Localization of high frequency noise with traditional hearing aids and personal sound amplification products (PSAPs)

Abstract

The aim of this study was to compare the speech-in-noise and localization performance of a personal sound amplification product (PSAP) versus a traditional hearing aid. The devices used in this study were the Soundhawk CS 50+ PSAP and the Oticon Nera miniRITE hearing aid. In order to qualify for the study, participants must have had a slight to moderately severe bilateral symmetric sensorineural hearing loss (PTA of 20 – 45 dB HL at 500 and 1000 Hz, PTA < 65 dB HL at 2000 Hz and 4000 Hz). Two qualifying participants with a slight to moderately severe sensorineural hearing loss were fit with both devices (PSAPs and HAs). Their speech-in-noise and localization ability was evaluated in three conditions (unaided, aided with PSAP, and aided with HA). 23 additional participants with varying degrees of hearing loss were evaluated in the unaided condition only. Speech-in-noise ability was assessed using the coordinate response measure (CRM) in order to obtain a speech identification threshold in the co-located and spatially separated condition. In the two qualifying participants, thresholds were obtained in the unaided, aided with PSAP, and aided with HA conditions. In both participants, the spatially separated condition produced a lower threshold for the unaided condition. However, the PSAP condition resulted in a lower speech identification threshold in the co-located condition. This indicated a loss of spatial advantage when using the PSAPs. Due to the small sample size, no statistical analyses were completed. In the 23 participants that completed the task in the unaided condition only, the spatially separated condition produced significantly lower speech identification thresholds compared to the co-located condition. Participants’ localization ability was also evaluated using a high frequency stimulus (3150 Hz). A 13-speaker array was used, and participants were asked to identify which speaker the stimulus originated from. In the two qualifying participants, the task was performed in all three conditions (unaided, aided with PSAP, and aided with HA). In participant 1, localization performance was best in the unaided condition, and worst in the PSAP condition. In participant 2, localization performance was best in the HA condition, and worst in the PSAP condition. In all three listening conditions, performance was better for speaker locations in front of the listener and declined as the speaker locations moved toward the periphery. Again, because of the small sample size, no statistical analyses were done. In the 23 participants that completed the task in the unaided condition only, the 3150 Hz stimulus was used in addition to a broadband stimulus. Participants had better overall localization performance when the broadband stimulus was used. With both stimuli, performance was better for speaker locations in front of the listener and declined as the speaker locations moved toward the periphery. Due to the small sample size of qualifying participants, no comparisons can be made to previous studies assessing the localization ability of PSAPs compared to traditional HAs. However, the results from the 23 participants who performed the tasks in the unaided condition show good agreement with previous localization research. Further research may consider less strict qualification criteria to obtain a larger sample size.