Text Captioning Buffers Against the Effects of Background Noise and Hearing Loss on Memory for Speech
Text Captioning Buffers Against the Effects of Background Noise and Hearing Loss on Memory for Speech
复制标题
DOI：
10.1097/aud.0000000000001079
复制
发表时间：
2022-01-01
期刊：
影响因子：
3.7
通讯作者：
Lohani, Monika
中科院分区：
文献类型：
--
作者：
Payne, Brennan R.;Silcox, Jack W.;Lohani, Monika
目标：日常言语理解经常发生在感知要求较高的环境中，例如，由于背景噪音和正常的与年龄相关的听力损失。由此产生的语音信号退化会增加听力的努力，这会对随后的记忆和理解产生负面的下游影响，即使语音是可理解的。在两个实验中，我们探讨了真实辅助文本字幕语音的呈现是否可以抵消背景噪音和听力障碍对语音记忆多项测量的负面影响。设计：在实验 1 中，听力正常的年轻成年人 (N = 48) 听句子以进行立即回忆和延迟再认记忆。语音在安静或两个级别的背景噪音中呈现。句子要么仅以语音形式呈现，要么以文本标题语音形式呈现。因此，实验遵循 2（有字幕与无字幕）x 3（无噪音、+7 dB 信噪比、+3 dB 信噪比）受试者内设计。在实验 2 中，一组听力敏锐度水平不同的老年人（年龄范围：61 至 80，N = 31）完成了与实验 1 相同的实验任务。对于这两个实验，通过广义线性混合效应模型分析了即时回忆、识别记忆准确性和识别记忆置信度。此外，我们在实验 2 中检查了作为听力敏锐度函数的个体差异。结果：在实验 1 中，我们发现，与单独的语音相比，在年轻听力正常的听众中呈现真实的文本字幕语音显示出改善的即时回忆和延迟识别记忆准确性和置信度。此外，文本字幕减弱了背景噪音对所有语音记忆结果的负面影响。在实验 2 中，我们在不同听力水平的老年人样本中复制了相同的结果模式。此外，我们发现，文本字幕的呈现减弱了听力损失对老年人言语记忆的负面影响。结论：总的来说，这些发现强烈表明，同时呈现文本可以抵消努力倾听对语音记忆的负面影响。至关重要的是，字幕的好处从即时单词记忆扩展到长期句子识别记忆，这种好处不仅适用于听力损失的老年人，也适用于听力正常的年轻听众。这些发现表明，文本字幕对记忆的好处是强大的，并且在支持声学挑战环境中的语音听力方面具有潜在的广泛应用。
Objective: Everyday speech understanding frequently occurs in perceptually demanding environments, for example, due to background noise and normal age-related hearing loss. The resulting degraded speech signals increase listening effort, which gives rise to negative downstream effects on subsequent memory and comprehension, even when speech is intelligible. In two experiments, we explored whether the presentation of realistic assistive text captioned speech offsets the negative effects of background noise and hearing impairment on multiple measures of speech memory. Design: In Experiment 1, young normal-hearing adults (N = 48) listened to sentences for immediate recall and delayed recognition memory. Speech was presented in quiet or in two levels of background noise. Sentences were either presented as speech only or as text captioned speech. Thus, the experiment followed a 2 (caption vs no caption) x 3 (no noise, +7 dB signal-to-noise ratio, +3 dB signal-to-noise ratio) within-subjects design. In Experiment 2, a group of older adults (age range: 61 to 80, N = 31), with varying levels of hearing acuity completed the same experimental task as in Experiment 1. For both experiments, immediate recall, recognition memory accuracy, and recognition memory confidence were analyzed via general(ized) linear mixed-effects models. In addition, we examined individual differences as a function of hearing acuity in Experiment 2. Results: In Experiment 1, we found that the presentation of realistic text-captioned speech in young normal-hearing listeners showed improved immediate recall and delayed recognition memory accuracy and confidence compared with speech alone. Moreover, text captions attenuated the negative effects of background noise on all speech memory outcomes. In Experiment 2, we replicated the same pattern of results in a sample of older adults with varying levels of hearing acuity. Moreover, we showed that the negative effects of hearing loss on speech memory in older adulthood were attenuated by the presentation of text captions. Conclusions: Collectively, these findings strongly suggest that the simultaneous presentation of text can offset the negative effects of effortful listening on speech memory. Critically, captioning benefits extended from immediate word recall to long-term sentence recognition memory, a benefit that was observed not only for older adults with hearing loss but also young normal-hearing listeners. These findings suggest that the text captioning benefit to memory is robust and has potentially wide applications for supporting speech listening in acoustically challenging environments.