Humans are able to discriminate sounds with frequencies that are separated by no more than 0.1%. This level of performance arises from the exceptional sensory capacities of our cochlea, which detects stimuli through an array of sensory hair cells, spatially organized to follow a gradient of preferred frequency (a tonotopic map). This map is conserved downstream of the cochlea, implying that cochlear neurons must connect hair cells to the cochlear nuclei in the brain with high precision so as to retain the performance of the auditory system. This project addresses the question as to how this precision is achieved. In particular I propose that hair cell activity - both prehearing spontaneous, and stimulus dependent - might play a fundamental role in the fine tuning of this wiring. I will silence hair cell activity and measure the effect of this on their afferent innervation. To do this, I will express exogenous hyperpolarizing channels in the hair cells, using a hair cell specific promoter to prevent any effect on other cell types. Activity dependent plasticity might be temporally delimited in development. I will therefore render our hair cell specific promoter inducible by using a tetracycline responsive element, so as to restrict silencing to a defined time frame. Furthermore, I will make stochastic mosaics using X-chromosome inactivation to randomly silence half the hair cells. I will then test if the silencing of a given hair cell affects the afferent innervation of its neighbors, which would mean that hair cells compete for innervation through their activity. These experiments will provide a new framework to study the wiring of the cochlea, through which new questions may be addressed at the level of the afferent neurons and downstream. The medical implications of this are two-fold: it will help us understand pathologies related to the improper innervation of the cochlea and to develop strategies to improve the wiring of cochlear implants in the brain.