Widex Office of Research in Clinical Amplification (ORCA-USA), Lisle, Illinois, USA.
Ear Hear. 2020 Sep/Oct;41(5):1282-1293. doi: 10.1097/AUD.0000000000000847.
Understanding how signal processing influences neural activity in the brain with hearing loss is relevant to the design and evaluation of features intended to alleviate speech-in-noise deficits faced by many hearing aid wearers. Here, we examine whether hearing aid processing schemes that are designed to improve speech-in-noise intelligibility (i.e., directional microphone and noise reduction) also improve electrophysiological indices of speech processing in older listeners with hearing loss.
The study followed a double-blind within-subjects design. A sample of 19 older adults (8 females; mean age = 73.6 years, range = 56-86 years; 17 experienced hearing aid users) with a moderate to severe sensorineural hearing impairment participated in the experiment. Auditory-evoked potentials associated with processing in cortex (P1-N1-P2) and subcortex (frequency-following response) were measured over the course of two 2-hour visits. Listeners were presented with sequences of the consonant-vowel syllable /da/ in continuous speech-shaped noise at signal to noise ratios (SNRs) of 0, +5, and +10 dB. Speech and noise stimuli were pre-recorded using a Knowles Electronics Manikin for Acoustic Research (KEMAR) head and torso simulator outfitted with hearing aids programmed for each listener's loss. The study aid programs were set according to 4 conditions: (1) omnidirectional microphone, (2) omnidirectional microphone with noise reduction, (3) directional microphone, and (4) directional microphone with noise reduction. For each hearing aid condition, speech was presented from a loudspeaker located at 1 m directly in front of KEMAR (i.e., 0° in the azimuth) at 75 dB SPL and noise was presented from a matching loudspeaker located at 1 m directly behind KEMAR (i.e., 180° in the azimuth). Recorded stimulus sequences were normalized for speech level across conditions and presented to listeners over electromagnetically shielded ER-2 ear-insert transducers. Presentation levels were calibrated to match the output of listeners' study aids.
Cortical components from listeners with hearing loss were enhanced with improving SNR and with use of a directional microphone and noise reduction. On the other hand, subcortical components did not show sensitivity to SNR or microphone mode but did show enhanced encoding of temporal fine structure of speech for conditions where noise reduction was enabled.
These results suggest that auditory-evoked potentials may be useful in evaluating the benefit of different noise-mitigating hearing aid features.
了解信号处理如何影响听力损失患者大脑中的神经活动，这与旨在减轻许多助听器佩戴者在噪声环境中言语理解缺陷的特征的设计和评估有关。在这里，我们研究了旨在提高言语可懂度的助听器处理方案（即定向麦克风和降噪）是否也能改善听力损失老年患者的言语处理的电生理指标。
该研究采用双盲、自身对照设计。一项由 19 名听力损失的老年成年人（8 名女性；平均年龄=73.6 岁，范围=56-86 岁；17 名有经验的助听器使用者）组成的样本参与了实验。在两个 2 小时的访问过程中，测量了与皮质（P1-N1-P2）和皮质下（频率跟随反应）处理相关的听觉诱发电位。在信号噪声比（SNR）为 0、+5 和+10dB 的情况下，听众会听到连续语音形状噪声中的辅音-元音音节 /da/序列。语音和噪声刺激使用配备有每个听众损失的助听器的 Knowles Electronics Manikin for Acoustic Research (KEMAR) 头和躯干模拟器进行预录制。研究辅助程序根据 4 种条件进行设置：（1）全向麦克风，（2）全向麦克风+降噪，（3）定向麦克风，（4）定向麦克风+降噪。对于每个助听器条件，语音从位于 KEMAR 前 1 米（即方位角 0°）的扬声器以 75dB SPL 发出，噪声从位于 KEMAR 后 1 米（即方位角 180°）的匹配扬声器发出。记录的刺激序列根据条件对语音水平进行标准化，并通过电磁屏蔽的 ER-2 耳插入换能器呈现给听众。呈现水平根据听众研究辅助程序的输出进行校准。
听力损失患者的皮质成分随着 SNR 的提高和定向麦克风和降噪的使用而增强。另一方面，皮质下成分对 SNR 或麦克风模式不敏感，但在启用降噪时，对语音的时间精细结构的编码增强。
这些结果表明，听觉诱发电位可能有助于评估不同降噪助听器功能的益处。
