Network analysis is nowadays employed in so many fields, ranging from economics to engineering, to investigate any kind of system in terms of its elements and their relationships. Sociology is probably the discipline that has mainly exploited and improved this approach, and all the complex methods and statistics behind it. In neuroscience, it is mainly used to investigate neuronal connections. The perspective supplied by network analysis is incredibly powerful because it allows localizing roles and constraints within very complicated systems. Some elements can be crucial as hubs of local influences, some others can be important as bridges between distant components. Some elements can be very sensitive to local changes, some others can be more independent from neighboring variations. Systems can be complicated because of the many elements and relationships (hundreds or thousands or millions of nodes) or because of the nature of their relationships (many variables influencing the relationships). Or both. In any case, network analysis is an amazing tool to step into the organization of functional or structural systems. When applied to anatomical elements, we can talk of Anatomical Network Analysis. In its basic form, nodes are the anatomical elements, and relationships can be just their physical contact. The network, therefore, represents a topological model, in which the position and neighboring properties of each component are the evolutionary result of a selection balancing the architecture of the system, generating at the same time limitations and constraints. In evolutionary neuroanatomy and paleoneurology, too often each macromorphological change in a specific cortical region is directly (and speculatively) interpreted as a functional (cognitive) adaptation. However, brain geometry is also influenced by skull constraints, and each cortical region is also sensitive to spatial variations of its neighboring cortical elements. Therefore, some anatomical changes may be due to intrinsic changes (for example, cell proliferation or growth), while others may be due to secondary extrinsic factors (for example, pressures and strains of the neighboring environment). In a first attempt to apply network analysis to brain macroanatomy, one year ago we analyzed the cortical organization as described by the traditional Brodmann map, finding an interesting correspondence with the general subdivision in “lobes” and cranial fossae, and further integration of the parieto-occipital block. We have now published a more detailed study, on the regions generally considered in evolutionary neuroanatomy and paleoneurology, as to characterize their architectural roles and relevance in terms of overall topology. The posterior cortex is more integrated than the frontal regions, with the precentral gyrus acting as a spatial hinge between these two districts. Topological complexity matches, in this case, sulcal complexity. The temporal cortex is particularly sensitive to distinct cortical influences. We also considered a preliminary model including the main endocranial (bone) elements, evidencing the role of the anterior fossa and frontal bone in influencing the prefrontal cortical morphology. This first targeted analysis represents an invitation to go beyond simplistic approaches to brain morphological changes in paleoneurology and evolutionary neuroanatomy. Yet, it is just a kickoff. More steps are on the way.
