Jury:
Mme ROE Anna Wang, Professeure, Universite de Zhejiang, Hangzhou (Chine)
M.VANDUFFEL Wim, Professeur, Universite Catholique de Louvain (Belgique)
Mme ANGELUCCI Alessandra, Professeure, Universite de l’Utah, Salt Lake City (USA)
Mme BEN HAMED Suliann, Directrice de Recherche, CNRS Lyon ;
M.DERRINGTON Edmund, Professeur des Universites, Universite Lyon 1
M.HAYASHI Takuya, Professeur Associe, RIKEN Center for Biosystems Dynamics Research, Kobe (Japon)
M.KENNEDY Henry, Directeur de Recherche, INSERM Lyon
M.KNOBLAUCH Kenneth, Directeur de Recherche, INSERM Lyon
Invited :
DEHAY Colette, DR CNRS, SBRI
HIBA Bassem, CR, ISC – Marc Jeannerod Invité
Abstract
The work performed during this thesis includes two projects related to the macaque cortical connectome. The thesis opens with an in-depth review of spatial embedding of the cortex, work that has been largely carried out in the host team in the past 20 years, and which are fundamental for two projects that constitute the thesis. The first project investigates the impact of retinotopic subdivisions of the visual cortex on the primate visual cortical connectome. The second project compares the cortical connectome revealed by diffusion MRI tractography with that derived from ground truth retrograde tract tracing.
In the first project, 21 injections were performed in different retinotopic subdivisions of areas V1 and V2. This revealed important regularities of projection strength across the cerebral cortex to upper, central and lower visual fields of V1 and V2 with respect to trajectory distance and areal origin. Projections to the central and upper visual fields are significantly stronger from ventral stream areas, while peripheral and lower field projections are stronger from the dorsal stream. Peripheral injections revealed projections from 15 areas that are not labeled by central injections. These observations align with numerous published physiological and psychophysical findings. These results suggest that the top-down pathways not only serve multisensory integration but might also play a role in active vision.
In the second project, we compare the weighted inter-areal cortical connectivity in the macaque using high-resolution dMRI based tractography in 9 macaque brains with retrograde tracer from a novel and considerably extended connectivity matrix of 42 areas. With respect to tractography we calculated the connection weight revealed as well as the parentage of false positives and false negatives using a ground truth tract tracing 42×42 connectivity matrix. By using surface mapping technique, we further investigated the non-parcellated spatial distributions of tract tracing and tractographic streamlines on the brain surface, which allowed the characterization of the relative localization of true positives, false positives and false negatives. Our results suggest that tractography and tract tracing are underpinned by very different processes which while leading to superficial similarities, are characterized by fundamental differences which may be reasonable given the numerous false positives and false negatives which characterizes tractography. The observed differences in the weight-distance distributions suggest that tractographic connectomes will fail to correctly identify the properties of the inter-areal cortical connectome.
