第二语言学习是人脑可塑性的重要体现。随着人类社会的发展和国际交流的提升，第二语言学习日益普遍和重要。英语是最为广泛学习的第二语言，仅我国就有接近4亿英语学习者（Wei & Su, 2012）。对大多数学习者来说，第二语言学习是一项极具挑战性的任务。由于第二语言学习受到母语影响，并与各种学习经验存在复杂关联，因此，解析母语对第二语言学习的影响，并揭示特定学习经验的作用，对于理解人脑可塑性、二语学习机制具有重要理论意义，并对促进第二语言学习具有重要现实意义。

阅读学习是第二语言学习的重要领域。第二语言阅读学习的第一步是文字解码。只有准确、快速地完成文字解码，学习者才能节省认知资源，以投入经由阅读的学习和认知加工中。不过，文字解码会受特定文字形-音对应规则的影响。英语和汉语是两种差异巨大的文字。英语是拼音文字，有效解码英语单词依赖于学习者自动化地整合单词中的字母与其对应的语音，因此字母-语音自动化整合是英语单词解码的关键。与此相对，汉语是非拼音文字，汉字的任何笔画、部件均不表征音位，汉语阅读并不涉及类似英语字母-语音整合那样精细到音位水平的形-音对应与转换。已有研究表明，我国汉语母语者在英语阅读解码的关键指标——字母-语音自动化整合时存在困难，母语背景制约了汉语母语者的英语字母-语音自动化整合。因此，本论文拟系统研究汉语母语者在英语阅读解码中脑的字母-语音自动化整合模式及其与英语学习熟练度、英语学习方式（沉浸式与非沉浸式）、其他学习经验（如音乐学习）、以及训练经验的关系，以期为完善第二语言学习理论、研发有效促进汉语母语者英语学习的方法提供新的观点和证据。

本论文结合脑电和认知行为技术，采用横断的组间比较和纵向训练设计，针对汉语母语者英语字母-语音自动化整合不足及其与学习经验的关系，我们完成了四项研究，探讨了英语熟练度、英语学习方式、非语言特异的音乐学习经验、以及训练经验等因素对汉语母语者英语阅读解码中脑的自动化加工模式的作用，取得了以下系列研究结果。

研究1考察了英语熟练度能否调节母语背景对汉语母语者英语字母-语音自动化整合的影响。英语母语者（n = 15）、一般英语熟练度（n = 12）和高英语熟练度（n = 15）的汉语母语者参与了研究。高英语熟练度的汉语母语者和英语母语者的英语熟练度水平达到匹配。两组汉语母语者均没有英语母语国家沉浸式学习经验和专业音乐学习经验，在英语学习起始年龄和学习年限上匹配。研究采用跨通道失匹配负波（Mismatch Negativity: MMN）范式（Froyen et al., 2008），对比了听觉单通道条件（Auditory-only）和两个视听双通道条件（AV0和AV200）间偏差刺激诱发的MMN差异。与听觉单通道条件诱发的MMN相比，如果视听双通道条件诱发的MMN显著增强则表明字母-语音自动化整合成功；反之，则表明失败。结果表明，英语母语者在AV0下诱发的MMN显著大于听觉单通道条件下的MMN，说明其成功整合了英语字母-语音。与英语母语者的表现相反，一般英语熟练度和高英语熟练度的汉语母语者在视听双通道条件下的MMN均显著小于听觉单通道条件的MMN，说明二者的英语字母-语音自动化整合失败，进一步验证了母语背景深刻制约着汉语母语者英语阅读解码中脑的自动化加工。

研究2探讨了英语沉浸式学习经验对汉语母语者英语字母-语音自动化整合的作用。研究采用同样的跨通道MMN范式，比较了英语母语者（n = 19）、有英语沉浸式学习经验（n = 23）和无沉浸式学习经验（n = 20）的汉语母语者。有沉浸式学习经验的汉语母语者在英语为官方语言的国家居住的平均时间是4.65年（SD = 3.2），无沉浸式英语学习经验的汉语母语者均没有在以英语为官方语言的国家生活的经验。与无沉浸式学习经验的汉语母语者相比，有沉浸式学习经验的汉语母语者英语学习起始年龄显著较小，学习年限显著较长。跨通道MMN结果显示，有沉浸式学习经验的汉语母语者和英语母语者表现相同，即都可以成功自动化地整合英语字母-语音，而无沉浸式学习经验的汉语母语者依然失败。即使在控制英语学习起始年龄和学习年限后，该结果保持不变。进一步，我们发现有沉浸式学习经验的汉语母语者和英语母语者采用的加工机制也相同，都表现出显著增强的MMN和delta振荡（自下而上的听觉-知觉加工机制），以及theta振荡（自上而下的认知加工机制），而无沉浸经验的英语学习者没有上述特点。此外，沉浸环境中的口语经验（听和说）和英语字母-语音自动化整合显著相关，提示口语经验对于学习者克服母语背景制约、建立目标语言要求的脑的自动化反应具有重要价值。

