Department of Otolaryngology, Wexner Medical Center, The Ohio State University, Columbus.
Department of Otorhinolaryngology, The University Medical Center Groningen, University of Groningen, the Netherlands.
J Speech Lang Hear Res. 2021 Feb 17;64(2):683-690. doi: 10.1044/2020_JSLHR-20-00496. Epub 2021 Jan 25.
Purpose This preliminary research examined (a) the perception of two common sources of indexical variability in speech-regional dialects and foreign accents, and (b) the relation between indexical processing and sentence recognition among prelingually deaf, long-term cochlear implant (CI) users and normal-hearing (NH) peers. Method Forty-three prelingually deaf adolescent and adult CI users and 44 NH peers completed a regional dialect categorization task, which consisted of identifying the region of origin of an unfamiliar talker from six dialect regions of the United States. They also completed an intelligibility rating task, which consisted of rating the intelligibility of short sentences produced by native and nonnative (foreign-accented) speakers of American English on a scale from 1 () to 7 (). Individual performance was compared to demographic factors and sentence recognition scores. Results Both CI and NH groups demonstrated difficulty with regional dialect categorization, but NH listeners significantly outperformed the CI users. In the intelligibility rating task, both CI and NH listeners rated foreign-accented sentences as less intelligible than native sentences; however, CI users perceived smaller differences in intelligibility between native and foreign-accented sentences. Sensitivity to accent differences was related to sentence recognition accuracy in CI users. Conclusions Prelingually deaf, long-term CI users are sensitive to accent variability in speech, but less so than NH peers. Additionally, individual differences in CI users' sensitivity to indexical variability was related to sentence recognition abilities, suggesting a common source of difficulty in the perception and encoding of fine acoustic-phonetic details in speech.
目的 本初步研究考察了（a）对言语区域方言和外国口音中两种常见变音指标的感知，以及（b）在聋前、长期使用人工耳蜗（CI）的个体和正常听力（NH）个体之间，变音指标处理与句子识别之间的关系。
方法 43 名聋前、成年人工耳蜗使用者和 44 名 NH 个体完成了一项区域方言分类任务，该任务包括从美国的六个方言区域中识别一个不熟悉说话者的原籍。他们还完成了一个可理解性评分任务，即对美国英语母语者和非母语者（外国口音）所发短句子的可理解度进行评分，分值从 1（）到 7（）。将个体表现与人口统计学因素和句子识别分数进行比较。
结果 人工耳蜗使用者和 NH 个体都难以进行区域方言分类，但 NH 个体的表现明显优于人工耳蜗使用者。在可理解性评分任务中，人工耳蜗使用者和 NH 个体都认为外国口音的句子不如母语句子可理解；然而，人工耳蜗使用者感知到母语和外国口音句子之间的可理解度差异较小。对口音差异的敏感性与人工耳蜗使用者的句子识别准确性相关。
结论 聋前、长期使用人工耳蜗的个体对言语中的口音变化敏感，但不如 NH 个体敏感。此外，人工耳蜗使用者对变音指标变化的敏感性存在个体差异，这与句子识别能力有关，表明在言语中对细微的声学-语音细节的感知和编码存在共同的困难来源。
