Syntactic complexity in adolescents' elicited and semi-spontaneous speech

This new paper (co-authored with Elliot Holmes) has been accepted for publication in the Journal of Child Language. It is based on research carried out initially with Lydia Gunning and Ekaterini Klepousniotou, as part of the pilot project Assessing core language skills in pre-adolescents in the Born-in-Bradford cohort.

Abstract: 

Little is known of syntactic development during early adolescence. Is complex syntax truly mastered at the beginning of secondary education? Are there differences between the knowledge of complex structures and the ability to use them?

Building on a critical review of the literature on syntactic complexity, we investigate these questions in a sentence repetition task and a narrative task, performed by 126 monolingual and bilingual 11- to-13-year-olds educated monolingually and growing up in areas affected by socioeconomic deprivation in the UK.

We demonstrate substantial individual variation associated with syntactic complexity in both tasks, and propose a novel index of syntactic complexity: the proportion of full subordinate clauses. Combining predictive regression models and random forest analyses, we identify the most important predictors of performance.

Drawing on these exploratory findings, we propose a research program to investigate difficulties associated with complex syntax in this age group. 

De Cat, C. and Holmes, E. (in press) Syntactic complexity in adolescents' elicited and semi-spontaneous speech. Journal of Child Language. 

Using Q-BEx to identify children needing specialist referral

Our study “Using Q-BEx to Identify Risk for Language Impairment in Bilingual Children” has been accepted for publication in the Journal of Speech, Language and Hearing research.

Alternative risk factor indices were derived from the Risk Factor and Proficiency modules differing in the weight and detail given to language proficiency. Our aim was to identify the optimal one to include in the Q-BEx backend calculator (which processes the data into spreadsheets and individual child reports).

Each index was applied to two datasets comprised of five- to eight-year-old children: one with independent diagnosis of Developmental Language Disorder (DLD)/Typical Development (TD) (109 bilingual children tested in France) and one of largely all-comer children (278 bilingual and monolingual children tested in France, the Netherlands, and the United Kingdom). We compared how these alternative indices predicted (likely) DLD/TD status, and structural language outcomes, assessed with LITMUS tools, designed for bilingual children.

Optimal balance of sensitivity and specificity was achieved by an index composed of four equally weighted components (age of first word, age of first sentence, early development concerns, and strongest speaking skills between the child’s languages).

• Clinical Application: The optimal risk factor index is now used by the Q-BEx backend calculator to trigger “Red Flags.” It serves as a first-level screening tool to identify children requiring formal diagnostic assessment via LITMUS or other specialist tools.
• Consistency: The index proved reliable across different linguistic environments and both monolingual and bilingual populations, ensuring objective triage for inclusive research sampling. 

De Cat, C., Tuller, L., Gusnanto, A., Kašćelan, D., Prévost, P., Serratrice, L., & Unsworth, S. (2026). Using Q-BEx to Identify Risk for Language Impairment in Bilingual Children. Journal of Speech, Language, and Hearing Research https://doi.org/10.31234/osf.io/57fyb_v2 (OSF preprint, accepted version)

How detailed do measures of bilingual language experience need to be? A cost-benefit analysis using the Q-BEx questionnaire

What is the optimal level of questionnaire detail required to measure bilingual language experience? This empirical evaluation compares alternative measures of language exposure of varying cost (i.e., questionnaire detail) in terms of their performance as predictors of oral language outcomes. The alternative measures were derived from Q-BEx questionnaire data collected from a diverse sample of 121 heritage bilinguals (5- to 9-years of age) growing up in France, the Netherlands and the UK. Outcome data consisted of morphosyntax and vocabulary measures (in the societal language) and parental estimates of oral proficiency (in the heritage language). Statistical modelling exploited information theoretic and cross-validation approaches to identify the optimal language exposure measure. Optimal cost-benefit was achieved with cumulative exposure (for the societal language) and current exposure in the home (for the heritage language). The greatest level of questionnaire detail did not yield more reliable predictors of language outcomes.

The preprint is available on the OSF, along with the data and script. This paper was accepted for publication in Bilingualism: Language and Cognition.

Individual language experience determinants of morphosyntactic variation in heritage and attriting speakers of Bosnian and Serbian: A causal inference approach

Using a causal inference approach, we explored the relationships among the language experience determinants of morphosyntactic sensitivity, to identify the factors that indirectly and directly cause its acquisition or maintenance in immigration contexts. We probed the sensitivity to Serbian/Bosnian clitic placement violations with a self-paced listening task, in a diverse group of bilinguals in Norway (n = 71), born to immigrant parents, or having emigrated in childhood or adulthood. The outcomes included a metalinguistic violation detection score and a listening/processing time difference between licit and illicit structures.

