A study on dynamic information processing system in the hierarchical structure of the hearing sense

本研究假设听觉的层次结构由听觉硬件、听觉软件以及知识和意识数据库组成。听觉硬件和软件分别模拟听觉外周和中枢神经系统的功能。这项研究的目的是描述层次结构中的动态信息处理。1997财年，动态信息处理系统被提出。进行了心理声学实验，由受试者对不同调幅声音之间的差异性进行评级。ABS(综合分析)分析表明，差动系统和/或调制滤波器组的特性是决定硬件性能的主要因素。在软件方面，声音之间的欧几里得距离似乎是根据听觉神经系统的时频刺激模式来计算的。在1998财年，我们研究了知识和意识数据库对动态处理的影响。采用SD方法在以下三种条件下进行了环境声心理声学评价实验：(1)无任何信息；(2)有声源上的语言信息；(3)有视觉信息。对实验结果进行因子分析，得出了每个实验的稳定的七个因素，尽管实验条件不同。另一方面，语言和视觉信息对声音知觉的影响表现为声音因素得分的剧烈变化。因此，这些跨通道实验表明了数据库对声音知觉的影响。

A hierarchical structure of the hearing sense, which consists of auditory hardware, auditory software, and knowledge and consciousness database, is assumed in this study. Auditory hardware and software simulate the functions of the auditory periphery and the central nerves system, respectively. The aim of this study is to describe the dynamic information processing in the hierarchy.In FY1997, the dynamic information processing system was addressed. Psychoacoustical experiments were carried out in which dissimilarities among various amplitude-modulated sounds were rated by subjects. An ABS (Analysis By Synthesis) analysis revealed that the characteristics of the differential system and/or the modulation filter bank are dominant to determine the property of the hardware. As to the software, Euclidean distances among the sounds seem to be calculated for the time-frequency excitation patterns in the auditory nerves system.In FY1998, the effects of the knowledge and consciousness database on the dynamic processing were examined. Psychoacoustical experiments to evaluate environmental sounds were conducted by the method of SD under the following three conditions : (1) without any information, (2) with the verbal information on the sound sources, and (3) with the visual information. The results were analyzed by factor analysis, and the stable seven-factors were derived for each experiment in spite of the differences in the experimental conditions. On the other hand, the effects of the verbal and visual information on the perception of the sounds were shown as the drastic changes in factor scores for the sounds. Consequently, the effects of the database on the sound perception were indicated by these experiments with cross modality.