A story of stories: a large-scale cross-linguistic study of young children’s narratives

A story of stories: a large-scale cross-linguistic study of young children’s narratives | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article A story of stories: a large-scale cross-linguistic study of young children’s narratives Natalia Gagarina, Manuel Bohn, Ute Bohnacker, Sharon Armon-Lotem, and 36 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-5672219/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract This study investigates children’s narrative skills, using the Multilingual Assessment Instrument for Narratives (MAIN), a theory-based tool adapted to 100 languages. How do children’s overall narrative skills, including the use of factual and inferred components, and the development of complete episodes, differ across languages and between monolinguals and bilinguals? To answer this, we examined 2608 comparable fictional narratives of 1189 monolingual and bilingual children aged 3-13 years, speaking 33 languages. Children told or retold stories elicited with parallel picture sequences. Narratives were analyzed for inclusion of five essential story components (internal states as initiating events, goals, attempts, outcomes, internal states as results), overall narrative structure, use of factual and inferred information, and presence of complete episodes. Monolinguals initially outperformed bilinguals, but bilinguals’ narrative skills developed more rapidly and eventually surpassed those of monolinguals, with all children performing better in story retelling than storytelling, especially in their L2. Across languages, children were more likely to produce factual components (outcomes/attempts) than inferred ones (internal states/goals). The likelihood of producing a complete narrative increased with age; monolinguals consistently outperformed bilinguals, even under stricter coding criteria. Patterns of use for factual and inferred components were consistent across languages, suggesting universal strategies in narrative development. Humanities/Language and linguistics Social science/Development studies Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Introduction The ability to produce comprehensible oral stories or narratives is an indispensable part of adequate functioning in society. It forms the basis of human communication and has an impact on social well-being and thus, health 1 , 2 . Fledgling narrative skills emerge in the early preschool years and have been said to show to strongly develop up to age nine 3 , 4 , 5 , 6 , 7 . Narrative skills are crucial for later preschool and primary school development, bridging oral and literate language 8 . They are strongly linked to early literacy, reading, and overall school and life success 9 , 10 , 11 , 12 , 13 . Given their importance for social communication, it is not surprising that there is a wealth of studies on the acquisition of narratives, but these studies mainly examine narrative development in monolingual children and predominantly for English and, to some extent, Spanish 14 , 6 , 15 , 16 , 17 , 18 , 19 , 20 . These studies are built on theories of narrative organization known as story grammars, which are inspired by the structure of Western fairytales and are not always applicable to non-WEIRD (Western Educated Industrialized Rich Democratic) contexts. One the most prominent theories continues to be Stein and Glenn’s (1979) 14 story grammar, according to which a well-formed narrative includes a number of components: a setting (which introduces the characters and describes the non-linguistic context of a story in time and space), an initiating event, internal response, internal plan, attempt, direct consequence and a reaction. Empirical studies of narrative acquisition have however grappled with identifying these proposed components in authentic narrative data, yielding non-comparable results. Other approaches have used different numbers and types of components 21 , 22 , 6 , 23 , 24 , 17 . These theories have in common that they cannot always be verified against empirical narrative material. Other challenges are methodological. Whilst there are many publications on the higher-order organization of narratives, no study has provided results based on transcribed quantified data for more than 240 children 15 , 16 , 25 , 20 , 26 , 3 , 27 , in fact, frequently sample size is only 3 to 50 children, across varying age ranges 28 , 3 , 29 . Some studies report results for telling, i.e. story generation from the pictorial stimuli, others for the retelling of a story that an adult had first told to a child, and some also compare these two different modes, but not on same elicitation material 27 . Moreover, most picture-based acquisition studies have used elicitation materials that are convenience-based and not linguistically motivated, such as random wordless picture books, e.g., Frog where are you? 28 , 3 about a boy and his run-away pet frog. Alternatively, single pictures, cartoons, or other picture sequences are used that are often eurocentric and culture-specific, e.g., the Bus Story Test featuring a ‘bobby’ policeman, a red double-decker bus and bucolic scenery – all typical for the UK 29 , or the Test of Narrative Language , depicting a meal at an American fast-food chain or aliens in a spaceship landing in a park 30 . Further, such elicitation materials do not include sufficiently parallel pictorial stimuli to assess the two languages of bilingual children in a comparable manner. In order to overcome these theoretical and methodological shortcomings, as part of the COST Action IS0804 Language Impairment in a Multilingual Society: Linguistic Patterns and the Road to Assessment , a research group of linguists, developmental psychologists and speech-language pathologists from over twenty