【作者】 胡学平； 胡娜； 许珍珍； 胡敏； 王祥鹏； 赵佳； 陈安涛；

【Author】 Xueping Hu;Na Hu;Zhenzhen Xu;Min Hu;Xiangpeng Wang;Jia Zhao;Antao Chen;Key Laboratory of Cognition and Personality of Ministry of Education, Faculty of Psychology, Southwest University;

【机构】 西南大学心理学部认知与人格教育部重点实验室；

【摘要】 人类的嗓音携带了大量的社会信息,因此嗓音辨别在人际交流和社会互动中具有重要的社会意义。在嗓音辨别过程中,个体需要将注意力从无关信息转移到目标导向的嗓音信息,把感知到的声学特征映射到说话者特定的嗓音模式上,然后基于先前存贮的声学-语音印记与随之呈现的嗓音刺激进行匹配、计算并比较差异,最后才能判断出这些嗓音是否源自同一说话人。简言之,个体可能运用其认知控制能力排除无关信息的干扰并执行嗓音差异的比较。但已有研究多集中于语言经验的差异比较,即个体对其母语的嗓音辨别是否好于非母语嗓音的辨别。这些研究没能揭示个体的认知控制能力如何参与嗓音辨别,认知控制和语音经验在嗓音辨别中是何种关系。因此本研究结合嗓音辨别same-different任务和高低听觉Stroop任务来探讨这些问题。实验采用国际语音语料库中的汉语和韩语语音材料,并通过时间反转技术排除语音的语义信息并确保语音不可理解。实验以听觉Stroop不一致试次与一致试次的反应时差值进行认知控制能力高低分组,以此分别对嗓音辨别任务在声学特征和说话人变异等加工层面进行语言间的差异比较。40名无韩语经验大学生的实验结果发现,1)在嗓音辨别任务中,汉语和韩语在声学特质的反应时和d’指标上存在显著差异;在嗓音辨别任务中,两种语言在说话人变异的反应时上差异边缘显著,在d’上差异显著。当组合方式中的刺激存在声学结构或说话人身份不一致时,个体需要调用更多的注意认知控制等资源来匹配、比较并完成判断。2)高低听觉Stroop结果表明,语义和音量分别在一致性(一致性,不一致性)反应上的反应时和正确率均存在显著差异;基于一致性反应时差值中位数的分组,认知控制能力分组在语义和音量上均存在显著的组间差异。3)基于认知控制能力分组的分析发现,在声学结构和说话人身份加工的反应时差值指标上均存在语言内和语言间的个体差异,而在辨别力指标上则只存在语言间的差异。因此,结合当前结果和相关文献,我们认为认知控制参与了嗓音辨别,但是受到语音经验的调节。本研究首次通过实验明确了认知控制参与嗓音辨别的认识,加深了对语音经验调节嗓音辨别的理解。更多还原

【Abstract】 Human voice carries lots of social information, so voice discrimination is crucial for human communication and social interactions. In the process of voice discrimination, listeners should shift attention from irrelevant information to goal-directed information, and map the acoustic pattern onto phonetic idiosyncrasies of the speaker. Then listeners compare the pre-existed acoustical-phonetic imprint of voice with the followed ones, and finally make a judgement whether these voices spoke from the same speaker or not. Briefly, listeners might exert the ability of cognitive control to eliminate all distractions from the irrelevant information and execute difference comparison on voice discrimination. However, numerous studies for voice discrimination have focused on the difference between language experiences, that is, whether listeners have better behavior performance at discriminating voices in their native language than a foreign language or not. Therefore, whether the cognitive control involves voice discrimination, the relationship between cognitive control and phonological experience on voice discrimination are still an open issue. Therefore, the same-different voice discrimination and high-low auditory Stroop task were been recruited to solve these problems. The Mandarin Chinese and Korean speech stimuli selected from the international speech corpus and then edited with time-reversed method to exclude the influence of linguistic comprehension. The median value of RTincongruent-RTcongruent in the speech volume blocks of auditory Stroop task was been selected as a grouping criterion for the ability of individual cognitive control. Based on the grouping, the difference between languages were been analyzed in the process of acoustic feature and speaker variation processing for the voice discrimination task. Forty college students without any experience of Korean participated in the current study, and then their behavior performance dates were been collected and been analyzed. The results showed that 1) the significant differences were found between the Chinese and Korean voice discrimination in the reaction time and d′for acoustic feature processing. There was a marginal significance between the Chinese and Korean voice discrimination in the reaction time, but significant difference was observed between languages in the d′for speaker variation processing. Notably, the listener had a same worse behavior performance for non-and native language when the stimuli combination contained incongruent stimuli in the level of voice-acoustic structure or voice identity, thus the more attention and cognitive control resources were recruited to finish voice discrimination. 2) Compared with congruent trials in the high-low auditory Stroop task, significantly longer reaction times were found in the incongruent trials for the semantic and speech volume judgement, respectively. Moreover, there was a significant difference for the ability of cognitive control in speech volume judgement based on grouping, and the result of semantic judgement was same too. 3) Based on the grouping, there were statistically significant individual differences for intra-and inter-languages in the difference value of reaction time between the different voice-acoustic and same voice-acoustic or different speaker and same speaker. And the inter-language difference was only observed in the discriminability index d-prime. Therefore, combined the current study with related literatures, we proposed that cognitive control is engaged in voice discrimination and modulated by phonological experience. Finally, the current study firstly confirmed the knowledge of cognitive control involves in voice discrimination, and further deepen our comprehension regarding phonological experience modulates voice discrimination.更多还原