Biosciences institute, Newcastle University, United Kingdom.
Biosciences institute, Newcastle University, United Kingdom; Institute of Neuroscience and Psychology, University of Glasgow, United Kingdom.
Hear Res. 2022 Sep 1;422:108524. doi: 10.1016/j.heares.2022.108524. Epub 2022 May 16.
Speech-in-noise difficulty is commonly reported among hearing-impaired individuals. Recent work has established generic behavioural measures of sound segregation and grouping that are related to speech-in-noise processing but do not require language. In this study, we assessed potential clinical electroencephalographic (EEG) measures of central auditory grouping (stochastic figure-ground test) and speech-in-noise perception (speech-in-babble test) with and without relevant tasks. Auditory targets were presented within background noise (16 talker-babble or randomly generated pure-tones) in 50% of the trials and composed either a figure (pure-tone frequency chords repeating over time) or speech (English names), while the rest of the trials only had background noise. EEG was recorded while participants were presented with the target stimuli (figure or speech) under different attentional states (relevant task or visual-distractor task). EEG time-domain analysis demonstrated enhanced negative responses during detection of both types of auditory targets within the time window 150-350 ms but only figure detection produced significantly enhanced responses under the distracted condition. Further single-channel analysis showed that simple vertex-to-mastoid acquisition defines a very similar response to more complex arrays based on multiple channels. Evoked-potentials to the generic figure-ground task therefore represent a potential clinical measure of grouping relevant to real-world listening that can be assessed irrespective of language knowledge and expertise even without a relevant task.
在听力受损人群中，常报告存在噪声下言语理解困难的问题。最近的研究已经确定了与言语感知相关的通用声音分离和分组的行为测量方法，但这些方法不需要语言。在这项研究中，我们评估了潜在的临床电生理（EEG）指标，包括中央听觉分组（随机图形-背景测试）和噪声下言语感知（言语噪声测试），并在有和没有相关任务的情况下进行了评估。听觉目标在背景噪声（16 个说话者噪声或随机生成的纯音）中呈现，在 50%的试验中，目标由图形（纯音频率和弦随时间重复）或言语（英语名称）组成，而其余试验只有背景噪声。在参与者呈现目标刺激（图形或言语）时，记录 EEG，同时参与者处于不同的注意状态（相关任务或视觉分散任务）。EEG 时域分析表明，在 150-350ms 的时间窗口内，两种类型的听觉目标的检测都产生了增强的负响应，但只有在分散注意条件下，图形检测才产生了显著增强的响应。进一步的单通道分析表明，简单的顶点到乳突采集与基于多个通道的更复杂阵列产生了非常相似的响应。因此，通用图形-背景任务的诱发电位代表了一种与现实听力相关的潜在临床分组测量方法，即使没有相关任务，也可以评估其与语言知识和专业知识无关的情况。