countries (from Northern Europe to South Africa) developed a multidimensional theoretical model for use with diverse monolingual and bilingual populations, different languages, and cultures. This model has been translated into an open-source narrative elicitation tool Multilingual Assessment Instrument for Narratives 31 , 32 (MAIN) which is now successfully being used by more than 3200 registered users in 97 languages in 67 countries, primarily with monolingual and bilingual children aged three to twelve. MAIN includes picture-based stories, instructions for use, scoring sheets, short questionnaires, examples of well-formed narratives, etc. ( https://main.leibniz-zas.de ). This linguistic model operates within and across several dimensions of causally-temporally organized minimal constituents of events, methodically translated into episodes. Episodes are centered around protagonists (in this model these are typically animals that are familiar to children across a variety of cultures, see Fig. 1). An episode consists of five integral components: an initiating internal state, a goal, an attempt, an outcome, and a resulting internal state. As stories typically consist of several episodes, our model includes three episodes to provide multiple opportunities for a storyteller to verbalize these components. Their quantification is the first dimension of the model. The categorization of these components into two types, factual and inferred, is the second dimension of the model. Factual components are directly observable events, such as a protagonist’s physical actions (Attempts) and their immediate consequences (Outcomes). For example, in an episode, one might observe a cat climbing a tree (Attempt) and catching a chick by the leg (Outcome). As opposed to factual components, inferred components encode less obvious aspects of the story since they are not directly visible. This component type focuses on the underlying intentions (Goals), emotions and mental states (Internal States) that drive the protagonists’ actions, and requires a deeper understanding and interpretation of the protagonists’ motivations and reactions. For instance, in the episode above, the cat’s wish to pounce (Goal), its happiness to catch its prey, and the chick’s fear (Internal States) need to be inferred. The combination of inferred and factual components elements ensures that an episode is not just a series of random events but a causally connected sequence. Also not random is their combination: in our model, one of the two inferred initial components (Internal State as initiating event or Goal) that trigger an episode must be combined with the ensuing actions (encoded by Attempt and Outcome). When a storyteller is able to combine inferred and factual components into an episodic sequence in this order one can say that s/he can verbalize a complete episode, and by extension – tell a well-formed story. This is the third dimension. In Fig. 1a pictorial ‘translation’ of the linguistic model is exemplified by the stories Baby Birds (top) and Baby Goats (bottom). In sum, the linguistic model operates across multiple dimensions and emphasizes the causal-temporal organization of narrative events by combining factual and inferred components. By dissecting an episode into its five key components, it offers a systematic approach to understanding and analyzing stories. This structured framework reveals the underlying elements that make storytelling an integral part of human culture and cognition. It has been adapted to and used in over 90 languages in different countries all over the world, which attests to the versatility of the model. While stories are told and listened to by children around the globe, the number of children growing up with stories in more than one language is steadily increasing. It is therefore astonishing that not very much is known about how bilingual children develop narrative skills – how bilingual children tell a picture-based narrative to a naïve listener – in their two languages (home language, L1, and second/societal language, L2), but also in comparison with monolinguals. For the acquisition of a range of language domains, such as the lexicon and morphosyntax, monolingual children have been generally claimed to outperform bilingual age peers. While the empirical evidence for the acquisition of vocabulary strongly points in this direction 33 , 34 , 35 , 36 , results concerning monolinguals vs bilinguals are far more mixed for other language domains, and for narrative skills there has been too little empirical data so far for robust generalizations. Also mixed is the evidence for monolingual vs bilingual narrative development. Some researchers have put forward the idea that bilingual children should not lag behind monolinguals in how they structure narratives because this can be considered a more general cognitive language skill that can be transferred between languages 37 , 38 , 39 , 40 , 25 . This cognitive skill has been downplayed by researchers who instead put lexical skills at the forefront of narrative ability. They argue that a threshold in expressive vocabulary must be reached, firstly, for well-formed narratives to be produced, and secondly, for cross-linguistic carry-over of hierarchical structure to occur in a bilingual child 41 , 42 , 43 . For the monolingual and bilingual acquisition of narratives then, there are conflicting theoretical predictions, and the empirical evidence is scarce as yet. This is where the present study comes in. We investigate 2608 narratives told or retold by 1189 children age 3.41 to 13.83 years (mean age = 7.47, SD = 2.50), speaking 33 languages. We report on monolinguals and bilinguals, and for bilinguals, we compare their L1 and L2. In doing this, we apply the above-mentioned multidimensional model and investigate the quantification of