Bastarrika Ainhoa, Davidson Douglas J
Basque Center on Cognition, Brain and LanguageDonostia, Spain; Department of Linguistics and Basque Studies, University of the Basque CountryGasteiz, Spain.
Basque Center on Cognition, Brain and Language Donostia, Spain.
Front Hum Neurosci. 2017 Jan 24;11:12. doi: 10.3389/fnhum.2017.00012. eCollection 2017.
The present study used magnetoencephalography (MEG) to investigate how Spanish adult learners of Basque respond to morphosyntactic violations after a short period of training on a small fragment of Basque grammar. Participants ( = 17) were exposed to violation and control phrases in three phases (pretest, training, generalization-test). In each phase participants listened to short Basque phrases and they judged whether they were correct or incorrect. During the pre-test and generalization-test, participants did not receive any feedback. During the training blocks feedback was provided after each response. We also ran two Spanish control blocks before and after training. We analyzed the event-related magnetic- field (ERF) recorded in response to a critical word during all three phases. In the pretest, classification was below chance and we found no electrophysiological differences between violation and control stimuli. Then participants were explicitly taught a Basque grammar rule. From the first training block participants were able to correctly classify control and violation stimuli and an evoked violation response was present. Although the timing of the electrophysiological responses matched participants' L1 effect, the effect size was smaller for L2 and the topographical distribution differed from the L1. While the L1 effect was bilaterally distributed on the auditory sensors, the L2 effect was present at right frontal sensors. During training blocks two and three, the violation-control effect size increased and the topography evolved to a more L1-like pattern. Moreover, this pattern was maintained in the generalization test. We conclude that rapid changes in neuronal responses can be observed in adult learners of a simple morphosyntactic rule, and that native-like responses can be achieved at least in small fragments of second language.
本研究采用脑磁图（MEG）来探究西班牙巴斯克语成年学习者在接受一小部分巴斯克语语法的短期训练后，对形态句法违反情况的反应。参与者（n = 17）在三个阶段（预测试、训练、泛化测试）接触违反和对照短语。在每个阶段，参与者听简短的巴斯克语短语，并判断其正确与否。在预测试和泛化测试期间，参与者没有收到任何反馈。在训练阶段，每次反应后都会提供反馈。我们还在训练前后进行了两个西班牙语对照阶段。我们分析了在所有三个阶段中记录的对关键单词的事件相关磁场（ERF）。在预测试中，分类低于随机水平，我们发现违反和对照刺激之间没有电生理差异。然后，向参与者明确教授了一条巴斯克语语法规则。从第一个训练阶段开始，参与者能够正确分类对照和违反刺激，并且出现了诱发的违反反应。尽管电生理反应的时间与参与者的第一语言效应相匹配，但第二语言的效应大小较小，并且地形分布与第一语言不同。第一语言效应在听觉传感器上双侧分布，而第二语言效应出现在右额叶传感器上。在训练的第二和第三阶段，违反 - 对照效应大小增加，地形演变为更类似于第一语言的模式。此外，这种模式在泛化测试中得以维持。我们得出结论，在简单形态句法规则的成年学习者中可以观察到神经元反应的快速变化，并且至少在第二语言的小片段中可以实现类似母语的反应。
