Interpeak characterizations for spontaneous otoacoustic emissions

One manifestation of the “active ear" is the presence of spontaneous otoacoustic emission (SOAE), which also exhibit salient connections to perception such as threshold microstructure. Historically, SOAE modeling efforts initially focused upon a single limit-cycle oscillator. However, SOAE spectra from a given ear typically exhibit multiple peaks, and more current models consider a spatially distributed tonotopic system with various types of coupling. SOAEs have nonstationary properties (e.g., amplitude and frequency modulations), which may be crucially tied to the coupling of active elements in the ear. Thus, to better biophysically constrain models, this study seeks to improve characterization of general non-stationary features of SOAE peaks as well as interrelations of such between them. Given the ubiquitous nature of SOAE across the animal kingdom, we analyze SOAE waveforms from a variety of species exhibiting disparate inner ear morphologies (e.g., human, barn owls, Anolis lizards). This manuscript provides a preliminary account of our analyses and focuses on the Anolis lizard. Upon filtering in the spectral domain, we characterize temporal properties of individual peaks, including possible amplitude-modulation (AM) and frequency-modulation (FM). Further, we perform correlation analyses of such between peaks to determine types of interactions and how such might vary across time. Initial results are consistent with previous reports (e.g., [1, 2]) in that an SOAE interpeak correlations for a given ear are idiosyncratic: Sometimes peaks (adjacent or not) exhibit correlated (positive or negative) AM and/or FM fluctuations with delays up to the order of milliseconds (typically longer for humans, shorter for lizards), while sometimes no correlation is observed. We attempt to frame these results within the broader context of specific SOAE modeling approaches.