除特定的第二语言沉浸式学习经验外，对跨通道整合具有较高精细要求的非语言特异性学习经验是否也有助于汉语母语者的大脑建立英语阅读解码的关键活动模式呢？音乐学习经验就是一种涉及大量精细视听跨通道整合加工的非语言特异性经验。为此，研究3探讨了专业的音乐学习经验能否促进汉语母语者的英语字母-语音自动化整合。具有专业音乐学习经验的汉语母语者（n = 15）、无音乐学习经验的汉语母语者（n = 15）和英语母语者（n = 15）参与了研究。两组汉语母语者在英语学习起始年龄、学习年限和英语熟练度上匹配。结果表明，具有专业音乐学习经验的汉语母语者可以成功自动化地整合英语字母-语音，其整合机制与英语母语者不同。具有专业音乐学习经验的汉语母语者在字母-语音同时和延时条件下的theta振荡均显著增强，提示音乐学习经验可能采用自上而下的认知加工机制。而英语母语者在字母-语音同时条件下发现了增强的delta振荡，且该振荡活动与其单词认读和假词拼读存在显著相关，说明英语母语者可能采用了自下而上的听觉-知觉加工机制。

母语背景深刻制约着汉语母语者的英语字母-语音自动化整合，处在阅读学习初始阶段的汉语儿童在经过数年的英语学习后，仍然无法熟练掌握英语字母-语音对应规则。如果进一步提供强化的英语字母-语音训练（特异性经验）和音乐训练（非特异性经验），能否有效改善其英语阅读加工呢？为此，研究4对小学2-4年级的汉语母语儿童开展了研究。首先，我们发现在接受训练前，汉语母语儿童表现为听觉单通道条件的MMN和视听双通道条件的MMN无显著差异，即没有表现出类似于汉语母语成人显著减弱的跨通道MMN效应，说明由于儿童阶段的母语经验相对有限，其大脑还没表现出强烈的受到母语背景制约的特征。其次，我们开展了随机对照训练研究，探讨英语训练和音乐训练经验的作用。英语训练组（n = 50）、音乐训练组（n = 37）和控制组（n = 36）在训练前的年龄、父母受教育水平和家庭年收入上匹配。训练时长为每周2次，每次1小时，共50次，持续一学年。结果发现，经过一学年的训练后，英语训练组和音乐训练组都可以自动化地整合英语字母-语音，但二者采用的加工机制存在不同。音乐训练组在两种视听双通道条件（AV0和AV200）下的theta振荡显著增强，提示音乐训组可能采用自上而下的认知加工机制；英语训练组在两种视听双通道条件（AV0和AV200）下的delta和theta振荡均显著增强，说明英语训练组可能采用自下而上的听觉-知觉加工和自上而下的认知加工相结合的机制。

基于上述系列研究结果，本论文结论如下：（1）当母语和二语的文字体系存在较大差异时，母语背景深刻制约着汉语成人英语阅读解码的关键指标——字母-语音自动化整合，处在阅读初始阶段的汉语儿童未表现出成人学习者那样的受到母语背景强烈制约的脑自动化加工模式。（2）第二语言沉浸式学习经验，特别是沉浸环境中听和说的口语经验有助于汉语母语者克服母语背景的限制，从而能像英语母语者那样进行英语阅读解码中脑的自动化加工；（3）语言特异的训练经验与音乐这一非特异性经验均有助于汉语母语者按照目标语言的要求建立起字母和语音间的自动化连接，但是音乐学习经验的作用机制与语言经验不同。（4）当缺乏第二语言沉浸经验和非特异学习经验（如音乐学习经验）时，二语熟练度不能显著调节母语背景对二语阅读解码中脑自动化加工模式的影响。

Second language learning is important for brain plasticity. With the development of human society and the improvement of international communication, second language learning is becoming more and more important. English is the most widely learned second language. There are more than 390 million people who learn English as a second language in China (Wei & Su, 2012). For most learners, second language learning is challenging. Second language learning is affected by native language, and it has complex associations with various learning experiences. Therefore, it is necessary to investigate the effects of native language and learning experiences on second language learning. It not only has important theoretical significance for understanding brain plasticity and the mechanism of second language learning, but also has important practical significance for promoting second language learning.

Second language reading is an important area for second language learning. Word decoding is the first step of second language reading. When the word decoding became accuractely and automatically, then the learners could process more complex reading and cognitive activity. Moreover, language system may influence word decoding. English and Chinese are two different language systems. English is an alphabetic language, letter-sound automatic integration is crtical for English word decoding. The learners need to integrate each of letters in the word and its corresponding speech sound. However, different with English reading, Chinese is a non-alphabetic language, and Chinese character reading does not involve letter-sound integration. Thus, it is difficult for native Chinese speakers to integrate English letter-sound automatically. Therefore, the present study combined both EEG and cognitive behavior tests, and aimed to investigate whether different learning experiences could help native Chinese speakers’ brain overcome the language background constraints and integrate English letter-sound automatically. The learning experiences include second language proficiency, second language immersion learning experience, music learning experience, and training experience. The results are as follows.

