备受瞩目的语言习得与认知神经科学交叉论坛即将拉开帷幕！
这不仅是一次交叉学科的盛会，更是思想碰撞、知识交融的绝佳机会，我们将有幸接触国内外知名专家学者们的最新研究成果。
让我们继续深入了解出席本次论坛的部分专家和他们的报告内容，共同期待这场学术盛宴。
Boping Yuan教授
剑桥大学丘吉尔学院院士、博士生导师，上海交通大学特聘教授
主要研究领域：
理论语言学框架下的二语和多语习得、双语与人类健康
报告题目：
二语输入中的语示分类及其理论和教学意义：以汉语二语为例
报告摘要：
学界普遍认为，在二语者接触的目标语语料输入中，语示（language cues）对于触发二语者大脑中二语语法的重新配置和提高二语水平是至关重要的。然而，二语语法中语言特征重新配置所需要的语示，可能会由于一语的影响而变得模糊不清，其显现度和可见度在目标语输入中可能会有所不同。并非所有的语言输入都可以完全转化为吸收，这是因为在语言输入过程中会出现相关语示的缺失、不足、甚至误导，同时相关语示在语言输入中的可见度和显现度也会影响相关语示的吸收。本文以汉语二语为例，根据其显现强度和可见度，将语言输入中的语示分为不同类型，并分别将其命名为宏观型语示 (macro-cues)、微观型语示 (micro-cues)、纳米型语示 (nano-cues) 和语示缺失 (absence of cues)。这些类别构成一个具有不同显现强度和可见度的语示连续体。我们认为这种对二语输入中语示的分类有助于汉语二语语法教学策略的选取；它显示了保留传统语法教学的必要性，特别是针对涉及微观型语示、纳米型语示、 语示缺失语言现象的语法教学，同时这也对汉语二语习得研究具有重要的理论意义，因为它为我们提供一种依据，让我们可以用来解释汉语二语习得中某些语言现象为什么容易习得，难以习得，习得迟缓或无法成功。
Shuai Li教授
美国卡内基梅隆大学教授
主要研究领域：
汉语作为第二语言的习得研究及二语语用学
报告题目：
An Exploratory Study on Developing a Small Language Model for Assessing Request-making in L2 Chinese
报告摘要：
Pragmatic competence refers to the ability to conduct socially appropriate communication. Recently, researchers have explored the application of Natural Language Processing techniques to assessing pragmatics. With the rise and evolution of generative AI technologies, there is substantial potential for further advancements in this area. The present study represents an initial effort in this direction.
Methodologically, none of the existing large language model were specifically designed for the purpose of language assessment. Because the data used for training the LLMs were not tagged for pragmatics, assessment quality may be in jeopardy. To address this limitation, we adopted the few-shot learning technique in machine learning to train a small language model to develop an AI-based tool for assessing speech act production of requests in L2 Chinese. We used 440 carefully tagged request utterances collected from 22 Chinese native speakers to develop and train our small language model. Tagging involved all essential pragmalinguistic components of request making. We then used the SLM to evaluate 1,240 request sequences collected from 62 American learners of L2 Chinese. We plan to discuss our findings and the implications for training AI models for assessing L2 pragmatic competence.
张积家教授
广西师范大学特聘教授、博士生导师
主要研究领域：
认知心理学、语言心理学
报告题目：
The Influence of Hunan Dialect on Speech Perception of Phonetic Contrasts /n/-/l/ and /f/-/h/ in Mandarin and Ambiguity Resolution
报告摘要：
It is an indisputable fact that dialects have affected the perception of phonetic categories in Mandarin. In Hunan dialects,the confusion of phonetic contrasts /n/-/l/ and /f/-/h/ is one of the primary reasons for the difficulty of Hunan students in learning Mandarin. Therefore, an empirical study on the the perception of phonetic contrasts for Hunan students must be carried out. This study respectively employs university students who grew up in Hunan and northern area to participate in two experiments.
