Department of Cognition, Emotion, and Methods in Psychology, Faculty of Psychology, University of Vienna, Vienna, Austria.
Acoustics Research Institute, Austrian Academy of Sciences, Vienna, Austria.
J Assoc Res Otolaryngol. 2022 Feb;23(1):119-136. doi: 10.1007/s10162-021-00821-3. Epub 2021 Nov 23.
Normal-hearing (NH) listeners rely on two binaural cues, the interaural time (ITD) and level difference (ILD), for azimuthal sound localization. Cochlear-implant (CI) listeners, however, rely almost entirely on ILDs. One reason is that present-day clinical CI stimulation strategies do not convey salient ITD cues. But even when presenting ITDs under optimal conditions using a research interface, ITD sensitivity is lower in CI compared to NH listeners. Since it has recently been shown that NH listeners change their ITD/ILD weighting when only one of the cues is consistent with visual information, such reweighting might add to CI listeners' low perceptual contribution of ITDs, given their daily exposure to reliable ILDs but unreliable ITDs. Six bilateral CI listeners completed a multi-day lateralization training visually reinforcing ITDs, flanked by a pre- and post-measurement of ITD/ILD weights without visual reinforcement. Using direct electric stimulation, we presented 100- and 300-pps pulse trains at a single interaurally place-matched electrode pair, conveying ITDs and ILDs in various spatially consistent and inconsistent combinations. The listeners' task was to lateralize the stimuli in a virtual environment. Additionally, ITD and ILD thresholds were measured before and after training. For 100-pps stimuli, the lateralization training increased the contribution of ITDs slightly, but significantly. Thresholds were neither affected by the training nor correlated with weights. For 300-pps stimuli, ITD weights were lower and ITD thresholds larger, but there was no effect of training. On average across test sessions, adding azimuth-dependent ITDs to stimuli containing ILDs increased the extent of lateralization for both 100- and 300-pps stimuli. The results suggest that low-rate ITD cues, robustly encoded with future CI systems, may be better exploitable for sound localization after increasing their perceptual weight via training.
正常听力 (NH) 听众依赖于两个双耳线索,即耳间时间差 (ITD) 和强度差 (ILD),用于方位声音定位。然而,人工耳蜗 (CI) 听众几乎完全依赖于 ILD。原因之一是,目前的临床 CI 刺激策略不传达明显的 ITD 线索。但是,即使在使用研究接口以最佳条件呈现 ITD 时,CI 听众的 ITD 敏感性也低于 NH 听众。由于最近已经表明,当只有一个线索与视觉信息一致时,NH 听众会改变他们的 ITD/ILD 权重,鉴于他们每天都接触到可靠的 ILD 但不可靠的 ITD,这种重新加权可能会增加 CI 听众对 ITD 的低感知贡献。六位双侧 CI 听众完成了为期多天的偏向训练,通过在没有视觉强化的情况下进行 ITD/ILD 权重的预测量和后测量来视觉强化 ITD。使用直接电刺激,我们在单个双耳位置匹配的电极对上呈现 100- 和 300-pps 脉冲串,以各种空间一致和不一致的组合传达 ITD 和 ILD。听众的任务是在虚拟环境中对刺激进行偏向。此外,在训练前后测量了 ITD 和 ILD 阈值。对于 100-pps 刺激,偏向训练略微但显著增加了 ITD 的贡献。训练既不影响阈值,也与权重无关。对于 300-pps 刺激,ITD 权重较低,ITD 阈值较大,但训练没有效果。平均而言,在测试会话中,为包含 ILD 的刺激添加与方位相关的 ITD 会增加两种 100- 和 300-pps 刺激的偏向程度。结果表明,未来的 CI 系统可以更有效地编码低速率的 ITD 线索,通过训练增加其感知权重后,这些线索可能更有利于声音定位。