Structural Equation Models revealed that literacy (as reading practices) was among the most influential determinants of the ability to detect violations, while Bosnian/Serbian use across contexts and age of bilingualism onset determined violation sensitivity in processing. We identified a significant threshold of societal language (SL) exposure at age 8. Rather than SL exposure before this age precluding bilinguals from developing and maintaining morphosyntactic sensitivity, this threshold seems to reflect a protective effect against attrition which intensifies the later after age 8 SL exposure starts. The length of residence in Norway did not determine attrition, suggesting that heritage and attrited speakers should be considered on a continuum rather than as distinct bilingualism profiles. 

DOI: https://doi.org/10.1075/lab.24016.tom

Unpacking language richness as a predictor of bilingual children’s language proficiency

Bilingual children’s language abilities are influenced by the richness of their language experience, which is typically estimated using parental questionnaires and often expressed as a composite score based on frequency-based variables (e.g., time spent reading). We evaluated whether the composite richness score in the Q-BEx questionnaire was fit for purpose. Data were collected from 173 bilingual children aged between 5 and 8 in three different countries (France, the Netherlands and the UK). Parents completed the Q-BEx questionnaire and children completed proficiency tasks in their societal language. We analysed the predictive power of the original score in comparison to several alternatives, derived using a principal components analysis. We found that (i) these alternatives were no more informative than the original, (ii) scores including interlocutor diversity and proficiency in addition to frequency-based measures fared better, (iii) the latent variables underlying richness were comparable across languages, and (iv) whether SES was included made little difference.

Preprint DOI: https://doi.org/10.31219/osf.io/rquvc

Accepted for publication in the Journal of Child Language (August 2025)

So many variables, but what causes what? An invitation to Causal Inference Modelling

 De Cat, C., & Unsworth, S. (in press). So many variables, but what causes what? Journal of Child Language. (manuscript)

This is a reply to Johanne Paradis’ keynote article (Sources of individual differences in the dual language development of heritage bilinguals, JCL), which provides a comprehensive overview of factors influencing bilingual children’s dual language abilities. In this commentary, we provide an illustration of how the complex relationships between the variables discussed in Paradis’s keynote article could be conceptualised within a causal inference approach. We offer a modest starting point by summarising key features of causal inference modelling and by illustrating how it might help us better understand what causes what.

Using Latent Class Analysis to identify multilingual family profiles

Wood, Megam, Gunning, Lydia, & De Cat, Cécile. (2023). Achieving representativity in opportunity sampling: the ‘Bradford effect’ in the multilingual families Covid-19 survey. International Journal of Multilingualism, 1–22. https://doi.org/10.1080/14790718.2023.2284291

Typically, families from ethnic minority backgrounds and socioeconomic disadvantage are underrepresented in research. Using secondary data from a survey of the language practices of multilingual families during the first wave of the Covid-19 pandemic, we sought to ascertain whether the large proportion of Bradford-based respondents affected the representativeness of the sample and/or the patterns of responses. Respondents were objectively categorised into five latent profiles, based on the demographics of the household and families’ engagement with the Heritage Language (HL). The three of the household profiles considered “engaged” with their HL reported more positive attitudes to multilingualism. Their language practices also more markedly changed during the Lockdown compared to families who were less engaged with their HL. One of the five profiles was deemed to represent those usually considered “Hard-to-reach” in research (i.e., ethnic minority, low socio-economic status). The “Hard-to-reach” and “Unengaged” profiles were mainly represented by Bradford families. We argue that the exceptional engagement of Bradford communities in research resulted in better representativity of family profiles in this national survey.

How to quantify bilingual experience? Findings from a Delphi consensus survey

De Cat, C., Kašćelan, D., Prévost, P., Serratrice, L., Tuller, L., Unsworth, S., & consortium, Q.-B. (2023). How to quantify bilingual experience? Findings from a Delphi consensus survey. Bilingualism: Language and Cognition, 26(1), 112-124. doi:10.1017/S1366728922000359

While most investigations of bilingualism document participants’ language background, there is an absence of consensus on how to quantify bilingualism. The high number of different language background questionnaires used by researchers and practitioners jeopardises data comparability and cross-pollination between research and practice. Using the Delphi consensus survey method, we asked 132 panellists (researchers, speech and language therapists, teachers) from 29 countries to rate 124 statements on a 5-point agreement scale. Consensus was pre-defined as 75% agreement threshold. After two survey rounds, 79% of statements reached consensus. The need for common measures to quantify bilingualism was acknowledged by 96% of respondents. Agreement was reached to document: language exposure and use, language difficulties, proficiency (when it cannot be assessed directly), education and literacy, input quality, language mixing practices, and attitudes (towards languages and language mixing). We discuss the implications of these findings for the creation of a new tool to quantify bilingual experience.

