The hidden structural complexity of the human cerebral cortex

The cerebral cortex exhibits an intrinsic structural complexity of folding. Fractal geometry may be used to describe the structural complexity of objects (such as the cerebral cortex) that show self-similarity throughout a proper range of spatial scales. The potential value of fractal dimension, estimated from T1-weighted MRI images, has already been widely demonstrated in previous literature, but how it is related to the cyto- and myeloarchitecture of the cerebral cortex is still unknown. In this study, we computed the fractal dimension of the six cortical layers, which present various cellular compositions and distributions, developmental trajectories, connections, physiology, and functional roles. The procedure for determining fractal dimension indices was applied to segmentations obtained from the public BigBrain dataset that consists of an ultrahigh-resolution three-dimensional (3D) model of a human brain based on histologically processed postmortem brain reconstruction. Results denote an increasing trend in fractal dimension values from the outermost (layer I) to the innermost layers (the infragranular layers V and VI) of the cortex in both hemispheres. Layers V and VI are composed of sparse, low-density pyramidal neurons and fusiform cells. Differently from the supragranular layers (i.e., layers I, II, and III), the infragranular layers are characterized by a high content of myelin belonging to the projection fibers. Consequently, the structural complexity of the cerebral cortex seems to be coupled with the projection fibers network, rather than with neuronal density, and the presence of associative and commissural fibers.