Introduction

The default mode network (DMN) is unique in supporting internal mental processes in humans. Its deactivation is critical for engaging in cognitive processes (Raichle et al., 2001). An homologous network to the DMN was proposed in multiple non-human primate species [1-4]. Despite the similarities of the human and non-human primate DMN, qualitative differences are also consistently present in non-human primate studies [5]. The mPFC is deemed to support a gateway function in self representation in humans [6]. However, in most non-human primates, the mPFC has not been found to play a role in the DMN undermining the proposed role of the primate DMN in self-directed thought or introspection. Only in chimpanzees, has a DMN been defined where fMRI found that medial prefrontal regions were strongly connected to the PCC as in humans [7]. Together, these studies suggest a possible evolutionary gap between hominid and non-hominid primates (NHoPs) DMN connectivity, especially in the medial prefrontal area.

Methods

Fourteen anaesthetized mouse lemurs were scanned using an EPI sequence on an 11.7T [4]. Four common marmosets were scanned awake and anesthetized on a 9.4T [8] and thirteen anaesthetized macaques were scanned on 3T Siemens MAGNETOM Skyra scanner. Forty-eight healthy humans resting-state fMRI images were generated by Castellanos et al. [9]. Images from each species were co-registered to a standardize space (AFNI) for statistical analyses [10].

Results

We first used identical group dictionary analysis on fMRI data and seven components were set for all analyses to extract similar networks as in the functional atlas of Yeo et al. [11]. For each species, the seven large-scale networks were labelled according to the literature and to their anatomical features (Fig.1). In humans, we extracted networks identified in the literature as the DMN and the fronto-parietal network (FPN) (Fig. 1A). In macaques, the same analysis revealed a cortical network encompassing mainly PCC (posterior cingulate cortex), parietal, and dlPFC (dorso-lateral prefrontal cortex) areas identified as the FPN (Fig. 1B) as well as a network involving mPFC and areas surrounding superior temporal sulcus identified as the fronto-temporal network (FTN). In marmosets, we extracted a component encompassing PCC, parietal, and dlPFC areas identified as the fronto-parietal network (FPN; Fig. 1C). We also found evidence for the FTN encompassing the mPFC and temporal areas (Fig. 1C). In mouse lemurs, we determined FPN from a cortical network encompassing the PCC, the parietal posterior cortex, and anterior lateral frontal regions (Fig. 1D) and FTN encompassing mPFC, dlPFC, and the middle temporal cortex (Fig. 1D). Each large-scale network was transformed into a mask and concatenated together, producing a 3D atlas for each species (Fig. 2, each atlas can freely be downloaded at http://www.nitrc.org/projects/prim_func_2020/). Cerebral clusters were spatially separated and attributed to a unique label.
Fingerprint comparison
Fingerprint analysis was performed on key clusters of the fronto-parietal network/DMN and functional connectivity between these clusters was calculated using correlation coefficients. The mean connectivity matrix between the five clusters identified as Temporal (Temp, Fig. 3B), PCC (Fig. 3C), dlPFC (Fig. 3D), and Parietal (Par, Fig. 3E) was performed. Cosine similarity can be used as an index indicating the degree of similarity between two fingerprints. Permutation testing was performed on the cosine similarity indexes. A low cosine similarity associated to low p-value (p<0.05) suggests differences in connectivity profiles. We compared the fingerprints using permutation cosine similarity on similar clusters and between every pairs of species. NHoPs showed greater cosine similarity than humans (Fig 3. F-G).
Comparison of pairwise correlations
We performed a pairwise comparison of the relative differences between functional clusters of interest (Fig 2). We established that in macaques, marmosets, and mouse lemurs, the functional connection between PCC-mPFC was lower than between PCC-dlPFC (Fig. 4A) and the opposite relationship was observed in humans (Fig. 4A). This opposite relationship was also confirmed by comparing the correlation coefficients between Par-mPFC and Par-dlPFC (Fig. 4 B). To increase the spatial accuracy of the pairwise comparison, we replicated the analysis using three anatomical atlases from the literature [12-14]. Similar results were obtained, except that in humans, connectivity between mPFC (area 9) and dlPFC (8Ad) to PCC (23b) were not statistically different (Fig. 4C). To control for possible effects of anaesthesia, which some animals received, we compared their results with animals who did not receive anaesthesia. We found that relative differences of connectivity measured between mPFC, dlPFC to PCC (23b-9; 23b-8A) under anaesthesia, while still present were somewhat underestimated.

Conclusion

Identification of a common network anatomical architecture as well as functional connections supports the proposal that evolutionary rules apply distinctly to humans and NHoPs. First, the robust connection between mPFC-PCC was found to be unique to human (or hominid) specialization. Second, area 8Ad was found to be a common/ancestral frontal area of all primates’ DMN(-like). Globally, characteristics of high order networks in all non-hominid primate species examined are closer to each other than to humans. These new evolutionary rules of the DMN connectivity could open new research areas to explain the cognitive gap between hominid and non-hominid primates as well as improve the use of meaningful translational models of neuropathology.

Acknowledgements

No acknowledgement found.