DISCRIMINATION AND IDENTIFICATION OF AUDITORY PATTERNS

听觉模式的辨别和识别

• 批准号: 6093556
• 负责人: CHARLES S WATSON
• 金额: $ 6.73万
• 依托单位: TRUSTEES OF INDIANA UNIVERSITY
• 依托单位国家: 美国
• 财政年份: 1983
• 资助国家: 美国
• 起止时间: 1983-08-01 至 1999-09-29
• 项目状态: 已结题
• 来源: https://reporter.nih.gov/project-details/6093556
• 关键词: auditory discrimination; auditory stimulus; auditory threshold; behavioral /social science research tag; form /pattern perception; hearing tests; human subject; mathematical model; psychoacoustics; psychological models; sound frequency

The remarkable ability of the human auditory system to extract meaning from continuous human speech can be more fully understood in relation to listeners' abilities to "hear out" the details of complex sounds in general. The proposal research will determine the limits of auditory discrimination and identification abilities of normal listeners for a variety of complex sounds. These experiments first extend earlier work with tonal patterns and develop a more recent line of investigation with spectrally complex sounds. The proposed studies with tonal patterns are divided into three categories, each addressing a basic issue in the discrimination or identification of complex sounds. These are 1) the ability to discriminate between complex patterns with various types of spectral and temporal microstructure; 2) the ability to recognize relational properties that form the invariant that define a pattern under various types of transformations; and 3) the ability to resolve pattern detail under different levels of pattern uncertainty (or familiarity). A second series of experiments will investigate the generality of the proportion-of-the-total-duration (PTD) rule, which has been shown to be a remarkably strong predictor of the deductibility of changes in components of a variety of unfamiliar complex sounds. Two recent theoretical model for the discrimination of complex sounds, one based on informational content and the other on the physical properties of the stimulus are able t o predict the general form of the PTD-rule results. Experiments will be conducted to determine which of these models provides the more comprehensive account of the ability to resolve the details of complex sounds. A third series of experiments will determine the abilities of listeners to resolve the details of dynamically varying, spectrally complex sounds, using non-speech stimuli that include some of the properties of speech. These studies will investigate the effects of various types of spectral complexity, including a new class of experiments on listeners' abilities to recognize the characteristics of complex filters. An improved version of the Test of Basic Auditory Capabilities will also be developed.

人类听觉系统提取意义的非凡能力 从连续的人类语言中可以更充分地理解， 听众“听出”复杂声音细节的能力， 将军 建议研究将确定听觉的极限 听力正常者的辨别和识别能力 各种复杂的声音。 这些实验首先扩展了早期的工作 与音调模式，并开发一个更近的调查线， 光谱复杂的声音 建议的音调模式研究是 分为三类，每一类都涉及一个基本问题， 辨别或识别复杂的声音。 这些是（1） 区分具有各种类型的复杂模式的能力 光谱和时间微观结构; 2）识别能力 构成定义模式的不变式的关系属性， 各种类型的转换;和3）解决模式的能力 在不同的模式不确定性（或熟悉程度）下的细节。 第二系列实验将研究 总持续时间比例（PTD）规则，已被证明是 一个非常强的预测变化的扣除， 各种不熟悉的复杂声音的组成部分。 最近的两 复杂声音辨别的理论模型，一个基于 信息内容和另一个对物理性能的 刺激能够预测PTD规则结果的一般形式。 将进行实验，以确定这些模型中的哪一个提供 更全面的说明解决细节的能力， 复杂的声音 第三个系列的实验将确定 收听者解析动态变化的细节的能力， 频谱复杂的声音，使用非语音刺激，包括一些 言语的特性。 这些研究将调查 各种类型的频谱复杂性，包括一类新的 听力实验的特点， 复合过滤器 听力基础测验的改进 能力也将得到发展。