The impact of auditory feedback on error monitoring and phonetic category representation in a second language

外语学习者（L2）经常保留发音错误，即使在大量的目标语言经验之后。拟议的项目旨在解释一个明显的悖论，即学习者往往能够察觉到同伴的发音错误，但他们无法纠正自己的生产相同的错误。虽然大量的研究关注的是社会经济因素影响学习L2的次优成绩，涉及语音处理的一般机制的认知因素仍然广泛未被探索。没有一个L2的声音采集的现有模型足以占一个持久的感知生产不对称的L2。拟议的项目将通过评估这种不对称的两个潜在原因来缩小这一差距：（1）使用听觉反馈进行错误监测的缺陷，以及（2）语音类别表征的性质。(1)本文将讨论第二语言学习者是否因为没有意识到自己的错误而在纠正他们的产出时遇到困难。它将被测试是否有更多的困难，发现和处理错误时，听自己的声音相比，其他学习者的声音。(2)将决定某些发音错误的经验如何影响语音类别和词汇表征之间的联系。它还将评估是否新建立的L2类别，不存在于母语（L1）或L1类别，不存在于L2，但影响灵活性的看法和生产在这两种语言。方法上的创新是测试在线改变听觉反馈下的第二语言产出。也就是说，通过信号处理技术，参与者将获得与其实际生产不匹配的反馈信号。由此产生的补偿性清晰度将说明错误监测以及类别表征的性质，因为这种方法对L1中语音类别的分布敏感-并且如将在这里确定的-潜在地也在L2中。该项目的长期意义在于，它有助于认知神经科学和言语处理中关于言语产生和感知关系的中心辩论。它将揭示听觉反馈和错误监控在二语加工中的作用。此外，它有可能为语言教学策略提供信息。

Learners of a foreign language (L2) often retain pronunciation errors even after extensive experience with the target language. The proposed project aims at explaining the apparent paradox that, frequently, learners are able to perceive pronunciation errors of fellow learners, yet they are unable to correct the same errors in their own production. Although a large amount of research is concerned with socio-economic factors affecting suboptimal attainment in learning an L2, cognitive factors involving general mechanisms of speech processing remain widely unexplored. None of the current models of L2 sound acquisition sufficiently accounts for a persistent perception-production asymmetry in the L2. The proposed project will close this gap by assessing two potential causes for this asymmetry: (1) deficits in using auditory feedback for error monitoring, and (2) the nature of phonetic category representations. (1) will address whether L2 learners have difficulties in correcting their productions because they are not aware of their own errors. It will be tested whether listeners have more trouble detecting and processing errors when listening to their own voice compared to other learners' voices. (2) will determine how experience with certain mispronunciations affects the association between phonetic categories and lexical representations. It will also assess whether newly established L2 categories that are not present in the native language (L1) or L1 categories that are not present in the L2, nevertheless affect flexibility perception and production in both languages. The methodological innovation here is to test L2 production under online-altered auditory feedback. That is, participants will be presented with a feedback signal that, through signal-processing techniques, mismatches their actual production. Resulting compensatory articulation will speak to error monitoring as well as to the nature of category representations as this method is sensitive to the distribution of phonetic categories in L1 - and as will be determined here - potentially also in L2. The long-term significance of the project is that it contributes to a central debate in cognitive neuroscience and speech processing about the relation of speech production and perception. It will reveal the workings of auditory feedback and error monitoring in L2 processing. In addition, it has the potential to inform strategies in language teaching.