This publication is one of the outputs from the Q-BEx project.

Open Science practices in applied linguistics: results from a preliminary survey

 With Men Liu, we have analysed and interpreted the results from the brief survey I carried out in spring 2021. These have been written up as a chapter for this forthcoming volume: Plonsky, L. (Ed.). Open Science in Applied Linguistics. John Benjamins. It can be downloaded from OSF preprints.  The data and code are also available via the /OSF.

A constellation of continua: Reconceptualising bilingualism, autism and language research

Kašćelan, Drasko and De Cat, Cécile (2022). A constellation of continua: Reconceptualising bilingualism, autism and language research.  Linguistic Approaches to Bilingualism 12(1): 59-64. https://doi.org/10.1075/lab.21069.kas or pdf 

This is a commentary on the keynote article "Bilingual language development in autism" by Philippe Prévost and Laurie Tuller. 

A review of questionnaires quantifying bilingual experience in children: Do they document the same constructs?

First published output from the Q-BEx project: 

Kašćelan, Drasko, Prévost, Philippe, Serratrice, Ludovica., Tuller, Laurie, Unsworth, Sharon, and De Cat, Cécile (2022).  A review of questionnaires quantifying bilingual experience in children: Do they document the same constructs? Bilingualism: Language and Cognition 25(1): 29-41. Preprint available on the OSF. DOI: https://doi.org/10.1017/S1366728921000390 

Abstract: Questionnaires documenting children's bilingual experience have been used frequently in research on language and cognitive development. However, there has been little investigation of the comparability between these tools. In this review, we (i) provide a list of available questionnaires used to quantify bilingual experience in children; (ii) identify the components of bilingual experience documented across questionnaires; and (iii) discuss the comparability of the measures used to operationalise these components. In doing so, we review 48 questionnaires and identify 32 overarching constructs, manifested as 194 components, and we calculate the frequency with which they are documented. Finally, by focusing on a subset of overarching constructs (language exposure and use, activities, and current language skills), we observe high variability in how they are operationalised across tools. These findings highlight the need for greater transparency in how we document bilingualism and for more comparable measures.

Assessing EAL Children’s Oral Language: A Survey of Professional Practices in the UK

Originally in preparation for a grant application, we (i.e., a group of language acquisition researchers from across the UK) carried out a survey of professional practices in the language and cognitive assessment of young children in the UK. The full results are reported in the first version of this preprint, which was then superseded by a version focusing just on language.  More extensive summaries of the responses is available via the supplementary materials. The final list of authors includes Emily Oxley, Allegra Cattani, Vicky Chondrogianni, Caroline Floccia, Laurence White and  Cécile De Cat.

What does the CELF Sentence Structure test really measure in monolinguals and bilinguals?

This paper is now available in the Journal of Child Language via FirstView

Abstract

The Sentence Structure sub-test (SST) of the Clinical Evaluation of Language Fundamentals (CELF) aims to “measure the acquisition of grammatical (structural) rules at the sentence level”. Although originally designed for clinical practice with monolingual children, components of the CELF, such as the SST, are often used to inform psycholinguistic research. Raw scores are also commonly used to estimate the English proficiency of bilingual children. This study queries the reliability of the SST as an index of children's ability to deal with structural complexity in sentence comprehension, and demonstrates that cognitive complexity induces a considerable confound in the task, affecting 5- to 7-year-old monolinguals (n = 87) and bilinguals (n = 87) alike.

Award for "Predicting proficiency in bilingual children"

 I'm deeply honoured that this paper has been awarded the Albert Valdman award 2020 for outstanding publication in Studies in Second Language Acquisition.

De Cat, Cécile (2020). Predicting language proficiency in bilingual children. Studies in Second Language Acquisition, 42(2): 279-325.  (doi:10.1017/S0272263119000597)

Delphi consensus survey on how to document bilingual experience

 The Delphi consensus survey we carried out as part of the Q-BEx project is now available as a preprint: 

De Cat, C., Kašćelan, D., Prevost, P., Serratrice, L., Tuller, L., & Unsworth, S. (2021, February 5). Delphi consensus survey on how to document bilingual experience. Available from osf.io/ebh3c

Highlights:

• We review 48 questionnaires quantifying bilingualism in children 
• We report the frequency of documenting 32 constructs and 194 components across tools 
• We reveal low comparability in several estimates (e.g., exposure and use)
• We argue for greater transparency and comparability of bilingualism measures

