Sound transduction occurs when mechanical stimuli open mechanoelectrical transduction (MET) ion channels located in specialised organelles in the internal ear, the hair bundles. These clusters of stereocilia are interconnected with molecular springs, the tip links, which stretch in accordance with the bundle’s deflection and project the mechanical force onto each channel. Biophysical measurements reveal that the relationship between the applied force and the bundles’ displacement is nonlinear. The canonical gating-spring model posits that this nonlinearity stems from a MET channel’s conformational change upon gating —the gating swing— which reduces the tension in the tip link. Although this view has been very successful, the inferred values of the gating swing are unrealistically large and experiments indicate that each tip link connects to two MET channels rather than one, raising questions about the interaction between them. Furthermore, the channels are not where they had been anticipated originally. No current model of MET can account for the number and location of MET channels or explain the origin of the large gating swing. In this thesis, I develop and analyse a new model of mechanotransduction in which a tip link connects to two channels that are mobile in the membrane. Their gating and positions are coupled by elastic forces in the lipid bilayer arising from the hydrophobic mismatch between each channel and the bilayer core. A large effective gating swing emerges in this framework from cooperative behaviour between pairs of channels. We reproduce the observed open probability, force and stiffness curves as functions of hair-bundle displacement, using only realistic biophysical parameters. Furthermore, we show how channel reclosure leads to bundle displacements during fast adaptation. We then validated experimentally the order of magnitude of some of the most relevant parameters of our model, such as the stiffness of the tip link, and measured the pivoting stiffness of individual stereocilia. For this task we used a recently developed technique called Hopping Probe Ion Conductance Microscopy that allows contactless mea-surements of stiffness in biological samples. We further explored the implications of our model and applied it to the study of hair bundle development. By doing so we we were able to account for phenomena that so far had lacked or had unsatisfactory explanations.