of the baboon brain have indicated that sex may account for variation in overall brain volume and sulcal length (Atkinson et al., 2015; Rogers et al., 2007), differences in the area of the corpus collosum (Phillips and Kochunov, 2011), and whole brain gray and white matter volumes (Franke et al., 2017). Franke et al. (Franke et al., 2017) found that males had significantly higher absolute gray matter volume than females but did not examine regional differences or gray matter covariation. Atkinson et al. (Atkinson et al., 2015) reported that sex accounted for a significant proportion of variance in the length of the central sulcus, lateral fissure, lunate, and superior temporal sulcus. In chimpanzees, an SBM analysis revealed significant sex differences in gray matter covariation, including the cerebellum, primary visual, anterior temporal, and frontopolar cortices (Hopkins, W. D. et al., 2019). Similar to chimpanzees, here we hypothesized that there would be significant sex differences in one or more of the baboon SBM components.