Entrainment effects and information processing in coupled oscillator models of auditory biomechanics

The ability to process sound as a means of gaining information about the environment is shared by many species of animals. The auditory system of many vertebrates contains non-linear and active elements, hair cells, in the inner ear. These are considered elementary for high sensitivity, high selectivity, and wide dynamic range. However, the implications of the complex interplay between active hair cell function and inner ear mechanics on information processing are not yet fully understood. Here we use the basic properties of vertebrate ears to develop a numerical approach that captures active and nonlinear elements with the goal of quantifying effects of entrainment on information processing. We show that entrainment and the generation of clusters affect information processing, and that inclusion of these phenomena extends the scope of commonly used models of hearing. The results show ubiquitous phenomena that are robust against variation of model parameters. We show that the encoding of information emerges as a global effect in the system rather than locally in individual elements. These findings have implications for our understanding of tonotopy in, e.g., the peripheral auditory system, and link active and nonlinear dynamic behavior with information encoding in the inner ear. This will help the identification of the key principles that underlie hearing in complex acoustical environments with strict physiological restrictions. The purpose of this study is to bridge entrainment effects with information processing in biological sensory systems. The results show the impact of self-organizing (spontaneous) and entrainment (externally driven) phenomena on the sensitivity, frequency selectivity, and information encoding capacity of the system. The frequency-specific sensitivity to external stimulation provides an explanation for physiological data recorded from the vertebrate auditory system. The link to information theory allows us to quantify the benefit of entrainment of sensory detection performance. Based on these results, acoustical, mechanical, and behavioral data can be evaluated within the same framework. This enables exploitation of the results for diagnostic and technological applications in, for example, audio signal processing.