Behavioral and pharmacological enhancement of neural plasticity in the adult visual cortex

During its early development, on a short time-window called the critical period, the brain structures and functions are very sensitive to sensory inputs. This sensitivity is supported by a state of enhanced plasticity and results from a high excitability of the brain. Once the critical period closes, due to a massive inhibition to balance the hyper-excitability, and the more the brain matures into adulthood, this plasticity becomes weaker. Some degree of plasticity remains after this initial critical period and plays a critical role in learning and brain repair. However, the weaker efficiency of plasticity in the adult brain can hinder recovery such as in adult amblyopia. Previous studies on rodents have demonstrated that a specific class of neuromodulators can increase perceptual learning and thus enhance visual plasticity. In the present thesis, we seek to understand the underlying neural mechanisms of visual plasticity when the perceptual learning is coupled with a neuromodulator, in adult non-human primates. To characterize such an effect, we used functional magnetic resonance imaging (fMRI), pharmacology and behavioral observations. We thus designed a longitudinal study to characterize the brain basal state (time point T1), then its changes in functional connectivity and activation patterns after a behavioral top-down manipulation, i.e. a reward manipulation biasing decision making in favor of specific parts of the visual field (time point 2). Finally, we characterized the changes occurring when this behavioral reward-based manipulation was coupled with a neuromodulator intake (time point 3). In order to enhance a plasticity, this neuromodulator has to re-instate neural excitability. We thus hypothesized that a selective serotonin reuptake inhibitor (SSRI), and specifically fluoxetine, which decreases GABA levels in the brain, hence resulting in a global (though complex) decrease of cortical inhibition, would contribute to enhance plasticity in the adult brain. Perceptual learning in the presence of fluoxetine produced marked behavioral and neural signatures of learning, thus indicating an enhanced plasticity. In a first study (behavioral), involving three distinct visual tasks coupling top-down and bottom-up mechanisms, we demonstrated that under fluoxetine luminance vision is degraded as well visual spatial resolution, while sensitivity to reward increases. In a second study (fMRI), we investigated the whole brain functional connectivity changes following an extensive reward-based learning of novel spatial contingencies and with the impact of fluoxetine. We highlight that the general decorrelation between the ventral and dorsal visual pathways observed following reward-based perceptual learning is further enhanced in the presence of the SSRI. We propose that these cortical changes underlie the enhanced sensitivity to rewarded spatial locations observed at the behavioral level and contribute to the maximization of behavioral outcome. Our third and last study (fMRI), further supports this view, by showing that while activations patterns in V2 track the spatial priority map constructed through the reward-based learning, these effects are greatly enhanced by fluoxetine. Quite remarkably, these changes are specific to the lower visual field cortical representations. We additionally show that fluoxetine degrades spatial resolution coding in the visual cortex, supporting our behavioral observations in study 1. Overall, we describe the neural and network mechanisms through which fluoxetine contribute to the re-instatement of visual cortical plasticity in the adult brain and their behavioral correlates. This has major implications in the clinical field in the context of sensory and cognitive recovery or learning.