词汇是习得一门语言的基础，不论是儿童还是成人，通过阅读学习词汇都是一种重要的学习方式。阅读情境中的词汇学习受到阅读材料特性的影响，前人研究大多关注于阅读材料的限制性强度和阅读材料主题的多样性对词汇学习的影响，而日常生活中很多阅读材料均带有情绪色彩，但截止目前还没有研究考察阅读材料的情绪性对阅读情境中词汇习得的影响。除了阅读材料的特性外，生词在阅读材料中需要接触多少次才能习得词汇也是研究者重点关注的问题，但前人较少有研究考察阅读材料的特性对生词接触频率效应的调节作用。本研究拟操纵语篇的情绪性和生词在语篇中接触频率这两个变量，以汉语-英语双语者为被试，探讨阅读语篇的情绪性和生词接触频率对第二语言（二语）词汇学习的影响，并从词汇内隐知识和外显知识两个角度对习得效果进行探查。

本研究共包括五个实验。实验1我们编制了带有不同情绪色彩的短语篇，并把这些语篇作为阅读的材料。我们把英语生词（英语假词）嵌入不同情绪色彩的语篇中，让被试在连续五天的时间里分别阅读带有不同情绪色彩的语篇。实验1采用被试间设计，即三组被试分别阅读积极情绪语篇、消极情绪语篇和中性情绪语篇。被试每天从阅读的语篇中接触3次英语生词，五天共接触15次。从第二天阅读结束开始，连续进行词汇知识测试，其中，采用字形再认和语义回忆任务测查词汇外显知识的习得，采用字形启动和语义启动任务测查词汇内隐知识的习得。实验2采用了被试内的实验设计，即同一组被试阅读带有不同情绪色彩的语篇，学习程序和词汇知识的测查任务等同实验1。为了排除重复测试次数对实验结果可能造成的影响，实验3仅在第五天完成阅读后进行词汇知识的测试，测试的方式同实验1和实验2。实验4和实验5，我们采用了更为实时、更为敏感的眼动技术考察阅读语篇的情绪性和生词接触频率对二语词汇学习的影响。两者的不同在于实验4采用了实验2、实验3的短语篇材料，实验5采用了更加具有生态效度的长语篇（小说改编而来）。实验4和实验5同样采用字形再认和语义回忆任务测查词汇外显知识，采用眼动追踪的数据查词汇的内隐知识。

研究结果表明，对于词汇外显知识而言：（1）语篇的情绪性是影响二语词汇外显知识习得的重要因素。具体表现为：和中性语篇相比，带有情绪色彩的语篇更有利于词汇意义的习得。（2）语篇中生词接触频率也是影响词汇外显知识习得的重要因素，重要的是，生词接触频率的作用受语篇情绪性特性的影响，具体表现为，达到同样的学习效果，阅读带有情绪色彩语篇比阅读中性情绪材料的语篇，需要接触生词的次数要少。例如，实验1和实验2中当关键词出现12次时，积极情绪语篇和消极情绪语篇下语义回忆的正确率已超过50%，但中性语篇直到15次时仍未达到50%。对于词汇内隐知识而言：实验4 结果表明，当生词出现12次和15次时，积极语篇和消极语篇会对词汇通达起到促进作用；实验5结果显示生词在接触频率较多时（20次左右），不论在哪种类型的语篇材料，被试均可建立生词的词汇表征，但仍表现出消极语篇对词汇通达的促进作用。

综合上述研究结果，我们得出以下结论：（1）阅读语篇的情绪性不仅影响二语外显词汇知识的习得，同样影响二语内隐词汇知识的习得。对外显知识习得的影响主要表现在语义回忆任务中，表现为阅读带有积极和消极情绪色彩语篇，尤其是带有积极情绪色彩语篇，更容易在阅读情境中习得语篇中生词的含义。对于词汇的内隐知识，当生词接触20次左右时，被试能够建立起词汇表征，语篇情绪性因素的促进作用，主要体现在词汇通达阶段。（2）阅读情境中，生词接触频率的作用受到语篇情绪性特性的影响，具体表现为，达到同样的学习效果，阅读带有情绪色彩语篇比阅读中性情绪材料的语篇，需要接触生词的次数要少。（3）生词接触频率对词汇外显知识和内隐知识的影响轨迹不同，生词接触频率对词汇外显知识习得产生作用的时间早于词汇内隐知识，对词汇内隐知识的作用主要体现在语义整合阶段，对词汇通达阶段的促进作用会更多的受到语篇情绪性的调节。

本研究首次考察了阅读语篇的情绪性对阅读情境中二语词汇学习的影响，并将其和生词的接触频率进行结合，从词汇外显知识习得和内隐知识习得两个角度进行考察，在本研究的基础上我们尝试提出语篇的情绪性如何影响二语词汇学习的理论模型以解释实验的结果。

Vocabulary is the basis of acquiring language. Whether children or adults, learning vocabulary through reading is an important way of learning. This kind of vocabulary learning is affected by the characteristics of reading materials. Previous studies mostly focused on the impact of the restrictive degree of reading materials and the diversity of reading materials on vocabulary learning. However, in daily life, many reading materials are emotional. But few previous studies have found the impact of the contexts’ emotion on vocabulary acquisition till now. In addition to the characteristics of reading materials, the exposure frequency of novel words in reading materials is also the focus of researchers, but few previous studies have investigated the regulatory effect of the characteristics of reading materials on the number of novel words. This study intends to manipulate the two variables of the contexts’ emotion and the exposure frequency frequency of novel words in the context. Taking Chinese-English bilinguals as participants, this study discusses the impact of the reading contexts’ emotion and the exposure frequency frequency of novel words in the contexts on second language vocabulary learning, and explores the acquisition effect from the perspectives of vocabulary implicit knowledge and explicit knowledge.

