C. Otoacoustic emissions

SFOAEs and DPOAEs were analyzed with a least squares fitting (LSF) technique (Long et al., 2008; Kalluri and Shera, 2013; Abdala et al., 2015). To apply LSF modeling, the time waveform recorded at the microphone was segmented into chunks or analysis windows, and models for the stimulus tones and the OAE were created. Signals of interest within specified analysis windows were then determined by a least squares fit, which minimizes the sum of the squared residuals between the model and the data. The LSF analysis can be thought of as a moving bandpass filter with varying center frequencies corresponding to the frequency of the OAE (Long et al., 2008). OAEs evoked with swept-tones and analyzed with an LSF technique produce estimates of amplitude and phase that are comparable to those generated with discrete tones (Long et al., 2008; Kalluri and Shera, 2013; Abdala et al., 2015). For measurement of the total DPOAE (including both distortion and reflection parts), the optimal LSF analysis-window duration is 125 ms when combined with a sweep rate of 0.5 octaves/s (Abdala et al., 2015). However, to isolate the nonlinear distortion part of the DPOAE, a longer window must be applied (500 ms) with a relatively fast sweep (such as the rate used here) to effectively eliminate the longer-latency reflection energy, leaving the nonlinear distortion component (Long et al., 2008; Abdala et al., 2015). Only the distortion component of the total DPOAE was used in the joint OAE profile. Whenever the acronym DPOAE was used in the Results section or in figures, it refers only to isolated distortion part of the total DPOAE. For analysis of the SFOAE, LSF analysis windows were set at 100 ms with 2 kHz/s sweep rate. For both emission types, the noise floor was estimated by taking the difference between adjacent sweep pairs and applying the LSF to this difference.