Effects of Cross-Linguistic Interactions on Perception of Speech

Project Information

E-CLIPS

Grant agreement ID: 799554

DOI

10.3030/799554

Project closed

EC signature date 18 July 2018

Start date 1 July 2019

End date 30 June 2021

Funded under

• EXCELLENT SCIENCE - Marie Skłodowska-Curie Actions

Total cost

€ 158 121,60

EU contribution

€ 158 121,60

Coordinated by

BCBL BASQUE CENTER ON COGNITION BRAIN AND LANGUAGE
Spain

Periodic Reporting for period 1 - E-CLIPS (Effects of Cross-Linguistic Interactions on Perception of Speech)

Reporting period: 2019-07-01 to 2021-06-30

Summary of the context and overall objectives of the project

What is the problem/issue being addressed?

Bilinguals often communicate with one another through spoken interactions in a non-native, second language (L2). Typical environmental factors that compromise the quality of the speech signal, like noise, can have an even greater detrimental effect on the comprehension of an L2. The ability to accurately recognize L2 words, in noise, depends on available context as well as individuals’ prior lexical knowledge. For bilinguals, this lexical knowledge includes words in two language systems, which interact with one another. The effects of cross linguistic interactions on L2 word recognition depends on the similarity between lexical forms across languages, however, how these effects are modulated by noise is not known. The overall goal of this proposal was to elucidate the functional and neural architecture that supports accurate L2 word recognition as a function of cross linguistic interactions, context, and listening condition. To this end, we had three multi-methodological objectives (behavioral, fMRI, electrophysiological) that addressed questions about the influence of L1 on L2 word recognition, in noise, as a function of crosslinguistic lexical similarity.

Why is it important for society?

The ability to communicate through spoken language is fundamental to human interactions and quality of life. A surge of technological developments have facilitated cross-cultural spoken interactions, on a global scale, among bilingual speakers. However, typical listening environments are often susceptible to noise and other types of interference that affect the quality of the speech signal, and are most detrimental for non-native language comprehension. Thus, understanding the factors that affect second-language listening, their interactions with noise, and the mechanisms that can be used to compensate for noise is of great value to our evermore globalized society.

What are the overall objectives?

Objective 1 characterizes the functional architecture underlying the mapping of sound onto words, in L2 listeners, as a function of its potential interactions with L1 lexical forms.

Objective 2 identifies the underlying network of brain regions, and their functional connections, that support accurate word recognition, in noise as a function of cross linguistic interactions and semantic feedback using fMRI.

Objective 3 aimed to explore the neurophysiological basis of the temporal dynamics that contribute to cross linguistic effects on L2 word recognition, using electrophysiological measures.

Work performed from the beginning of the project to the end of the period covered by the report and main results achieved so far

WP1
Description
The goal of Work Package 1 was to test the hypothesis that crosslinguistic effects on L2 word recognition, in noise, depends on the phonological overlap between translation equivalents. Specifically we investigated whether translation equivalents were more susceptible to the number of phonological neighbors in L1, activated by the L1 translation, if they share phonological form. To this end, we manipulated listening condition, cognate status, and the phonological neighborhood density of the L1 translation.

Results
The results show that crosslinguistic interactions are modulated by noise and affect L2 auditory word recognition, in Spanish-English (L1-L2) bilinguals, as a function of phonological overlap. Specifically, the findings show that, in noise, phonological overlap between translation equivalents leads to impairments on L2 word recognition, for L1 translations that have a high number of phonological neighbors. This finding supports a previously made claim that during auditory word recognition, in noise, crosslinguistic interactions make L2 word recognition more susceptible to effects of crosslinguistic competition.

WP2

Description
The goal of Work Package 2 was to determine the underlying neural systems that support accurate L2 word recognition, in noise, as a function of cross linguistic phono-lexical overlap between translation equivalents, and semantic context.

Results
The results showed that the engagement of a set of regions, often referred to as the bilingual language control network, was upregulated when cross linguistic interactions promoted cross linguistic lexical competition and down regulated by semantic priming, suggesting that semantic access helps mitigate effects of cross linguistic lexical interactions that generate competition. Interestingly, the regions are similar to those previously identified in monolingual studies of lexical competition.

WP 3
Description
The original goal of Work Package 3 was to relate cross linguistic phonological effects to neural oscillations across different frequency bands using EEG. This work package was not completed due to Covid-19 restrictions. Based on the findings in the other two packages, we will pursue a related goal, for a future experiment.

General results, exploitation, dissemination:
The findings of this project has provided insight into some of the daily challenges, typical of the L2 listening experience and how they can be overcome to improve accurate word recognition, in noise. These results will advance theoretical models of bilingual lexical processing and contribute to a more detailed functional architecture that specifies the nature of the of cross linguistic interactions across, phonological, lexical, and semantic levels of processing. One set of findings has been published and the rest are being submitted for publication.

Progress beyond the state of the art and expected potential impact (including the socio-economic impact and the wider societal implications of the project so far)

The results of the current proposal have contributed to our understanding of bilingual auditory word recognition showing clearly that noise modulates the effect of cross linguistic interactions, as a function of phonological and semantic overlap. Specifically, we show that semantic feedback and cognitive control are two complementary systems that support L2 word recognition, in noise: When words share form, across languages, conditions that promote cross linguistic lexical competition, upregulate cognitive control, which can be downregulated by the availability of a predictable semantic context (see Figure in Section 4).

The findings help understand flexibility in bilingual spoken language processing, as listeners contend with interactions between their two languages. They will also advance theoretical models of bilingual lexical processing contributing to a more detailed functional architecture that specifies the nature of the of cross linguistic interactions across, phonological, lexical, and semantic levels of processing as a function of listening condition. Finally, they bridge two areas of research (speech perception and bilingualism) providing a more unified framework of lexical processing showing that many of the same flexible mechanisms that support flexible perception, more generally, are at play. As a result, the findings also have implications for education, technological advances, and clinical populations that suffer from hearing decline or impairments caused by brain injury.

Summary of findings