Quantifying bilingual experience: the state of the art

As part of the Q-BEx project we have produced a review of the state-of-the-art in bilingual experience questionnaires.  It is now available as a preprint:

Kašćelan, D., Prevost, P., Serratrice, L., Tuller, L., Unsworth, S., & De Cat, C. (2020). A Review of questionnaires quantifying bilingual experience in children: Do we document the same constructs? Available from the OSF: 10.31219/osf.io/w4kef  

Highlights from the paper:

• We review 48 questionnaires quantifying bilingualism in children
• 32 constructs and 194 components across tools are identified. We report how frequently they are documented across questionnaires.
• The review reveals low comparability in several estimates (e.g., exposure and use).
• We argue for greater transparency and comparability of bilingualism measures

Socio-economic status as a proxy for input quality in bilingual children?

This study investigates the effect of socio-economic status (SES) as a proxy for input quality, in predicting language proficiency. Different operationalizations of SES are compared, including simple measures (parental education and parental occupation) and complex measures combining two dimensions (among parental education, parental occupation, and deprivation risk). All significantly predict overall English proficiency scores in a diverse group of 5- to 7-year-olds acquiring English and another language. The most informative SES measure in that respect is shown to be a complex measure combining parental education and parental occupation. That measure is used in a second set of analyses showing that different aspects of language are affected differently by variations in SES and in language exposure.

Now published as open access in journal of Applied Psycholinguistics. 

Preprint: https://osf.io/5y2wt  Final publication: http://dx.doi.org/10.1017/S014271642000079X  

What does the CELF Sentence Structure test really measure?

ACCEPTED FOR PUBLICATION IN THE JOURNAL OF CHILD LANGUAGE in December 2020

The Sentence Structure sub-test (SST) of the Clinical Evaluation of Language Fundamentals (CELF) aims to “measure the acquisition of grammatical (structural) rules at the sentence level”. Although originally designed for clinical practice with monolingual children, components of the CELF, such as the SST, are often used to inform psycholinguistic research. Raw scores are also commonly used to estimate the English proficiency of bilingual children. This study queries the reliability of the SST as an index of children’s ability to deal with structural complexity in sentence comprehension, and demonstrates that cognitive complexity induces a considerable confound in the task, affecting 5- to 7-year-old monolinguals (n = 87) and bilinguals (n = 87) alike.

Preprint of the full paper   OnlineFirst version (June 2021)

Most disappointingly, Pearson did not grant permission to reproduce any of the pictorial prompts. Text descriptors are used instead in the paper. 

Predicting language proficiency in bilingual children

Now published in Studies in Second Language Acquisition 
doi:10.1017/S0272263119000597
The data and R code are available via the OSF (see link below)

ABSTRACT: Using advanced quantitative methods, this paper demonstrates that cumulative exposure to the school language is the best language experience predictor of proficiency in that language (as indexed by sentence repetition, lexical semantic and discourse semantic tasks) in a highly diverse group of 5- to 7-year-old bilingual children in monolingual education. An objective method is proposed to identify the amount of school language experience beyond which bilingual children are likely to perform within the monolingual range, and show that relative passivity in the home language does not translate into better school language proficiency. Socio-economic status is shown to interact in complex ways with language exposure, such that it is only above a certain level of exposure to the school language that the benefits of a more privileged background have a tangible impact on school language proficiency. To tease apart the effect of environmental predictors from the effect of cognitive factors, memory and cognitive flexibility measures are included as covariates in all analyses.

https://osf.io/f5q98/

Intervention effects in wh-chains: the combined effect of syntax and processing

This study in experimental syntax investigates the factors affecting the acceptability of embedded clauses featuring a left-dislocated phrase below a fronted wh-phrase. Sixty native speakers of French took part in an on-line acceptability judgment task including 45 critical items (with an intervening XP) and 20 baseline items (including grammatical and ungrammatical sentences with an embedded wh-dependency). Using Random Forest and Ordinal Regression analyses we demonstrate that Clitic Left Dislocated (CLLD) objects yield stronger intervention effects (except when they are pronouns) than CLLDed subjects. We argue this is due to excessive processing demands incurred when a wh-dependency features a CLLD chain that is not fully within its scope. A processing account
also explains why pronouns are not disruptive of wh-chains.

Barbosa, P., & De Cat, C. (2019). Intervention effects in wh-chains: the combined effect of syntax and processing. Glossa: A Journal of General Linguistics, 4(1), 127. DOI: http://doi.org/10.5334/gjgl.1005