narrative components, compare the distribution of factual and inferred types, and the combination of these components into well-formed narratives. Thus, the overarching aim of the study is to advance our understanding of how storytelling (assessed via single components and complete episodes) develops in monolingual and bilingual children from a multidimensional perspective. We derive four research questions: RQ1: What are the differences in children’s overall narrative skills across languages? RQ2: For bilingual children, how does the development of narrative skills in L1 compare to that in L2? RQ3: What is the general pattern in children’s use of factual and inferred components across the investigated languages? RQ4: How do complete episodes evolve across languages and in monolingual vs bilingual children? Results What are the differences in children’s overall narrative skills across languages? First, we focused on the developmental trajectories of mono- and bilinguals in overall narrative skills. We found strong effects for age, language status (mono- vs. bilingual) and story mode (telling vs. retelling the story). Across languages, children improved with age (fixed effect in Bayesian regression model: β = 0.68, 95% CrI = 0.47 to 0.91). An interaction between age and language status (β = -0.49, 95% CrI = -0.78 to -0.20) points to different developmental trajectories for mono- and bilinguals. At the lower end of the age range, monolinguals outperform bilinguals. The slope for bilinguals, however, is much steeper, such that they eventually come to outperform the monolinguals (see Fig. 2 ). Overall, children perform better when retelling the story, compared to when telling it (β = -0.36, 95% CrI = -0.41 to -0.32). Please note that the data coverage is different for mono- and bilinguals. There is no data for older monolinguals, but there is data for older bilinguals. The high level of performance of older bilinguals might be responsible for the interaction effect and if older monolinguals had been tested, the interaction effect might have been reduced or even absent. Thus, we also fitted a model with the interaction excluded, in which case, there was a slight overall advantage for monolinguals (β = 0.14, 95% CrI = 0.04 to 0.23). How does the development of bilinguals’ narrative skills in L1 compare to that in L2? Next, we analyzed only the data from bilingual children and focused on differences in overall narrative skills between L1 and L2. The results again showed a strong effect of age (β = 0.58, 95% CrI = 0.36 to 0.82) and story mode (β = -0.36, 95% CrI = -0.41 to -0.31). Most importantly, children perform better when tested in their L2 compared to L1 (β = 0.22, 95% CrI = 0.09 to 0.34). What is the general pattern in children’s use of factual and inferred components across the investigated languages? The third question asked about the different elements of the narrative. Instead of looking at the overall narrative skills, we analyzed whether there are differences between children speaking different languages in how likely they are to produce structural elements of the narrative – internal states, goals, attempts, and outcomes. Attempts and outcomes are considered to be factual components of the narrative while goals and internal states are inferred components. Across languages, children were more likely to produce factual compared to inferred components: outcomes were the most likely to be produced (β = 0.86, 95% CrI = -0.25 to 1.96), closely followed by attempts (β = 0.81, 95% CrI = -0.29 to 1.90). Internal states were substantially less likely than the previous two components (β = -0.51, 95% CrI = -1.61 to 0.56) and goals were the least likely to be produced (β = -0.88, 95% CrI = -1.99 to 0.20). Even though there was substantial variation across languages in how likely each component was produced, the overall pattern was largely the same (see Fig. 4 ). When analyzing the developmental trajectory for the different components across languages, we see that attempts and outcomes are more likely to be produced already in younger children. Goals and internal states are very unlikely to be produced at the same age. Internal states show a steeper slope compared to goals and are quite likely to be produced by older children (see Fig. 5 ). How do complete episodes evolve across languages and in monolingual vs bilingual children? The final question focused on differences between mono- and bilinguals’ production of complete narratives. That is, we asked if children produced at least one narrative including an attempt, a goal or an internal state, and an outcome. The results showed that the probability to produce a complete narrative increased with age (β = 0.90, 95% CrI = 0.57 to 1.25) and was higher when retelling the story (β = 0.48, 95% CrI = 0.36 to 0.60). Furthermore, monolinguals outperformed bilinguals (β = 0.28, 95% CrI = 0.05 to 0.50, see Fig. 6 ). These results did not change on a stricter coding of a complete narrative, i.e. when leaving out internal states and only coding an episode as complete when the child produced a goal, an attempt, and an outcome (age: β = 0.88, 95% CrI = 0.44 to 1.21; retelling: β = 0.60, 95% CrI = 0.45 to 0.76; monolingual: β = 0.31, 95% CrI = 0.00 to 0.61). Discussion The present study enhances our understanding of children’s storytelling development by applying a multidimensional model of narrative organization to a large pool of monolingual and bilingual data. Specifically, the study identifies similarities and differences in children’s overall narrative skills across languages, compares narrative skill development in L1 and L2 for bilingual children, examines the pattern in children’s use of factual and inferred components across languages, and analyzes the evolution of complete episodes. A dataset of over 2600 oral texts told or