Yamagishi Shimpei, Otsuka Sho, Furukawa Shigeto, Kashino Makio
Interdisciplinary Graduate School of Science and Engineering, Tokyo Institute of Technology, Yokohama, Kanagawa, 226-8503, Japan; Reseach Fellow of Japan Society for the Promotion of Science, Japan.
NTT Communication Science Laboratories, NTT Corporation, 3-1 Morinosato Wakamiya, Atsugi, Kanagawa, 243-0198, Japan.
Hear Res. 2017 Jul;350:244-250. doi: 10.1016/j.heares.2017.03.006. Epub 2017 Mar 18.
The two-tone sequence (ABA_), which comprises two different sounds (A and B) and a silent gap, has been used to investigate how the auditory system organizes sequential sounds depending on various stimulus conditions or brain states. Auditory streaming can be evoked by differences not only in the tone frequency ("spectral cue": ΔF, TONE condition) but also in the amplitude modulation rate ("AM cue": ΔF, AM condition). The aim of the present study was to explore the relationship between the perceptual properties of auditory streaming for the TONE and AM conditions. A sequence with a long duration (400 repetitions of ABA_) was used to examine the property of the bistability of streaming. The ratio of feature differences that evoked an equivalent probability of the segregated percept was close to the ratio of the Q-values of the auditory and modulation filters, consistent with a "channeling theory" of auditory streaming. On the other hand, for values of ΔF and ΔF evoking equal probabilities of the segregated percept, the number of perceptual switches was larger for the TONE condition than for the AM condition, indicating that the mechanism(s) that determine the bistability of auditory streaming are different between or sensitive to the two domains. Nevertheless, the number of switches for individual listeners was positively correlated between the spectral and AM domains. The results suggest a possibility that the neural substrates for spectral and AM processes share a common switching mechanism but differ in location and/or in the properties of neural activity or the strength of internal noise at each level.
双音序列（ABA_）由两种不同的声音（A和B）以及一个静音间隙组成，已被用于研究听觉系统如何根据各种刺激条件或脑状态来组织连续声音。听觉流不仅可以由音调频率的差异（“频谱线索”：ΔF，音调条件）引起，也可以由调幅率的差异（“调幅线索”：ΔF，调幅条件）引起。本研究的目的是探讨音调条件和调幅条件下听觉流的感知特性之间的关系。使用一个持续时间较长的序列（ABA_重复400次）来检验流的双稳性特性。引发分离感知的等效概率的特征差异比率接近于听觉滤波器和调制滤波器的Q值比率，这与听觉流的“通道理论”一致。另一方面，对于引发分离感知的相等概率的ΔF和ΔF值，音调条件下的感知切换次数比调幅条件下的多，这表明决定听觉流双稳性的机制在这两个领域之间不同或对其敏感。然而，个体听众在频谱域和调幅域之间的切换次数呈正相关。结果表明，频谱和调幅过程的神经基质可能共享一种共同的切换机制，但在位置和/或各级神经活动特性或内部噪声强度方面存在差异。
