Project 1: The acquisition of idiomatic language by second language learners
This project aims to investigate how specific teaching/training contributes to the acquisition of idiomatic expressions in a second language (L2). Even advanced L2 learners are often recognized as non-native speakers by their failure to process idiomatic language in a native-like fashion and the incorrect use of idiomatic expressions can cause misunderstandings or induce a processing burden in native listeners. These findings underline the importance of acquiring idiomatic language in L2. Previous investigations have indicated that studies abroad offering L2 immersion support the acquisition of idiomatic language. Against this background, it is interesting to investigate alternative ways of providing immersion conditions favouring explicit learning of idiomatic language in the L2.
In this project Dutch L2 learners receive specific instruction and intensive training in idiomatic language through an ASR-based CALL system. The data collected will provide insight into L2 idiomatic language acquisition. The knowledge acquired and the resources developed in this project can be used to the benefit of L2 teaching.
Project 2: Comprehension of idiomatic language in different stages of L2 learning
The aim of project 2 is to study the comprehension of L2 idiomatic language, as a function of frequency, transparency, language overlap and usage. Research into the processing and comprehension of idiomatic language has suggested that figurative language may be processed faster and in a different way than literal language. This project aims to investigate how exactly this type of language is processed in comprehension. There has been little research on bilingual comprehension of idiomatic language and the venture involves several methodological challenges. The present project aims to resolve these issues and to add to the body of knowledge on this topic.
Project 3: Production of idiomatic language in different stages of L2 learning
The aim of project 3 is to study the production of L2 idiomatic language, as a function of frequency, transparency and usage. One of the reasons why idiomatic spoken language has attracted the attention of researchers working in psycholinguistics and language acquisition is that idiomatic expressions appear to contribute to reducing cognitive load and promoting fluency. Our hypothesis is that frequency, transparency and usage are likely to affect the production of idiomatic language. However, there has been little research on how these variables impact L2 idiomatic language production, which is partly related to the complexity involved in setting up controlled studies of such phenomena. The present project contributes to filling this gap.
Project 4: A computational model of the processing of idiom-containing sentences
Idioms offer a challenging case for cognitive models of competition-based, incremental processing, whereby multiple possible completions of a sentence may be considered in parallel as that sentence is processed word-by-word. Specifically, existing cognitive models of this type tend to differentiate competing completions on the basis of differing syntactic parses. However, in the case of idioms, syntactic structure is a constant while what differs is typically the conceptual structure—whether or not the idiom is interpreted literally or figuratively.
In project 4, we develop and implement a new model of incremental processing that accounts for this competition between literal and figurative interpretations. We draw on the empirical results from the other projects and 1) fit the parameters of the model on a training set sampled from these results, and 2) demonstrate the efficacy of the model in predicting empirical results using a complementary test set. Importantly, the architecture of this model allows it to model processing in L2 speakers as well as in natives. For more information on computational modeling, click here.