Investigating potential interactions between envelope following responses elicited simultaneously by different vowel formants
Vijayalakshmi Easwar,Susan Scollie,David Purcell
DOI: https://doi.org/10.1016/j.heares.2019.05.005
2019-09-01
Abstract:Envelope following responses (EFRs) evoked by the periodicity of voicing in vowels are elicited at the fundamental frequency of voice (f0), irrespective of the harmonics that initiate it. One approach of improving the frequency specificity of vowel stimuli without increasing test-time is by altering the f0 selectively in one or more formants. The harmonics contributing to an EFR can then be differentiated by the unique f0 at which the EFRs are elicited. The advantages of using such an approach would be increased frequency specificity and efficiency, given that multiple EFRs can be evaluated in a certain test-time. However, multiple EFRs elicited simultaneously could interact and lead to altered amplitudes and outcomes. To this end, the present study aimed to evaluate: (i) if simultaneous recording of two EFRs, one elicited by harmonics in the first formant (F1) and one elicited by harmonics in the second and higher formants (F2+), leads to attenuation or enhancement of EFR amplitude, and (ii) if simultaneous measurement of two EFRs affects its accuracy and anticipated efficiency. In a group of 22 young adults with normal hearing, EFRs were elicited by F1 and F2+ bands of /u/, /a/ and /i/ when F1 and F2+ were presented independently (individual), when F1 and F2+ were presented simultaneously (dual), and when F1 or F2+ was presented with spectrally matched Gaussian noise of the other (noise). Repeated-measures analysis of variance indicated no significant group differences in EFR amplitudes between any of the conditions, suggesting minimal between-EFR interactions. Between-participant variability was evident, however, significant changes were evident only in a third of the participants for the stimulus /u/ F1. For the majority of stimuli, the change between individual and dual conditions was positively correlated with the change between individual and noise conditions, suggesting that interaction-based changes in EFR amplitude, when present, were likely due to the restriction of cochlear regions of excitation in the presence of a competing stimulus. The amplitude of residual noise was significantly higher in the dual or noise relative to the individual conditions, although the mean differences were very small (<3 nV). F-test-based detection of EFRs, commonly used to determine the presence of an EFR, did not vary across conditions. Further, neither the mean reduction in EFR amplitude nor the mean increase in noise amplitude in dual relative to individual conditions was large enough to alter the anticipated gain in efficiency of simultaneous EFR recordings. Together, results suggest that the approach of simultaneously recording two vowel-evoked EFRs from different formants for improved frequency-specificity does not alter test accuracy and is more time-efficient than evaluating EFRs to each formant individually.