Dynamic sounds and perceptual processes: Exploring the role of amplitude envelope in auditory perception

口技表演者经常利用知觉系统整合听觉和视觉信息的倾向，利用它来诱骗观众“听到”木偶说话。我已经证明，音乐家使用一种类似的方法来欺骗观众，让他们听到比他们的声学时长更长或更短的音符。这种使用自然音乐声音的新奇错觉与早期的研究不一致，早期的研究表明视觉不会影响对持续时间的听觉感知。在我目前的发现资助过程中，我的团队已经确定，这种差异在很大程度上源于一种被称为振幅包络的声学特性--声音的“随时间变化的形状”。 听觉实验经常依赖于具有“平坦”幅度包络的声音--表现出较长的持续期和/或突然的偏移。这些声音既负担得起也有必要使用自然声音所不具备的倾听策略。例如，虽然视觉不能与呈现平坦(即传统)振幅包络的声音相结合，但它确实与呈现动态衰减(即自然)振幅包络的声音相结合--例如我在前面描述的错觉中使用的那些。受到这一发现的吸引，我的团队开始对过去的听觉研究中使用的声音进行正式调查。到目前为止，我们已经记录了对没有自然幅度包络的声音的强烈偏见-大约93%的刺激遇到。总而言之，这些发现表明，有必要探索从人工音调实验中得出的理论和预测在多大程度上适用于实验室外听到的声音。 通过提供对听觉处理核心问题的新见解，本研究计划将为各种相关领域提供信息。神经科学家正在越来越多地探索多感觉感知的生物学基础，临床医生目前正在努力更好地了解典型整合过程失败的情况--这是许多自闭症谱系障碍患者的共同问题。据估计，这种具有挑战性的情况直接影响到50万加拿大人--更不用说他们的照顾者、家庭成员、教师和其他帮助他们的人了。这项提议还将有助于在医疗设备中设计有效的听觉人机界面。目前的行业标准要求这些警报使用具有平坦幅度包络的音调，从而导致无效和令人不快的交互。我的团队的研究将探索自然幅度包络的使用如何潜在地导致警报效果的改善，并澄清我们对涉及声音的联想记忆的基本理解。这具有重要的经济影响，因为加拿大工业部评估我们的医疗器械市场每年价值64亿美元。

Ventriloquists routinely exploit the perceptual system’s propensity to integrate auditory and visual information, using it to trick audiences into “hearing” puppets speak. I have shown that musicians use a parallel approach to trick audiences into hearing musical notes as longer or shorter than their acoustic duration. This novel illusion using natural musical sounds is at odds with earlier research suggesting that vision does not influence auditory perception of duration. Over the course of my current Discovery Grant, my team has established that this discrepancy stems in large part from an acoustic property known as amplitude envelope - a sound’s “shape over time.” Auditory experiments frequently rely on sounds with “flat” amplitude envelopes - exhibiting long sustain periods and/or abrupt offsets. These sounds both afford and necessitate listening strategies unavailable for natural sounds. For example, although vision does not integrate with sounds exhibiting flat (i.e. traditional) amplitude envelopes, it does integrate with sounds exhibiting dynamically decaying (i.e. natural) amplitude envelopes - such as those used in my previously described illusion. Intrigued by this discovery, my team began a formal survey of sounds used in past auditory research. To date we have documented a strong bias towards sounds lacking natural amplitude envelopes - approximately 93% of stimuli encountered. Together, these findings demonstrate a need for exploring the degree to which theories and predictions derived from experiments with artificial tones generalize to sounds heard outside the laboratory. By providing novel insight into core issues of auditory processing, this program of research will inform a variety of cognate fields. Neuroscientists are increasingly exploring the biological basis of multi-sensory perception, and clinicians are currently working to better understand situations in which typical integration processes fail - a common issue for many individuals with autism spectrum disorder. This challenging condition is estimated to directly affect 500,000 Canadians - not to mention their caretakers, family members, teachers, and others working to assist them. This proposal will also aid efforts to design effective auditory human-computer-interfaces in medical devices. Industry standards currently mandate these alarms use tones with flat amplitude envelopes, resulting in ineffective and unpleasant interactions. My team’s research will explore how the use of natural amplitude envelopes could potentially lead to improvements in alarm efficacy as well as clarify our basic understanding of associative memory involving sound. This holds important economic implications, as Industry Canada evaluates our medical device market at $6.4 billion annually.