Low-level information and high-level perception: the case of speech in noise.
Low-level information and high-level perception: the case of speech in noise.
复制标题
DOI：
10.1371/journal.pbio.0060126
复制
发表时间：
2008-05-20
期刊：
影响因子：
9.8
通讯作者：
Ahissar M
中科院分区：
文献类型：
--
作者：
Nahum M;Nelken I;Ahissar M
听觉信息以从精细到粗略的分层方案进行处理，从低级声学信息到高级抽象表示（例如语音标签）。我们现在要问的是，精细的声学信息（未在高水平上保留）是否仍可用于从噪声中提取语音。先前的理论表明，要么是低级信息的完全可用性，要么是受任务难度限制的可用性。我们提出了第三种替代方案，基于反向层次理论（RHT），该理论最初是为了描述处理层次和视觉感知之间的关系而得出的。 RHT 断言只有层次结构的较高级别才能立即被感知。直接获取低层信息需要特定的条件，并且只能以并发理解为代价才能实现。我们在一系列实验中测试了这三种视图的预测，在这些实验中，我们测量了利用低级双耳信息进行语音感知的好处，并将其与早期听觉系统模型的预测进行了比较。只有听觉 RHT 可以解释结果的完整模式，这表明类似的默认和权衡是视觉和听觉模态中的分层处理和感知之间关系的基础。感觉神经科学的核心问题之一是确定我们大脑中编码的最大数量的任务相关信息。人们通常假设所有这些信息都可用于做出感知决策。现在我们证明这个假设并不普遍成立。我们发现，当辨别或理解被噪声掩盖的语音时，通常只有高层皮质区域所代表的信息才能被感知。因此，当我们需要决定说话者是说“白天”还是“夜晚”时，我们很可能会成功进行这种区分。然而，当需要精细区分时（例如，“day”与“bay”），区分这两个词所必需的有关精细光谱和时间细节的信息只有在特殊条件下才能充分利用。这些条件包括，例如，刺激的系统重复，如心理声学实验中经常进行的那样，或者当人们消除理解的需要并专注于纯粹的识别时。这些条件是不生态的，并且在大多数日常情况下是无法承受的。准确感知和理解语音的能力取决于听觉处理的声学到语音层次结构中的信息丢失。大脑分层处理的一般理论预测了我们能够或不能同时做到这两点的条件。
Auditory information is processed in a fine-to-crude hierarchical scheme, from low-level acoustic information to high-level abstract representations, such as phonological labels. We now ask whether fine acoustic information, which is not retained at high levels, can still be used to extract speech from noise. Previous theories suggested either full availability of low-level information or availability that is limited by task difficulty. We propose a third alternative, based on the Reverse Hierarchy Theory (RHT), originally derived to describe the relations between the processing hierarchy and visual perception. RHT asserts that only the higher levels of the hierarchy are immediately available for perception. Direct access to low-level information requires specific conditions, and can be achieved only at the cost of concurrent comprehension. We tested the predictions of these three views in a series of experiments in which we measured the benefits from utilizing low-level binaural information for speech perception, and compared it to that predicted from a model of the early auditory system. Only auditory RHT could account for the full pattern of the results, suggesting that similar defaults and tradeoffs underlie the relations between hierarchical processing and perception in the visual and auditory modalities. One of the central questions in sensory neuroscience is the determination of the maximal amount of task-relevant information that is encoded in our brain. It is often assumed that all of this information is available for making perceptual decisions. We now show that this assumption does not hold generally. We find that when discriminating or understanding speech masked by noise, only the information that is represented at higher cortical areas is generally accessible for perception. Thus, when we need to decide whether the speaker said “day” or “night,” we are likely to succeed in this discrimination. However, when fine discriminations are required (e.g., “day” vs. “bay”), the information regarding the fine spectral and temporal details, which are necessary to discriminate these two words, can be fully utilized only under special conditions. These conditions include, for example, systematic repetitions of the stimuli, as often done in psychoacoustic experiments, or when one eliminates the need for comprehension and focuses on mere identification. These conditions are nonecological, and are not afforded in most daily situations. The ability to both perceive and understand speech accurately depends on the loss of information along the acoustic-to-phonological hierarchy of auditory processing. A general theory of hierarchical processing in the brain predicts the conditions under which we can or cannot do both.
登录
查看更多内容
影响因子：
64.8
作者：
Ahissar, M;Hochstein, S
通讯作者：
Hochstein, S
影响因子：
2.4
作者：
Bernstein, Leslie R.;Trahiotis, Constantine;Freyman, Richard L.
通讯作者：
Freyman, Richard L.
影响因子：
2.4
作者：
Bernstein, LR;van de Par, S;Trahiotis, C
通讯作者：
Trahiotis, C
影响因子：
2.5
作者：
Batra, R;Kuwada, S;Fitzpatrick, DC
通讯作者：
Fitzpatrick, DC
DOI：
10.1037/0096-1523.25.3.617
发表时间：
1999-06-01
影响因子：
2.1
作者：
Darwin, CJ;Hukin, RW
通讯作者：
Hukin, RW