Soutenance de thèse de Joachim Confais

27 mars 2013

titre : Timing in motor cortex : from cue anticipation to movement preparation


Abstract : Our actions are shaped by the spatial and temporal context in which they take place. With experience, we learn to extract the necessary spatial and temporal information, and use it to improve our sensorimotor behavior. During my thesis, I assessed the influence of the temporal context (i.e., the timing) on movement preparation, at the neuronal level. More specifically, I performed extracellular recordings in the motor cortex of two macaque monkeys while they were performing a delayed center-out reaching task. In our task, the cue carrying the spatial information about the movement and the cue requiring its execution were temporally distinct. Furthermore, their timing changed from trial to trial but was predictable, thus allowing the animals to anticipate their appearance. We recorded and analyzed about 1200 neurons and 650 local field potentials (LFPs). We found that prior to the onset of the cue, that is in absence of movement preparation, 40% of the motor cortical neurons showed an anticipatory activity, either ramping up or ramping down. The pattern of activity of these neurons (increasing or decreasing) was predictive of their activity during movement preparation, especially in the epoch following the spatial cue. The two categories of neurons were as much involved in motor preparation, as reflected in their degree of directional selectivity. In the period preceding the cue, neurons with the same pattern of anticipatory activity tended to co-vary positively from trial to trial, whereas neurons with opposite patterns were negatively correlated. We proposed the hypothesis that preceding the onset of an informative cue, the motor cortex enters into an optimal state, at the same time facilitating the processing of the cue and preventing the generation of an early unwanted movement. During movement preparation itself, between the spatial cue and the execution of the movement, the neuronal activity was deeply influenced by the delay duration. Specifically the LFP signals evoked by the spatial cue (Visual Evoked Potential, or VEP) and by the movement execution (Movement Related Potential, MRP) were inversely modulated by delay duration, the VEP being larger in short and the MRP larger in long delay trials. The firing rate modulation of neurons recorded during the same epoch followed the same pattern of delay selectivity, although not as strong as that of the LFP. The delay selectivity of both signals seemed rather independent of their selectivity to movement direction, but as important. However, the time course of the two selectivities was different, with the delay selectivity being more influent at the beginning of the delay, whereas the direction selectivity being more influent at the end of the delay. The delay selectivity may be the reflection of motor strategies or the influence of motivational factors. Lastly, we showed that in the context of a predictable GO signal, the ability to estimate the delay duration seems to be the best predictor for the variability of the animal’s performance trial by trial.

Jury de thèse : Marie-Claude Hepp-Reymond, Zürich (rapporteur) Aldo Genovesio, Rome (rapporteur) Franck Vidal, Marseille (examinateur) Jennifer Coull, Marseille (examinateur) Emmanuel Procyk, Bron (examinateur) Alexa Riehle, Marseille (directeur)

CNRS logo université Aix Marseille logo | plan du site | mentions légales | contact | admin | intranet | intcloud |