Next CLLEAR seminar: Thinking ahead in a second language: On the role of prediction in L2 processing

CLLEARThe next Centre for Linguistics, Language Education and Acquisition Research (CLLEAR) seminar will take place on Wednesday 9th May 2018 at 4pm in Room 1177, Building 65, Avenue Campus. The talk is entitled “Thinking ahead in a second language: On the role of prediction in L2 processing” and will be delivered by Theres Gruter from Hawaii. All welcome!

Here is the abstract for this seminar:
The role of prediction in native language (L1) processing has been investigated, and debated, extensively over the past couple of decades. Yet it is only in the last few years that prediction/anticipation in second language (L2) processing has become a topic of interest. In this talk, I will discuss how the investigation of prediction in L2 processing may help us move beyond the common but rather unsatisfying description of differences between L1 and L2 speakers as L2 learners having “a processing problem”. In recent and on-going research in our lab, we have used online (visual-world eye-tracking) and offline methods to probe to what extent L2 listeners engage in proactive ‘thinking ahead’ during sentence and discourse processing. Drawing on findings from studies targeting various linguistic cues that can give rise to anticipatory processing – including classifiers in Mandarin Chinese and grammatical aspect in English – I will argue that L2 speakers do not necessarily differ from L1 speakers in whether or not they engage in prediction, but in how and when they engage in prediction, and what information they use to generate expectations about upcoming information. Taken together, these findings suggest that prediction is a universal mechanism of human language processing (and behaviour more generally), and that L1 and L2 speakers make adaptive use of this mechanism depending on its utility given their knowledge and processing goals.

Next CLLEAR seminar: Methodological considerations in measuring ambiguous relative clause attachment strategies in bilinguals

CLLEARThe next Centre for Linguistics, Language Education and Acquisition Research (CLLEAR) seminar will take place on Monday 16th April 2018 at 15:30 in Room 1011, Building 67, Highfield Campus. The talk is entitled “Methodological considerations in measuring ambiguous relative clause attachment strategies in bilinguals” and will be delivered by Elena Valenzuela from the University of Ottawa. All welcome!

Here is the abstract for this seminar:
It has been argued that monolinguals and bilinguals differ in how they resolve ambiguities in relative clause attachment. Cuetos and Mitchell (1988) first noted that sentences as in (1) and (2), which contain a complex NP of the type “NP of NP” followed by a relative clause (RC), are parsed differently depending on the language:
(1) She kissed the brother(NP1) of the poet(NP2) that was on the balcony.
(2) Elle a embrassé le frère(NP1) du poète(NP2) qui était sur la balcon.
In English (1), the poet is on the balcony whereas in the same sentence in French (2), it is the brother who is on the balcony. Languages can be grouped according to the parsing strategy for monolinguals: high attachment (Spanish, French, Greek, Italian, Japanese, etc.) and low attachment (English, Arabic, Brazilian Portuguese, Romanian, etc.).

Dussias and Sagarra (2007) found that language dominance was the Spanish-dominant bilinguals with limited exposure to English preferred high attachment in both languages, while bilinguals with extensive exposure to English preferred low attachment in both English and Spanish. Valenzuela et al. (2015) examined the parsing strategies of bilinguals in code-switched sentences and also found that language dominance and exposure played the greatest role for parsing. However, in these previous studies, participants were all living in an English environment at the time of testing which may have influenced so-called language dominance. In our study we test French/English bilinguals living their dominant language environment.
This research examines parsing strategies in monolingual and code-switched sentences to address the following research questions:
i. Does language dominance play a role in parsing strategies?
ii. Does direction of the language code-switch affect processing?
iii. Does the direction of the language code-switch affect processing differently based on individual’s language dominance?

Two groups of bilinguals: French/English living in English dominant environment (n=15) and French/English living in French dominant environment (n=14) were tested on their parsing strategies of French, English, and French/English code-switched ambiguous relative clauses. Participants were given two experimental tasks: Sentence Judgment Task and Sentence Completion Task. Results show that across the board low attachment was preferred regardless of the language environment and language dominance of the participant. This may suggest that, as in Dussias and Sagarra (2007), language exposure plays an important role. However, this may also be due to the prolonged language contact situation in Canada resulting in an emerging dialect. Results will be discussed in terms of language dominance, frequency of language usage, and appropriateness of the methodology used.

