Fishman Yonatan I, Kim Mimi, Steinschneider Mitchell
Departments of Neurology and Neuroscience, and
Epidemiology and Population Health, Albert Einstein College of Medicine, Bronx, New York 10461.
J Neurosci. 2017 Nov 1;37(44):10645-10655. doi: 10.1523/JNEUROSCI.0792-17.2017. Epub 2017 Sep 27.
An important aspect of auditory scene analysis is auditory stream segregation-the organization of sound sequences into perceptual streams reflecting different sound sources in the environment. Several models have been proposed to account for stream segregation. According to the "population separation" (PS) model, alternating ABAB tone sequences are perceived as a single stream or as two separate streams when "A" and "B" tones activate the same or distinct frequency-tuned neuronal populations in primary auditory cortex (A1), respectively. A crucial test of the PS model is whether it can account for the observation that A and B tones are generally perceived as a single stream when presented synchronously, rather than in an alternating pattern, even if they are widely separated in frequency. Here, we tested the PS model by recording neural responses to alternating (ALT) and synchronous (SYNC) tone sequences in A1 of male macaques. Consistent with predictions of the PS model, a greater effective tonotopic separation of A and B tone responses was observed under ALT than under SYNC conditions, thus paralleling the perceptual organization of the sequences. While other models of stream segregation, such as temporal coherence, are not excluded by the present findings, we conclude that PS is sufficient to account for the perceptual organization of ALT and SYNC sequences and thus remains a viable model of auditory stream segregation. According to the population separation (PS) model of auditory stream segregation, sounds that activate the same or separate neural populations in primary auditory cortex (A1) are perceived as one or two streams, respectively. It is unclear, however, whether the PS model can account for the perception of sounds as a single stream when they are presented synchronously. Here, we tested the PS model by recording neural responses to alternating (ALT) and synchronous (SYNC) tone sequences in macaque A1. A greater effective separation of tonotopic activity patterns was observed under ALT than under SYNC conditions, thus paralleling the perceptual organization of the sequences. Based on these findings, we conclude that PS remains a plausible neurophysiological model of auditory stream segregation.
听觉场景分析的一个重要方面是听觉流分离——将声音序列组织成反映环境中不同声源的感知流。已经提出了几种模型来解释流分离。根据“群体分离”（PS）模型，当“A”和“B”音调分别激活初级听觉皮层（A1）中相同或不同频率调谐的神经元群体时，交替的ABAB音调序列被感知为单个流或两个单独的流。PS模型的一个关键测试是，即使A和B音调在频率上广泛分离，当它们同步呈现而不是交替模式呈现时，它是否能够解释A和B音调通常被感知为单个流的观察结果。在这里，我们通过记录雄性猕猴A1中对交替（ALT）和同步（SYNC）音调序列的神经反应来测试PS模型。与PS模型的预测一致，在ALT条件下观察到的A和B音调反应的有效音调拓扑分离比在SYNC条件下更大，从而与序列的感知组织平行。虽然其他流分离模型，如时间连贯性，并不被本研究结果排除，但我们得出结论，PS足以解释ALT和SYNC序列的感知组织，因此仍然是听觉流分离的一个可行模型。根据听觉流分离的群体分离（PS）模型，在初级听觉皮层（A1）中激活相同或不同神经群体的声音分别被感知为一个或两个流。然而，尚不清楚PS模型是否能够解释声音在同步呈现时被感知为单个流的情况。在这里，我们通过记录猕猴A1中对交替（ALT）和同步（SYNC）音调序列的神经反应来测试PS模型。在ALT条件下观察到的音调拓扑活动模式的有效分离比在SYNC条件下更大，从而与序列的感知组织平行。基于这些发现，我们得出结论，PS仍然是听觉流分离的一个合理的神经生理学模型。
