Challenging the notion of innate phonetic boundaries.

Challenging the notion of innate phonetic boundaries.
复制标题

DOI:
10.1121/1.1379078
复制
发表时间:
2001-08
期刊:
The Journal of the Acoustical Society of America
影响因子:
--
通讯作者:
S. Nittrouer
S. Nittrouer
中科院分区:
其他
文献类型:
--
作者:
S. Nittrouer

文献摘要

被引文献

相似文献

许多关于婴儿语音感知能力的研究表明,这些年轻的听者可以获得语音信号中的声学细节。由于这些研究使用的刺激可以用成人定义的语音类别来描述,因此作者得出结论,婴儿天生就能像成人一样将刺激识别为这些类别的成员。事实上,目前关于言语感知的主流观点认为,婴儿生来就对语音边界的普遍设置敏感,并且那些由环境语言支持的边界得以维持,而那些不受环境语言支持的边界则消失了。在这项研究中,46名婴儿和75名3岁儿童的辨别能力进行了测试,几个语音对比发生在他们的母语,使用自然和合成语音。能够区分任何特定对比的儿童的比例在不同的对比中有所不同,并且没有一个对比被任何接近所有儿童的东西所区分。虽然这些结果与其他人报告的结果没有不同,但这里的解释是,我们应该重新考虑先天语音类别和/或边界的概念。此外,自然语音和合成语音的成功率没有差异,因此一个次要的结论是,在语音实验中使用合成刺激不会对儿童产生不利影响。
Numerous studies of infants' speech perception abilities have demonstrated that these young listeners have access to acoustic detail in the speech signal. Because these studies have used stimuli that could be described in terms of adult-defined phonetic categories, authors have concluded that infants innately recognize stimuli as members of these categories, as adults do. In fact, the predominant, current view of speech perception holds that infants are born with sensitivities for the universal set of phonetic boundaries, and that those boundaries supported by the ambient language are maintained, while those not supported by the ambient language dissolve. In this study, discrimination abilities of 46 infants and 75 3-year-olds were measured for several phonetic contrasts occurring in their native language, using natural and synthetic speech. The proportion of children who were able to discriminate any given contrast varied across contrasts, and no one contrast was discriminated by anything close to all of the children. While these results did not differ from those reported by others, the interpretation here is that we should reconsider the notion of innate phonetic categories and/or boundaries. Moreover, success rates did not differ for natural and synthetic speech, and so a minor conclusion was that children are not adversely affected by the use of synthetic stimuli in speech experiments.