This study answers the above questions through five experiments. In Experiment 1, we compiled short passages with different emotion (positive, negative and neutral) as reading materials, and adopted the inter-subject design, that is, three groups of participants in EXP1: positive group, negative group and neutral group. Participants read only one kind of emotional contexts (embed English pseudowords) for five consecutive days. The participants were asked to read the context in a self-paced reading paradigm, and they were exposed to English pseudowords three times a day from the reading contexts, and 15 times in total in the five days. From the end of reading on the second day, vocabulary knowledge was tested continuously. Form recognition and semantic recall tasks were used to test the acquisition of explicit vocabulary knowledge, and form priming and semantic priming tasks were used to test the acquisition of implicit vocabulary knowledge. Within-subject experimental design was adopted in experiment 2, that is, the same group of participants read the three kinds of emotional contexts. The testing tasks of vocabulary knowledge were the same as experiment 1. In order to eliminate the possible influence of repeated tests on the experimental results, Experiment 3 only tested vocabulary knowledge after reading on the fifth day. The testing method was the same as experiment 1 and Experiment 2. In Experiments 4 and 5, we used more real-time and sensitive--eye movement techniques to investigate the effects of the emotion of reading context and the exposure frequency of novel words on second language vocabulary learning. The difference between the two is that experiment 4 adopted the artificial short contexts of Experiment 2 and Experiment 3, but Experiment 5 adopted the long text with more ecological validity (adapted from the real English novel). Similarly, the explicit knowledge of vocabulary was still tested by form recognition and semantic recall tasks, and the implicit knowledge of vocabulary was tested by eye tracking data.

The main findings of this study are as follows: (1) The emotion of context is an important factor affecting the acquisition of explicit knowledge of second language vocabulary. Specifically, compared with neutral context, positive and negative context is more effective to the acquisition of meaning. (2) The exposure frequency of novel words appeared in context is effective to the acquisition of explicit vocabulary knowledge. More importantly, the role of the exposure frequency of novel words is affected by the emotional characteristics of the text. In Experiment 1 and Experiment 2, when targets appeared 12 times, the accuracy of semantic recall in positive and negative discourse has exceeded 50%, but it has not reached 50% in neutral context until 15 times. In Experiment 4, positive and negative context have exceeded 40% at 12 times, and neutral discourse has not reached 40% until 15 times. For the implicit knowledge, the results of eye tracking experiment in Experiment 4 and Experiment 5 showed that the participants can establish the lexical representation of novel words in the process of natural reading. The results of Experiment 4 showed that when targets appeared 12 and 15 times, positive negative context will promote lexical access. The results of Experiment 5 showed that on the basis of sufficient occurrences of novel words (about 20 times), the participants can establish the lexical representation of novel words in any emotional context, but still show that negative context will promote the lexical access.

Based on the above research results, we draw the following conclusions: (1) the emotion of reading context affects not only the acquisition of explicit vocabulary knowledge, but also the acquisition of implicit vocabulary knowledge. The influence of explicit knowledge acquisition is mainly reflected in the meaning recall task. Reading contexts with positive and negative emotions is easier to acquire the meaning of novel words in the context. The acquisition of implicit knowledge is mainly reflected in reading contexts with negative emotion, which is more effective to the establishment of vocabulary representation of the novel words. (2) In the L2 contextual word leaning, the exposure frequency of occurrence of novel words is affected by the emotional characteristics of the context. Specifically, to achieve the same learning effect, compared with the neutral contexts, reading the emotional contexts need less exposure frequency. (3) The influence of novel words’ exposure frequency on the trajectory of lexical explicit knowledge and implicit knowledge is different. The effect of novel words’ exposure frequency on the acquisition of lexical explicit knowledge is earlier than lexical implicit knowledge. The effect on lexical implicit knowledge is mainly reflected in the stage of semantic integration, and the promotion on the stage of lexical access will be more regulated by the contexts’ emotion.Lexical representation can be established in the process of reading. When the exposure frequency of occurrence is not enough, although the lexical representation of novel words can not be established, but it can be found that positive and negative context can promote the lexical access. When the novel words appear enough times, there is no difference among the three emotional contexts in the semantic integration, it can still show the promoting effect of negative context on lexical access.

This study is the first to investigate the impact of the emotional characteristics of the reading contexts on L2 vocabulary learning, combined with the exposure frequency of targets, from the perspective of the explicit and implicit knowledge acquisition. On the basis of these studies, this study attempts to put forward a theoretical model of how emotional context affects the L2 contextual word learning.