Next CLLEAR seminar: Supporting Heritage Language Development

CLLEARThe next Centre for Linguistics, Language Education and Acquisition Research (CLLEAR) seminar will take place on Friday 9th March 2018 at 16:00 in Lecture Theatre B, Building 65, Avenue Campus. The talk is entitled “Supporting Heritage Language Development and will be delivered by Silvina Montrul from the University of Illinois at Urbana-Champaign. All welcome for the seminar and discussion!

Here is the abstract for this seminar:
Defined by their transmission across a generation, heritage languages are spoken by the bilingual children of immigrant parents. There is little consensus in the U.S. today about the language education of English Language Learners (ELLs) who I refer to as heritage language speakers: while some call for the early teaching of English, others insist that this can have negative repercussions for the full development and maintenance of the heritage language. Similar concerns affect heritage speakers in other parts of the globe. In this talk, I challenge widely held ideas about native language proficiency: namely, that once acquired early in childhood, a language is stable, especially in adults. I argue instead that native language proficiency can be shaped by the environment, and this is particularly true for U.S. bilinguals. In contrast to monolingual native speakers, the language mastery of heritage speakers by early adulthood is often significantly different from that of both their immigrant parents and native speakers in the home country. Heritage speakers, like all speakers, are born with the ability to learn one or more languages fully and indeed retain native ability in selected grammatical areas due to their early exposure to the language, when compare to second language learners who start acquisition of the second language much later, for example. I will show how insufficient use of the language during late childhood and adolescence can profoundly affect specific aspects of their command of grammar, interrupting the full development of the language and turning their native language into a second language. I will conclude with suggestions for supporting heritage language development during the language learning period.

Next CLLEAR seminar: “Once a native, always a native? Language attrition and constraints on bilingual development”

CLLEAR

The next Centre for Linguistics, Language Education and Acquisition Research (CLLEAR) seminar will take place on Wednesday 11th January 2017 from 4:00-5:30pm in Lecture Theatre C, Building 65, Avenue Campus. The talk is entitled “Once a native, always a native? Language attrition and constraints on bilingual development” and will be delivered by Professor Monika Schmid from the University of Essex. All welcome for the seminar and discussion!

Here is the abstract for this seminar:
Bilinguals are not, as François Grosjean so famously pointed out, “two monolinguals in one person”. They use language differently from monolinguals, they differ from them in terms of processing, of acquisition, in their performance on controlled tasks and so on. We know this to be true, and yet it does not seem to have informed our research to the degree that it should: When we try to assess proficiency levels, probe underlying representations, investigate language production or processing, and so on, among L2 users – we still tend to compare them, as far as possible, to a monolingual reference group. Does it make sense to compare two groups that we know a priori to be different in order to find out that they are indeed different?

I will argue that in order to answer some of the most pressing questions in bilingualism research nowadays, such as whether language acquisition in childhood is qualitatively or merely quantitatively different from language acquisition later in life, we should invoke L1 attrition as part of the bilingual equation. We can thus put the populations that we compare on an equal footing with respect to their being bilinguals. In other words, we should not compare monolinguals and bilinguals, but dominant and non-dominant languages. In the case of L1 attriters, the non-dominant language is the one which was acquired as the first and only language in childhood (and was thus not subject to any maturational constraints). In the case of L2 learners, the non-dominant language (ie., the language that we are interested in) was acquired later in life, after the first language had been established.

Such a comparison has the potential of separating those linguistic factors that are vulnerable to cross-linguistic interference in both early- and late-learned languages (and on which both populations differ from monolingual controls) from those that might indeed have been affected by some kind of a Critical Period (which are stable in attriters but variable in L2 speakers).

I will illustrate this argument with data from a number of ongoing investigations, using behavioral measures, free speech data and evidence from neuroimaging studies.

Next CLLEAR seminar: “The effect of computational complexity on L1 transfer: evidence from L2 Chinese attitude-bearing wh-questions”

The next Centre for Linguistics, Language Education and Acquisition Research (CLLEAR) seminar will take place on Tuesday 12 May 2015 from 5:00-7:00pm in Building 46, Lecture Theatre B, Room 2003, Highfield Campus. The talk is entitled “The effect of computational complexity on L1 transfer: evidence from L2 Chinese attitude-bearing wh-questions” and will be delivered by Dr. Boping Yuan (Cambridge University). All welcome! Read more…