Acoustic cues in auditory pattern analysis

描述（由申请人提供）：本提案的总体目标是识别和表征听觉过程，这些听觉过程限制了对不同水平刺激的频谱线索的检测和区分，包括听力正常和听力受损的听众。目的1使用剖面分析的心理声学范式检验刺激水平波动对频谱形状判别的影响。为了限制绝对水平线索的使用，通常在光谱形状判别的心理物理测试中引入水平线索，但初步数据表明，这种操作在1 kHz以上比在1 kHz以下具有更大的效果。这样的结果将对我们所知道的光谱形状的敏感性有重要的意义。目的2关注轮廓分析与元音感知之间的关系，包括水平波动对元音辨别的潜在影响。这一目的的实验将阐明这两种通常被认为密切相关的范式之间的联系，以及频谱形状识别的心理物理数据在多大程度上为我们对言语感知的理解提供信息。目的3探讨时间精细结构在蒙面语音感知中的作用，重点研究波动噪声下的语音识别。虽然最近的文献研究支持了时间精细结构在调制噪声中的听力中起关键作用的观点，但拟议的实验引入了控制条件并测试了尚未经过严格评估的合理替代假设。这个问题是特别相关的考虑到最近的兴趣发展人工耳蜗技术，保留良好的时间线索。目标4发展了新的线索范式，有可能为我们从波动背景中提取信号的能力提供新的见解。这种方法的一个目标是更好地理解观察掩蔽释放所需的先决条件。预计在这四个目标中提出的工作将澄清随机水平波动在自然声音中存在的频谱线索处理中的作用，例如背景噪声中潜在的语音识别。

DESCRIPTION (provided by applicant): The overarching goal of this proposal is to identify and characterize the auditory processes that limit detection and discrimination of spectral cues for stimuli that vary in level, including both normal- hearing and hearing-impaired listeners. Aim 1 examines the effect of stimulus level fluctuation on spectral shape discrimination using the psychoacoustic paradigm of profile analysis. Level rove is often introduced in psychophysical tests of spectral shape discrimination in order to limit the use of absolute level cues, but preliminary data indicate that this manipulation has greater effect above than below 1 kHz. Such a result would have important implications for what we think we know about sensitivity for spectral shape. Aim 2 concerns the relationship between profile analysis and vowel perception, including potential effects of level fluctuation on vowel discrimination. Experiments in this aim will clarify the association between these two paradigms, often assumed to be closely related, and the extent to which psychophysical data on spectral shape discrimination inform our understanding of speech perception. Aim 3 explores the role of temporal fine-structure in masked speech perception, with emphasis on speech recognition in fluctuating noise. Whereas recent studies in the literature provide support for the idea that temporal fine-structure plays a key role in hearing in modulated noise, the proposed experiments introduce control conditions and test plausible alternative hypotheses that have not yet been rigorously evaluated. This issue is particularly pertinent given recent interest in developing cochlear implant technology that preserves fine timing cues. Aim 4 develops novel cuing paradigms that have the potential to provide new insights about our ability to extract signals from fluctuating backgrounds. One goal of this approach is to develop a greater understanding of the preconditions necessary for observing masking release. It is anticipated that the work proposed in these four aims will clarify the role of random level fluctuation in the processing of spectral cues present in natural sounds, such as those underlying speech recognition in background noise.