Effects of Prosody while Disambiguating Ambiguous Japanese Sentences in the Brain of Native Speakers and Learners of Japanese: A Proposition for Pronunciation and Prosody Training

Abstract

Recently, the significant role that pronunciation and prosody plays in processing spoken language has been widely recognized and a variety of teaching methodologies of pronunciation/prosody has been implemented in teaching foreign languages. Thus, an analysis of how similarly or differently native and L2 learners of a language use pronunciation/prosodic cues needs to be further investigated, and the learnability of pronunciation/prosodic features should be explored. In this study, the role of prosody in Japanese sentence processing will be specifically explored among native speakers and L2 learners of Japanese whose native language is English. In Experiment 1 and 2, the effect of prosody during Japanese sentence processing was explored among native speakers using a psycholinguistic measurement. In Experiment 3, we compared L2 learners processing and judgment of Japanese sentences utilizing the brain-imaging technique Electroencephalography (EEG) together with a psycholinguistic measurement. In Experiment 1 and 2, native speakers of Japanese listened to globally ambiguous sentences that can be interpreted in two ways, and temporarily ambiguous sentences that have two different syntactic structures. They either rated how acceptable each sentence is or answered a comprehension question on each sentence. As for the globally ambiguous sentences, the results revealed that overall one type (`embedded-clause') of interpretation is preferred over the other type (`main-clause') of interpretation at the judgment given time pressure. Prosody guided their interpretations to a certain degree; however, it did not have a deterministic effect, especially for arriving at a `main-clause' interpretation. As for the temporarily ambiguous sentences, significant effects of prosody in parsing temporarily ambiguous sentences were found, with the results suggesting that while parsing affects processing, its role is not deterministic. In Experiment 3, native speakers and intermediate- to advanced-level L2 learners of Japanese listened to two types of temporarily ambiguous sentences read with two types of prosody and rated how acceptable each sentence was. Simultaneously, their brain activity was continuously recorded using EEG. The results revealed important similarities and differences among the native speakers and L2 learners' processing of these sentences. Both groups yielded a brain response that indicates the detection of prosodic break, and prosody was utilized at least to some extent. However, the patterns were different among the two groups, and the precise nature of the effects for the learners suggests that they have difficulties with processing non-default-type of structure (`main-clause' structure), and the congruent prosody for that structure (`main-clause prosody'). These results indicate that L2 learners have access to prosodic cues in sentence comprehension. On the other hand, the measurements of processing presented here suggest that these learners are not yet utilizing prosody in a native-like way, suggesting the utility of creating new ways to introduce prosody and its relation with the structure and meaning of Japanese sentences. It is suggested that teaching how to use prosodic cues in comprehending complex sentences with various types of sentence structures may develop L2 learners' ability to develop their oral communication skills.