Neural strategies of temporal coding for sensorimotor processing

At the cellular scale, single-neurons process information mainly through spikes or action potentials. At the system scale, the brain is an intricate network of interconnected structures. The aim of this thesis is to integrate the cellular scale and the system scale by studying how ensembles of neurons translate the temporal precision of the spikes and the spatial precision of the anatomical connections into sophisticated strategies of temporal coding. The first part the thesis is dedicated to the problem of temporal coding in ensembles of single neurons recorded from the primary somatosensory cortex of the rat. The integration of the temporal precision of spikes with the spatial precision of somatotopic maps leads to a novel model of somatosensory processing, in which the somatosensory system not only translates spatial information from the body into somatotopic responses in the brain, but also transforms this spatial information into fine temporal codes through a sophisticated radial (i.e. center-to-surround) decomposition of the incoming information. The second part of the thesis is dedicated to the problem of temporal coding in large populations of neurons studied through local field potentials from the human subthalamic nucleus. Here the integration between temporal and spatial precision is studied in terms of rhythmical activity, which arises from the specific network configuration of the subthalamic area. Again, this integrative approach lead to a novel model of subthalamic function, by which - in addition to the rate coding of the classical basal ganglia model - populations of subthalamic neurons use rhythms at different frequencies as a basic element for constructing temporal codes that contribute to the complex involvement of the basal ganglia in the pathophysiology of movement control. In conclusion, ensembles of neurons can integrate the temporal precision of the spikes and the spatial precision of the anatomical connections into sophisticated strategies of temporal coding for sensorimotor processing.