Study 1 explored whether second language proficiency could help native Chinese speakers integrate English letter-sound automatically. Native English speakers (n = 15), native Chinese speakers with intermediate English proficiency level (n = 12) and advanced English proficiency level (n = 15) were recruited. Native Chinese speakers with advanced English proficiency level and native English speakers were matched on English proficiency. All native Chinese speakers did not have any English immersion experience and professional music learning experience. Native Chinese speakers with intermediate English proficiency level and advanced English proficiency level were matched on English age of acquisition and learning duration. To examine the letter-sound automatic integration, the present study used Cross-modal MMN paradigm which was developed by Froyen et al. (2008). This paradigm includes an auditory-only condition and two audio-visual conditions (AV0 and AV200). The logic of this paradigm is that: compared with the MMN induced by auditory-only condition, the enhanced MMN induced by the audio-visual conditions indicates letter-sound automatic integration. The results found that native English speakers could integrate English letter-sound automatically, who performed the enhanced MMN in the AV0 condition compared with that of the auditory-only condition. However, native Chinese speakers with both intermediate and advanced English proficiency level failed to integrate English letter-sound automatically, who performed larger MMN in the auditory-only condition than that of the audio-visual conditions. The results revealed that native language background constrained native Chinese speakers’ English brain automatic responses.

Study 2 further explored whether second language immersion experience could help native Chinese speakers integrate English letter-sound automatically. The Cross-modal MMN paradigm was also used. Native English speakers (n = 19), native Chinese speakers with English immersion experience (n = 23) and without any English immersion experience (n = 20) were recruited. For native Chinese speakers with English immersion experience, their average length of residence in an English-speaking country was 4.65 years (SD = 3.2). Native Chinese speakers with English immersion experience began to learn English earlier and had longer English learning duration than native Chinese speakers without any English immersion experience. The results showed that both native English spearkers and native Chinese speakers with English immersion experience could integrate English letter-sound automatically. After controlled for age of acquisition and learning duration, the results still remained. In addition, both native English speakers and native Chinese speakers with English immersion experience use same processing mechanism. They showed enhanced MMN and delta oscillation (bottom-up auditory-perceptual processing), and theta oscillation (top-down cognitive modulation). The present study also found that Englsih oral experience (listening and speaking) is important for native Chinese speakers’ English letter-sound automatic integration.

Besides language-specific experience, whether non-language-specific learning experience with fine requirements for cross-modal integration could also help native Chinese speakers’ brain to establish critical skill for English decoding? Music experience is such a non-language-specific experience which involves a large number of fine audiovisual integration processing. Therefore, study 3 investigated whether professional music experience could facilitate native Chinese speakers’ English letter-sound automatic integration. Native Chinese speakers with professional music experience (n = 15) and without music experience (n = 15), and native English speakers (n = 15) were recruited. Native Chinese speakers with professional music experience and without music experience were matched on English age of acquisition, English learning duration, and English proficiency. The results showed that music experience could facilitate native Chinese speakers’ English letter-sound automatic integration. We further explored the mechanism of music experience. The results found that native Chinese speakers with professional music experience showed enhanced theta oscillations in both AV0 and AV200 conditions, indicating native Chinese speakers with professional music experience may use top-down cognitive modulation to integrate English letter-sound automatically. Native English speakers showed enhanced delta oscillation in AV0 condition, and delta oscillation significantly correlated with English word reading and pseudoword reading, indicating native English speakers may use bottom-up auditory-perceptual processing to process English letter-sound automatic integration.

Native language background deeply constrained native Chinese speakers’ English letter-sound automatic integration. The letter-sound integration training is important for native Chinese-speaking children who are in the initial stage of the reading. The reason is that the native Chinese-speaking children have not been constrained by the abundant native language experience. Study 4 investigated the role of English and music training experience for native Chinese-speaking children. First, we found that before training, there was no significant difference for MMN between the auditory condition and audio-visual conditions. Seocnd, we explored the effects of English and music training experiences. The English training group (n = 50), music training group (n = 37), and the control group (n = 36) were matched in terms of age, parental education level, and annual family income. The training duration is two times a week, each time for one hour, totally 50 hours. After training, both English and music training group could integrate English letter-sound automatically. We further explored the mechanisms of English and music training experience. The results found that music training group showed enhanced theta oscillation in both AV0 and AV200 conditions, indicating music training experience may use top-down cognitive modulation to process English letter-sound automatic integration. English training group showed enhanced delta and theta oscillations in both AV0 and AV200 conditions, indicating English training group may use both bottom-up auditory-perceptual processing and top-down cognitive modulation to integrate English letter-sound automatically.

In summary, the results of the present study were as follows. (1) Native language background deeply constrained native Chinese adults’ English letter-sound automatic integration. For native Chinese-speaking children, they did not show the constraint brain response like adults. (2) Second language immersion experience, especially English oral experience (listening and speaking) could help native Chinese speakers overcome native language background constraints and respond as required by the second language. (3) Both English and music training experiences could help native Chinese speakers build English letter-sound automatic integration. The mechanism of music learning experience is different with that of the English training experience. (4) Without second language immersion experience and music learning experience, English second language proficiency could not benefit native Chinese speakers’ English letter-sound automatic integration.