| 摘要: | 日常生活中的聲音訊息，例如音樂及語言，通常伴隨著背景噪音，噪音中聽辨（Hearing-in-Noise; HIN）表現，即如何在背景噪音中提取目標聲音，是一項重要的能力。噪音中聽辨能力經常隨著老化和某些疾病而衰退，因此，如何提高噪音中聽辨能力是聽覺認知的重要議題。研究顯示，受過音樂訓練者在噪音中的聽辨表現較未受過音樂訓練者具優勢。然而，受過音樂訓練者在音樂能力上可能存在一些差異，因此造成了在研究中探討音樂家HIN優勢不一致的結果。本研究測試了絕對音感（absolute pitch; AP），即在無基準音的情況可正確辨識獨立音名之能力，是否為影響音樂家在噪音中聽辨表現之因子。本研究第一部分為行為實驗，實驗中以高、中、低（無）絕對音感音樂家及非音樂家在四種訊噪比（SNR = 0, -3, -6, -9 dB）的噪音環境中進行音樂聽辨測驗（Music-in-Noise Task; MINT）及中文語句聽辨測驗（Speech-in-Noise Task; SINT），受試者需辨識噪音中的目標旋律或語句並依照指示做出回答（比較兩段目標旋律是否相同，或打出目標語句），實驗中含有不同子測驗，如加入視覺或空間提示等。此外，也探討了可能會影響HIN的相關因子，如工作記憶及相對音感能力。行為實驗結果顯示，在四種訊噪比下，音樂家的旋律辨識正確率皆顯著高於非音樂家，且其準確率與絕對音感程度呈正相關，但在語句辨識中，則未看到音樂家及絕對音感的優勢。此外，結果顯示音樂工作記憶的表現與音樂聽辨表現呈正相關，說明了絕對音感能力增進噪音中之音樂聽辨表現，亦可能與其音樂工作記憶能力相關。本研究第二部分以功能性核磁共振腦造影比較高度和低度絕對音感之音樂家，以及非音樂家在音樂及中文語句聽辨任務之腦區活化以及靜息態腦區活化及功能性連結，受試者需對三種訊噪比（SNR = No Noise, 0 dB, -9 dB）中的目標旋律或語句做清晰度評分（1~5，5為最清楚）。腦造影結果顯示，高絕對音感者在主要聽覺區（i.e., superior and middle temporal gyrus; STG and MTG）對旋律及語句皆有比較好的表徵，而可能進一步增強後續高層認知任務的處理。而在辨識旋律或語句腦區活化不同之處主要為：辨識語句時，除了聽覺區外，活化的腦區大致與認知功能有關（i.e., superior frontal gyrus; SFG, and insula），且這些區域的活化程度與絕對音感能力呈正相關，但三組在噪音中辨識語句時，主要利用相似的顳葉到頂葉的功能性連結;而在噪音中辨識旋律時，絕對音感者似乎憑藉著對音樂刺激的敏感度，在初級聽覺區（i.e., Heschl’s gyrus; HG）對音樂就有較好的神經編碼，進而增進聽覺與感覺區（i.e., supramarginal gyrus; SMG）之間的連結與整合，對噪音中的旋律清晰度評分也顯著較高。此外，在靜息狀態下的神經網絡也顯示了高絕對音感者在主要聽覺區的連結較強，說明了絕對音感能力帶來的神經可塑性。總結來說，實驗結果顯示絕對音感能力會增進HIN表現，但只限於音樂而非語言刺激，且伴隨著在主要聽覺區對音樂有較好的神經表徵以及功能性連結，因此，若想要以音樂訓練來改善在複雜噪音背景中理解聲音目標的能力，應將絕對音感能力列入考量的因素之一。;The sound we receive in our daily life, such as music and speech, is usually accompanied by background noise. Hearing-in-noise (HIN) ability, the ability to extract sound targets from background noise is thus crucial. HIN ability usually declines with age and in certain diseases, so how to effectively extract important information from background noise is an important issue in auditory cognition. Previous research has shown that people with musical training perform significantly better on HIN tasks. However, there may be some differences in musical ability among musicians which may contribute to the inconsistent findings observed in musician HIN advantage. Therefore, the current research investigated whether absolute pitch (AP), an innate musical ability to identify isolated notes, could be one factor that affect musician HIN performance. The first part of this study consisted of behavioral experiments. We examined musicians with high-, quasi-, non-AP ability and non-musicians on performance of Music-in-Noise task (MINT) and Mandarin Speech-in-Noise task (SINT) with four signal-to-noise ratios (SNRs = 0, -3, -6, -9 dB). Participants were asked to identify the target melody or sentence in noise and to respond according to the instructions (compare whether the two target melodies were the same, or type the target sentence). The experiment contained different subtasks, such as the addition of visual cues or spatial cues, etc. In addition, we also explored related factors that may affect HIN performance, including working memory and relative pitch ability. Behavioral results showed that musicians outperformed non-musicians in perceiving melody in noise, but not speech. Musicians conferred an advantage in identifying melodies under difficult conditions, with those possessing higher AP ability having better performance. Furthermore, results showed that accuracy in working memory recall of musical notes was positively correlated with performance accuracy on music-in-noise, but not speech-in-noise task. The second part of this study consisted of functional magnetic resonance imaging (fMRI) experiments. We investigated the brain activation and functional connectivity of musicians with high-AP, non-AP ability and non-musicians when performing music- and speech-in-noise tasks and in resting state. During fMRI experiment, participants were required to use the button box to rate the intelligibility of the target melody or sentence (1~5, 5 = most intelligible) under three SNR levels (SNRs = No Noise, 0 dB, -9 dB). Results showed that musicians with high AP ability may have better representation of melody or speech in primary auditory areas (i.e., superior and middle temporal gyrus; STG and MTG), which may further enhance subsequent high-level cognitive processing. With respect to the differences between identifying music or speech, except for auditory areas, the activated regions were related to cognitive functions (i.e., superior frontal gyrus; SFG, and insula) in speech perception in noise, and the degree of activation in these areas positively correlated with AP proficiency. However, we found that the three groups shared similar major functional connectivity between auditory and parietal regions for speech perception in noise. For music perception in noise, AP possessors relied on their sensitivity to musical stimuli to have better neural encoding of music in the primary auditory area (i.e., Heschl’s gyrus; HG), thereby enhancing the connection and integration between the auditory and sensory areas (i.e., supramarginal gyrus; SMG). Compared to non-AP musicians and non-musicians, AP musicians also had significantly higher intelligibility ratings on musical targets in noise. Furthermore, resting-state results revealed AP musicians had stronger connectivity in primary auditory regions, suggesting neuroplasticity shaped by absolute pitch ability. In summary, these results suggest that absolute pitch ability is associated with improved HIN perception for music but not speech stimuli, likely underlied by enhanced activation in primary auditory regions. Therefore, if music training were to be used to enhance speech intelligibility in noisy environments, absolute pitch ability should be one of the factors to be taken into consideration. |