In Experiment1a, 56 Hunan dialect speakers and 44 northern dialect speakers students were employed as subjects. The assimilation types of phonetic contrasts /n/-/l/ and /f/-/h/ were investigated by making phonetic judgement. In Experiment1b, 32 Xiang dialect speakers and 36 northern dialect speakers students were also employed as subjects. The main ways of lexical processing include auditory and visual channels at the level of speech comprehension. Therefore, Experiment 1b attempted to testify whether Hunan dialect speakers also had difficulty in perceiving sounds in single minimal pairs from the perspective of visual lexical processing. In Experiment2, 48 Hunan dialect speakers and 46 northern dialect speakers students were also employed as subjects. The purpose of Experiment 2 was to explore whether word context can achieve the disambiguation of phonetic contrasts for Hunan dialect speakers.
Experiment 1a exhibited that compared with the condition of irrelevant sound, both Hunan dialect speakers and northern dialect speakers had lower accuracy and longer response time under the condition of phonetic contrast. This result has revealed that phonetic contrasts (such as "na4") led to not only more time cost, but also a higher risk of making mistakes for both kinds of dialect speakers. Specifically, Hunan dialect speakers have more difficulty in perceiving and distinguishing phonetic contrasts than northern dialect speakers. Moreover, for Hunan dialect speakers, the result that both phoneme /n/ and phoneme /l/ had the low accuracy indicated the difficulty in distinguishing these two phonemes, which conforms to the type of SC assimilation.The accuracy of speech judgment of phoneme /f/ by Hunan dialect speakers was remarkably higher than that of phoneme /h/, but it was significantly lower than that of northern dialect speakers, manifesting that it was difficult to distinguish the phonetic contrast /f/-/h/, but it was relatively easier to differentiate the phoneme /f/ than phoneme /h/in Mandarin, which is in line with the type of CG assimilation. On the whole, the results of Experiment 1b has showed the same trend with Experiment 1a. Experiment 2 has demonstrated that compared with northern dialect speakers, Hunan dialect speakers had a longer response time and a lower rate of accuracy when making phonological judgment. Nevertheless, there was no significant difference in response time and accuracy between the two types of speakers under the condition of contextual phonetic contrast. These result proved that the difficulty in distinguishing phonetic contrasts /n/-/l/ and /f/-/h/ for Hunan dialect speakers can be disambiguated through lexical context, which indicated that the confusion of phonetic contrasts in Hunan dialect speakers is a process assimilation rather than a representational assimilation.
郭桃梅教授
北京师范大学认知神经科学与学习国家重点实验室教授、博士生导师
主要研究领域:
汉语及第二语言学习加工的认知神经机制研究
报告题目：
Look at a Leopard through a Tube: How a Bilingual’s Brain is Wired for Speaking Two Languages
报告摘要：
Proficient bilinguals can freely choose the appropriate language for communication according to the situation, and even switch back and forth between two languages. Given the fact that two languages are represented and processed within one brain, bilingualism provides a central through a tube at a leopard, i.e., investigating how speech is controlled by human brain. In this talk based on some of our recent work, I will try to address the following issues: How do bilinguals overcome interference from the non-target language and select correct words in the target language? How dose a bilingual brain work for producing words in one of the two languages? Are neural correlates for bilingual language control domain-general or language-specific? Overall, the findings of these studies contribute to our understanding of the universality and diversity across domains, and the neural plasticity in adult bilinguals.
明日将会继续更新专家预告，敬请期待来自引智基地的一线报道！
百问百答专栏
你有想对专家说的话吗？有想要和专家交流的问题吗？
是否对本次论坛的活动有疑问呢？
或者有哪些突然闪现的小tips.....
“百问百答”专栏将限时开通，收集来自您的奇思妙想，欢迎大家积极投递，我们期待与您的思想碰撞......
邮箱：clcb@blcu.edu.cn
