Le but de la thèse était d’avancer les connaissances sur les procédures diagnostiques ayant pour objectif  la latéralisation des déficits de la mémoire visuelle, ainsi que d’étudier la perception dans l´épilepsie du lobe temporal (ELT). La première étude a examiné l’aptitude de quatre tests de mémoire visuo-spatiale, comme méthode d´identification des dysfonctions du lobe temporal médial droit. Des méthodes statistiques variées, illustrent la capacité limitée de ces tests à latéraliser la zone dysfonctionnelle, même lorsque les facteurs cliniques liées à l’épilepsie ainsi que d’autres facteurs cognitives étaient contrôlés. La deuxième étude est fondé sur la première, et à pour but de valider un nouveau paradigme. Ainsi, nous avons examiné des patients avec ELT en préopératoire avec le Delayed Matching to Sample (DMS-48) test et en postopératoire avec deux versions parallèles que nous avions développées et standardisées auprès des sujets contrôles sains. Nos analyses démontrent que le DMS-48 et ces versions parallèles ont pu latéraliser la zone épileptogène avant et après l´intervention chirurgicale. La troisième partie comprime une expansion de la mémoire visuelle à la perception visuelle. Une étude de cas unique suggère que la reconnaissance de l´objet et l´imagerie mentale sont soutenues par des régions corticales proches, situées en contiguïté de la voie temporo-occipitale. De plus, la perception et l´imagerie de l´espace sont rendues possibles par des mécanismes en dehors de la voie ventrale. En outre, ces résultats semblent indiquer que des dysfonctions chroniques du système visuel peuvent être réversibles avec un contrôle des crises effectif. / The aim of the present thesis was to advance the knowledge of diagnostic procedures for lateralizing visual memory deficits and to study the characteristics of perception in temporal lobe epilepsy (TLE). The first study examined the appropriateness of four routinely used learning and reproduction visuo-spatial memory tests as an identification method for right mesial temporal lobe dysfunctions. Various statistical methods illustrate the tests poor capacity to lateralize the functional deficit zone, even when epilepsy-related clinical and other cognitive factors were controlled. The second study is built upon the results of the first, aiming to validate a new test paradigm for lateralizing right hippocampal dysfunctions. Thus we assessed mesial TLE patients preoperatively with the Delayed Matching to Sample (DMS-48) task and postoperatively with two parallel versions that we had developed and standardized in healthy controls. Our analysis suggests that the DMS-48 and its parallel versions were able to lateralize the epileptic onset zone pre- and postsurgically. The third part consists of an expansion from visual memory to visual perception. A study of single case suggests that visual object recognition and visual imagery are sustained by cortical areas located in proximity to the temporo-occipital ventral pathway and that perception and imagery for space is subserved by mechanisms, which are close anatomically, and outside the ventral path. Furthermore, the results seem to indicate that nonlesional paroxysmal activity in the posterior temporal lobe can cause chronic dysfunctions of the visual system, which may be reversible with effective seizure control.

No description available.

The aim of the present thesis was to advance the knowledge of diagnostic procedures for lateralizing visual memory deficits and to study the characteristics of perception in temporal lobe epilepsy (TLE). The first study examined the appropriateness of four routinely used learning and reproduction visuo-spatial memory tests as an identification method for right mesial temporal lobe dysfunctions. Various statistical methods illustrate the tests poor capacity to lateralize the functional deficit zone, even when epilepsy-related clinical and other cognitive factors were controlled. The second study is built upon the results of the first, aiming to validate a new test paradigm for lateralizing right hippocampal dysfunctions. Thus we assessed mesial TLE patients preoperatively with the Delayed Matching to Sample (DMS-48) task and postoperatively with two parallel versions that we had developed and standardized in healthy controls. Our analysis suggests that the DMS-48 and its parallel versions were able to lateralize the epileptic onset zone pre- and postsurgically. The third part consists of an expansion from visual memory to visual perception. A study of single case suggests that visual object recognition and visual imagery are sustained by cortical areas located in proximity to the temporo-occipital ventral pathway and that perception and imagery for space is subserved by mechanisms, which are close anatomically, and outside the ventral path. Furthermore, the results seem to indicate that nonlesional paroxysmal activity in the posterior temporal lobe can cause chronic dysfunctions of the visual system, which may be reversible with effective seizure control.