retold by 1189 monolingual and bilingual children from age three to thirteen, speaking 33 languages has been used for this purpose. Firstly, as regards RQ1 , we could document a substantial development of overall narrative skills across languages for children, irrespective of language status (monolingual, bilingual). This is not very surprising, and it corroborates the results of previous studies using much smaller samples and a range of different elicitation materials 3 , 16 , 25 , 15 , 18 . More surprising is our finding concerning RQ2 , namely that there are different developmental trajectories for the two groups (monolingual and bilingual) with an interaction between age and language status or ‘lingualism’ of a child. At younger ages, monolingual children exhibit better narrative abilities compared to their bilingual peers (we cannot relate these results to previous studies as there are no studies with such comparisons for linguistically motivated picture sequences). The bilingual children in our sample demonstrate a steeper developmental slope and surpass age-matched monolinguals in narrative performance. However, as noted in the results, it is unclear if the steeper slope for bilinguals would persist if there was data for older monolinguals. Future research should close this gap. As presented here, the findings could be explained by the strong dependence of the early narrative skills on the lexicon. On the one hand, there is evidence that early narrative structure is associated with lexical skills 20 , 42 , 44 , 25 , 45 . On the other hand, a wealth of studies has found that expressive vocabularies of young bilinguals in one language are on average smaller than for age-matched monolinguals 34 , 46 , 47 , 36 . Although the present study does not investigate the direct effect of expressive skills on narration, based on the existing evidence we suggest the following: As the linguistic encoding of the narrative structures relies on words, the smaller expressive vocabularies of young bilinguals may more strongly impede the verbalization of narrative hierarchical structures, leading to a perceived lag of bilinguals in this domain. With increased exposure, bilingual vocabulary extends beyond the threshold and allows for the carry-over of knowledge of narrative hierarchical structures. We also found that children, regardless of their lingualism (language status, i.e. monolingual, bilingual), perform better when retelling than when telling a picture-based narrative. A likely explanation lies in the priming effect of the ‘model’ story that the children listen to prior to retelling, which can bring out a better performance in the child. While this result of better performance for retelling than telling (i.e., story-generation from scratch) might not be mind-boggling, it has been never shown, first, for picture-based methodology grounded in a theoretical linguistic model, second, with structurally exactly parallel elicitation material for telling and retelling, and, third, for such a large number of children across languages and ages. The next result related to RQ2 showed better narrative performance in the L2 as compared to the L1 of bilingual children and a strong effect of age and story mode for both languages. The finding that bilingual children exhibit stronger development of narrative skills in the societal language compared to their home language underscores the influence of environmental exposure and social context on language development. Bilinguals have broader opportunities to use the societal language in various settings, such as schools, friends and others outside home interactions, as well as via media consumption. These frequent and diverse L” exposures lead to increased and diverse input and a richer set of communicative situations where bilinguals actively use their L2, typically at the expense of the L1. This leads to dominance in L2 (for example, for the lexicon 48 , 47 , 49 , 50 . Additionally, formal training in the educational institutions in L2 also supports narrative development in the L2 51,52 . In contrast, the home language (L1) is used in more limited and context-specific situations, providing fewer opportunities for practice and development. Since narrative skills can be transferred between languages, an effect of age is seen in both, L1 and L2 languages 37 , 42 , 38 , 43 , 25 . The general patterns in children’s use of factual and inferred components across the investigated languages are addressed by RQ3 . Children were found to be far more likely to include factual components, such as outcomes and attempts, in their narrative productions compared to inferred components like internal states and goals. Specifically, outcomes were the most frequently produced component, closely followed by attempts, while internal states and goals were substantially less likely to be included in children’s narratives, with goals being the least produced. This consistent pattern across languages suggests that factual components are more accessible or salient to all children, also younger ones, possibly due to their concrete nature and direct (straightforward) observability in stories 53 . In contrast, inferred components, which require understanding and expression of abstract concepts like intentions and emotions, appear to be more challenging for young children. Whilst even very young children are able to identify and infer meaningful relations between events to some degree and create mental network representations of these events, their inferential processes are generally not as developed as those of older children and adults. As their knowledge of social interaction and facts of the world increases, they cannot only connect concrete, physical events that occur closely together (i.e., attempts and outcomes) but also increasingly connect events that are