Cortical Population Coding and Network Effects of Cochlear Implant Stimulation During Behavior - Abstract Globally, over one million people rely on cochlear implants (CIs) to hear and listen. However, inter-individual variance is high and ranges from the effective absence of functional communication to almost typical hearing levels. Most CI users do not attain speech comprehension scores close to people with typical hearing for often unclear reasons, as there is a significant lack of predictors and objective markers for CI therapy success. To aid the NIDCD’s mission to help prevent, detect, diagnose and treat hearing disabilities, it is crucial that we understand more about when and why CIs do and do not work. While most current research focuses on device technology and the electrode-nerve-interface, it is widely appreciated that an integral part to CI rehabilitation is brain adaptability. In fact, clinical results showing increased speech understanding over time dependent on onset and duration of deafness strongly hint at a major role of experience-dependent brain plasticity and perceptual learning for CI outcomes. Despite this significance, the exact neural changes induced by chronic CI use remain largely unknown even though they can be recorded in animals, a gap that this proposal is aimed at narrowing. The long-term goal of this research is to advance and develop CI stimulation strategies in ways that optimally harness, utilize, and support brain function and plasticity. There is increasing evidence that sensory cortical ensembles form small functional networks crucial for experience- dependent plasticity and perceptual learning. Cortical rhythmic synchronization associated with such sensory circuits and local small networks is implicated with phoneme discrimination and speech understanding in noise in human CI users. The aims of this research proposal are constructed to identify those cortical population activity mechanisms that contribute to CI success. The experiments focus on how small networks and functional oscillations commonly activated during perceptual learning are disrupted during CI- guided learning. All aims will be achieved with daily brain activity recordings in acutely deafened mice learning to use CIs during behavior. We will first chronically record high-density intracortical activity as mice learn to interpret CI stimulation in a reward-based operant conditioning task over several weeks, hypothesizing that small network formation is disrupted due to increased correlated variability elicited by hypersynchronizing CI stimulation. We will further test cortical oscillatory activity measured by EEG as a predictor and objective marker of CI behavioral outcome in mice. All measures taken will be correlated with behavioral performance and learning speed to find candidate mechanisms of interest for future R01 studies tracking and manipulating detailed networks activated during CI-guided behavior to increase behavioral performance and learning speed. The completion of our aims is thus a critical first step in unlocking the full potential of CI technology by taking into account the brain’s heavy lifting during CI rehabilitation – an urgent endeavor given that an estimated 75 million Americans and 700 million people worldwide will suffer from moderate to complete hearing loss by 2050.
