Zhang Caicai, Shao Jing, Huang Xunan
Department of Chinese and Bilingual Studies, The Hong Kong Polytechnic University, Hong Kong SAR, China.
Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences, Shenzhen, China.
PLoS One. 2017 Aug 22;12(8):e0183151. doi: 10.1371/journal.pone.0183151. eCollection 2017.
Congenital amusia is a lifelong disorder of fine-grained pitch processing in music and speech. However, it remains unclear whether amusia is a pitch-specific deficit, or whether it affects frequency/spectral processing more broadly, such as the perception of formant frequency in vowels, apart from pitch. In this study, in order to illuminate the scope of the deficits, we compared the performance of 15 Cantonese-speaking amusics and 15 matched controls on the categorical perception of sound continua in four stimulus contexts: lexical tone, pure tone, vowel, and voice onset time (VOT). Whereas lexical tone, pure tone and vowel continua rely on frequency/spectral processing, the VOT continuum depends on duration/temporal processing. We found that the amusic participants performed similarly to controls in all stimulus contexts in the identification, in terms of the across-category boundary location and boundary width. However, the amusic participants performed systematically worse than controls in discriminating stimuli in those three contexts that depended on frequency/spectral processing (lexical tone, pure tone and vowel), whereas they performed normally when discriminating duration differences (VOT). These findings suggest that the deficit of amusia is probably not pitch specific, but affects frequency/spectral processing more broadly. Furthermore, there appeared to be differences in the impairment of frequency/spectral discrimination in speech and nonspeech contexts. The amusic participants exhibited less benefit in between-category discriminations than controls in speech contexts (lexical tone and vowel), suggesting reduced categorical perception; on the other hand, they performed inferiorly compared to controls across the board regardless of between- and within-category discriminations in nonspeech contexts (pure tone), suggesting impaired general auditory processing. These differences imply that the frequency/spectral-processing deficit might be manifested differentially in speech and nonspeech contexts in amusics-it is manifested as a deficit of higher-level phonological processing in speech sounds, and as a deficit of lower-level auditory processing in nonspeech sounds.
先天性失歌症是一种终生存在的、在音乐和言语中对音高进行精细处理的障碍。然而，目前尚不清楚失歌症是否只是特定于音高的缺陷，还是更广泛地影响频率/频谱处理，比如除了音高之外，对元音中元音共振峰频率的感知。在本研究中，为了阐明缺陷的范围，我们比较了15名说粤语的失歌症患者和15名匹配的对照组在四种刺激情境下对声音连续体的范畴知觉表现：声调、纯音、元音和语音起始时间（VOT）。声调、纯音和元音连续体依赖于频率/频谱处理，而VOT连续体则依赖于时长/时间处理。我们发现，在识别方面，就跨范畴边界位置和边界宽度而言，失歌症参与者在所有刺激情境下的表现与对照组相似。然而，在依赖频率/频谱处理的三种情境（声调、纯音和元音）中，失歌症参与者在辨别刺激时的表现系统地比对照组差，而在辨别时长差异（VOT）时表现正常。这些发现表明，失歌症的缺陷可能并非特定于音高，而是更广泛地影响频率/频谱处理。此外，在言语和非言语情境中，频率/频谱辨别受损似乎存在差异。在言语情境（声调元音）中，失歌症参与者在范畴间辨别中比对照组受益更少，表明范畴知觉降低；另一方面，在非言语情境（纯音）中，无论范畴间还是范畴内辨别，他们的表现总体上都比对照组差，表明一般听觉处理受损。这些差异意味着，频率/频谱处理缺陷在失歌症患者的言语和非言语情境中可能有不同表现——在言语声音中表现为高级语音处理缺陷,在非言语声音中表现为低级听觉处理缺陷。
