Top-down control of selective amplification

选择性扩增的自上而下控制

• 批准号: 9023532
• 负责人: Gerald, Jr Kidd
• 金额: $ 54.2万
• 依托单位: BOSTON UNIVERSITY (CHARLES RIVER CAMPUS)
• 依托单位国家: 美国
• 财政年份: 2014
• 资助国家: 美国
• 起止时间: 2014-04-01 至 2019-03-31
• 项目状态: 已结题
• 来源: https://reporter.nih.gov/project-details/9023532
• 关键词: Accounting; Acoustics; Attention; Auditory; Awareness; Binaural; Brain; Communication; Complex; Cues; Devices; Environment; Eye; Goals; Health; Hearing; Hearing Aids; Human; Hybrids; Judgment; Lead; Masks; Modification; Monitor; Performance; Persons; Process; Public Health; Sensorineural Hearing Loss; Series; Source; Speech; Stimulus; Task Performances; Testing; Work; attentional control; base; cost; design; experience; gaze; hearing impairment; impression; improved; knowledge translation; novel; novel strategies; prototype; research study; sound; speech recognition; success

DESCRIPTION (provided by applicant): In realistic, everyday acoustic environments human listeners - regardless of whether they have normal or impaired hearing - depend on the considerable processing power of the human brain to evaluate the "auditory scene". The various sources of sound must be parsed, evaluated for relevance and significance, and then attention must be directed to the desired source (the "target") while interfering sources (the "maskers") are discarded and ignored. Listeners with sensorineural hearing loss (SNHL) - even when wearing hearing aids - often experience extreme difficulty perceptually navigating the auditory scene, severely limiting their ability to communicate effectively. From an acoustic perspective, the designation of a sound source as "target" versus "masker" is arbitrary because it depends on the current - and changeable - internal state of the observer. Although the amplification of sounds by hearing aids provides the best (often the only) option for improving communication for listeners with SNHL, current hearing aids inherently fail to solve the source selection problem because they amplify target and masker sounds indiscriminately. The challenge is to devise a hearing aid that focuses only on those sounds the listener chooses to attend and suppresses competing sounds, responding to the wishes of the listener immediately, accurately, and effectively. The goal of the work proposed here is to evaluate a new approach to providing amplification for listeners with hearing loss that is based on the premise that only the listener can make the distinction between which sources to attend and which to ignore. The experiments employ a prototype "hearing aid" that combines an eye-tracking device with an array of microphones that forms a steerable acoustic beam. By sensing where the eyes are focused, the prototype device can steer the beam of amplification toward the desired source. In that sense, it implements "top-down" control of focused amplification for the purpose of enhancing sound source selection. The primary goal is to determine the conditions under which top-down control of selective amplification, as implemented by this "visually-guided hearing aid" (VGHA), can benefit persons with SNHL in complex, dynamic and uncertain listening environments. This goal is to be accomplished under two specific aims that explore 1) hypotheses about top-down control of selective amplification in multitalker sound fields, and 2) the inherent dilemma posed by spatially selective amplification for simultaneously attending to a target source while concurrently monitoring the environment for new sources. A new approach to solving this dilemma will be examined using a dual task. The experimental plan to test these hypotheses employs listeners with hearing loss, matched listeners with normal hearing, and normal-hearing listeners under degraded stimulus conditions simulating the spatial hearing deficits caused by SNHL. Performance under VGHA conditions will be compared and contrasted to that obtained under representative control conditions, and the experimental plan is designed explicitly to take into account the type of masking - energetic versus informational - that is present.

描述（由申请人提供）：在现实的日常声学环境中，人类听众-无论他们是否具有正常或受损的听力-都依赖于人类大脑的相当大的处理能力来评估“听觉场景”。必须对各种声音来源进行分析，评估其相关性和重要性，然后必须将注意力集中在所需的来源（“目标”）上，而干扰源（“掩蔽物”）则被丢弃和忽略。患有感音神经性听力损失（SNHL）的听众-即使佩戴助听器-通常也会在感知听觉场景时遇到极大的困难，严重限制了他们有效沟通的能力。从声学的角度来看，将声源指定为“目标”与“掩蔽物”是任意的，因为它取决于观察者当前和可变的内部状态。虽然助听器对声音的放大提供了改善SNHL听众沟通的最佳（通常是唯一）选择，但目前的助听器本质上无法解决声源选择问题 因为它们不加区别地放大目标和掩蔽声。目前的挑战是设计一种助听器，只关注听者选择听的声音，抑制竞争的声音，立即，准确，有效地响应听者的愿望。这里提出的工作的目标是评估一种新的方法，为听力损失的听众提供放大，这是基于只有听众可以区分哪些源参加和忽略的前提。这些实验使用了一个“助听器”的原型，它将眼睛跟踪设备与麦克风阵列结合起来，形成了一个可操纵的声束。通过感知眼睛聚焦的位置，原型设备可以将放大光束转向所需的光源。从这个意义上说，它实现了“自上而下”的集中放大控制，以增强声源选择的目的。主要目标是确定条件下，自上而下的选择性放大控制，实现这种“视觉引导助听器”（VGHA），可以受益于人SNHL在复杂的，动态的和不确定的听环境。该目标将在两个具体目标下实现，这两个具体目标探索1）关于多讲话者声场中的选择性放大的自上而下控制的假设，以及2）由空间选择性放大引起的固有困境，用于同时关注目标源，同时监视环境中的新源。一种新的方法来解决这个困境将使用双重任务进行检查。测试这些假设的实验计划采用听力损失的听众，匹配的听力正常的听众，和正常听力的听众在退化的刺激条件下模拟SNHL引起的空间听力障碍。VGHA条件下的性能将进行比较和对比，在代表性的控制条件下获得的，和实验计划的设计明确考虑到掩蔽的类型-能量与信息-这是存在的。