Developing a clinical objective measurement of binaural processing

Binaural (‘two eared’) hearing is essential to effective everyday listening, particularly in noisy real-life situations. Binaural hearing and the access to binaural cues – such as small differences in the timing (interaural time differences, ITDs) and level (interaural level differences, ILDs) of sounds at the two ears - enable us to locate sound sources and to hear out conversations in competing background noise, overcoming ‘the cocktail-party’ problem. ITDs comprised in the temporal-fine structure of acoustic signals are the most relevant binaural cue that underlies speech in noise perception skills. Nevertheless, in the presence of hearing impairments, binaural and in particular ITD processing are usually affected. To measure binaural and ITD processing several psychoacoustic techniques have been documented. However, these techniques are usually unfeasible to conduct in clinical settings, particularly in difficult-to-test population. Therefore, objective measurements of ITD processing have gained interest from researchers and clinicians due to their advantages and potential clinical applications.

In this dissertation, two measurements of ITD processing used in laboratory settings, the ‘acoustic change complex (ACC) to transitions in interaural phase differences (IPD)’ (Ross et al., 2007) and the ‘interaural phase modulation-following response (IPM-FR)’ (Haywood et al., 2015, McAlpine et al., 2016; Undurraga et al., 2016) are assessed in normal hearing adults using a 64-channel electroencephalography (EEG) system. These techniques are evaluated regarding its feasibility to being acquired using diverse stimuli parameters such as intensity levels, interaural asymmetries, interaural phase modulations (IPMs); recording parameters comprising reference and active electrodes; and post-processing parameters including the application of two denoising techniques. This, aiming to develop a feasible and reliable clinical objective technique that allows to assess ITD processing in populations with speech in noise perception deficits such as those with hearing loss, hearing aid users and cochlear implant recipients, in the future.

Results show that both ACC and IPM-FR can be elicited reliably at an intensity of 70 dBA using both symmetric and asymmetric IPD transitions - IPM configurations -. However, when introducing interaural asymmetries, the magnitudes of both the ACC and the IPM-FR show reduced amplitudes, excepting the IPM-FR elicited by a symmetric IPM configuration, in both laboratory and clinical-like setups. Results also demonstrate that mastoid electrodes are amongst the best electrodes to obtain large and reliable responses in both measurements. However, the use of a reference electrode on the forehead significantly reduces the magnitude of these responses when compared to a vertex reference. The application of denoising techniques such as the electrooculography (EOG) template matching suppression (Valderrama et al., 2018) and the denoise separation source (DSS, de Cheveigne and Simon, 2008) reduce substantially the levels of noise in the EEG recordings, and therefore improve the signal-to-noise ratio of both measurements, particularly when using a forehead reference.

Both ACC and IPM-FR are suitable to use in the clinic using these stimuli parameters, however, in the presence of subclinical interaural asymmetries using a +/-90° IPM does not affect the IPM-FR magnitude. These differences in the interaction between the IPM and the interaural asymmetries are thought to reflect two different neural underlying mechanisms of ITD processing in the IPM-FR paradigm. The recording parameters for both ACC and IPM-FR should be carefully selected in a real clinical setup, with a vertex reference recommended to use since provides larger magnitudes on both measurements compared to the forehead reference, with mastoid electrodes being suitable active recording electrodes in a clinical-like setup. Both post-processing techniques demonstrated to improve the SNRs and reduced the EEG noise, and they are recommended to use in the clinic, considering the technical modifications to apply EOG suppression, and the limitations on the use of DSS.