1Department of Linguistics, CLiPS Computational Linguistics and Psycholinguistics Research Centre, University of Antwerp, Antwerp, Belgium; and 2Division of Language and Communication Science, City University London, London, United Kingdom.
Ear Hear. 2017 Jul/Aug;38(4):475-486. doi: 10.1097/AUD.0000000000000406.
This longitudinal study examined the effect of emerging vocabulary production on the ability to produce the phonetic cues to prosodic prominence in babbled and lexical disyllables of infants with cochlear implants (CI) and normally hearing (NH) infants. Current research on typical language acquisition emphasizes the importance of vocabulary development for phonological and phonetic acquisition. Children with CI experience significant difficulties with the perception and production of prosody, and the role of possible top-down effects is, therefore, particularly relevant for this population.
Isolated disyllabic babble and first words were identified and segmented in longitudinal audio-video recordings and transcriptions for nine NH infants and nine infants with CI interacting with their parents. Monthly recordings were included from the onset of babbling until children had reached a cumulative vocabulary of 200 words. Three cues to prosodic prominence, fundamental frequency (f0), intensity, and duration, were measured in the vocalic portions of stand-alone disyllables. To represent the degree of prosodic differentiation between two syllables in an utterance, the raw values for intensity and duration were transformed to ratios, and for f0, a measure of the perceptual distance in semitones was derived. The degree of prosodic differentiation for disyllabic babble and words for each cue was compared between groups. In addition, group and individual tendencies on the types of stress patterns for babble and words were also examined.
The CI group had overall smaller pitch and intensity distances than the NH group. For the NH group, words had greater pitch and intensity distances than babbled disyllables. Especially for pitch distance, this was accompanied by a shift toward a more clearly expressed stress pattern that reflected the influence of the ambient language. For the CI group, the same expansion in words did not take place for pitch. For intensity, the CI group gave evidence of some increase of prosodic differentiation. The results for the duration measure showed evidence of utterance final lengthening in both groups. In words, the CI group significantly reduced durational differences between syllables so that a more even-timed, less differentiated pattern emerged.
The onset of vocabulary production did not have the same facilitatory effect for the CI infants on the production of phonetic cues for prosody, especially for pitch. It was argued that the results for duration may reflect greater articulatory difficulties in words for the CI group than the NH group. It was suggested that the lack of clear top-down effects of the vocabulary in the CI group may be because of a lag in development caused by an initial lack of auditory stimulation, possibly compounded by the absence of auditory feedback during the babble phase.
本纵向研究考察了新词汇生成对人工耳蜗植入（CI）婴儿和正常听力（NH）婴儿在咿呀学语和词汇双音节中产生韵律突出的语音线索的能力的影响。当前典型语言习得的研究强调了词汇发展对语音和语音习得的重要性。CI 儿童在感知和产生韵律方面存在显著困难，因此，可能的自上而下效应的作用对于这一人群尤为相关。
在纵向的音频-视频记录和转写中，确定并分段了 9 名 NH 婴儿和 9 名与父母互动的 CI 婴儿的孤立双音节咿呀学语和第一语言。从咿呀学语开始，每月进行一次录音，直到儿童的词汇量累计达到 200 个单词。在孤立的双音节元音部分测量了三个韵律突出的线索，即基频（f0）、强度和时长。为了表示一个音节在一个音节中的韵律差异程度，强度和时长的原始值被转换为比值，而 f0 则得出了半音感知距离的度量。比较了两组在每个线索的双音节咿呀学语和单词的韵律差异程度。此外，还检查了两组在咿呀学语和单词的重音模式类型上的群体和个体趋势。
CI 组的音高和强度距离总体上小于 NH 组。对于 NH 组，单词的音高和强度距离大于咿呀学语的双音节。特别是对于音高距离，这伴随着一种更清晰表达的重音模式的转变，反映了环境语言的影响。对于 CI 组，音高并没有出现同样的扩展。对于强度，CI 组显示出一些韵律差异增加的证据。时长测量的结果表明，两组都有话语结尾延长的迹象。在单词中，CI 组显著减小了音节之间的时长差异，从而出现了更均匀的、差异较小的模式。
词汇生成的开始并没有对 CI 婴儿产生同样的促进作用，使他们在产生韵律的语音线索方面，特别是在音高方面，表现得更好。有人认为，CI 组的时长结果可能反映出与 NH 组相比，CI 组在单词中的发音更困难。有人认为，CI 组缺乏词汇的清晰自上而下的影响可能是由于初始缺乏听觉刺激导致的发育滞后，这可能因在咿呀学语阶段缺乏听觉反馈而变得更复杂。