more distant and, in particular, abstract, internal events (i.e. goals, mental and other internal states, such as emotional reactions) 54 , 55 . In our dataset, the probability of mentioning internal states at the youngest age is low, but the developmental slope is steeper compared to goals, making them more common in older children’s narratives. Outcomes may be more important since they trigger the next action, function as the initiating force for the episode to come and are more crucial to move the story on. Our results are in concordance with previous studies which showed that outcomes are both a very commonly produced (e.g., 27 , 56 , 17 , 60 ) and a very commonly recalled component (in story recall, e.g., 14 , 57 , 58 . Earlier studies of the development of narrative structure have also found that attempts and outcomes are some of the earliest and most frequent components produced in children’s storytelling, whilst goals and internal states lag behind. However, unlike the present study, older work involved much smaller samples, did not include both monolinguals and bilinguals and often covered English only. Our findings show this pattern for theoretically-based pictorial stimuli, for a much larger dataset and across 33 languages. Generally, despite variation across languages in the likelihood of producing each narrative component, the overall developmental pattern remains largely consistent. This cross-linguistic consistency underscores potential universal aspects of narrative development in children. The final finding of our study, which concerns RQ4 , sheds light on the differences between monolingual and bilingual children in the production of complete narratives, defined as those including at least one episode consisting of an attempt, and internal state or a goal, and an outcome. Here we also found an effect of age, which is expected. Again, our findings reveal that monolingual children outperform bilingual children in producing complete narratives. This performance gap could be attributed to several factors. Monolingual children might have more extensive vocabulary and stronger narrative skills in their single language due to consistent and concentrated exposure. In contrast, bilingual children split their linguistic input and practice between two languages, which at first may slow their development of proficiency and narrative abilities in each individual language somewhat. An interesting additional outcome was that the likelihood of producing a complete narrative (which increased with age and was higher when children were retelling a story rather than telling one) remained consistent even when we applied stricter criteria for coding a complete narrative, i.e. when an episode was only considered complete if the child included a goal, an attempt, and an outcome, excluding internal states. This result shows that the original methodology, which suggested goal-based episodes as the core of a successful and comprehensible narrative, stands (cf. 59 ). The present study thus lays a novel theoretical and methodological foundation for the field of developmental narrative research and brings it up to a new level: It a) allows for the use of unified theory-driven pictorial stimuli that have been widely tested – also on non-WEIRD languages, b) confirms the goal-based approach to the episode as the fundamental unit of a well-formed narrative and as a basic dimension to evaluate children’s narrative skills, and c) shows how narrative theory can be developed, tested and applied both within and across languages with different populations to enhance our understanding of a main human behavior ability – the skill to tell stories. Methods Ethical compliance The research reported here complies with the ethical regulations of the respective countries and was carried out in accordance with the principles of the revised Helsinki Declaration. Participants 2608 narratives told or retold by 1189 monolingual and bilingual children age 3.41 to 13.83 years (mean age = 7.47, SD = 2.50), speaking 33 languages were examined. All children attended preschools or primary schools where the societal language is the language of schooling – this holds for all datasets. Prior to testing, all parents provided written consent agreeing to participate in the study. In addition, parents completed a questionnaire providing background information about the child, developmental language milestones, social status, medical history, and so forth. On the basis of the parental questionnaire, and conversations with the teachers, children with deficits in hearing, neurological development, cognitive development, with a diagnosis of language disorder or with articulation problems were excluded from the present study. The children came from a variety of social backgrounds. As socioeconomic status is determined in different ways in different countries it was not included in the analyses here. Narrative elicitation, transcription, and coding Telling and retelling were elicited with the MAIN instrument, which allows narrative elicitation in multiple languages from the same child. The structurally parallel, comparable picture-based Baby Goats and Baby Birds stories from MAIN 34 were used. The full set of guidelines for administration and narrative elicitation, story scripts and picture stimuli, scoring protocols, comprehension questions, and background questionnaire is available from http://www.zas.gwz-berlin.de/zaspil.html . Trained native speakers of the language in question administered the task. For bilingual children the task was administered twice, in a single-language setting (monolingual mode) in the two respective languages, with an interval of at least two days in between. At the beginning of the testing, a set of warm-up questions was asked. After that, the experimenter placed three colored envelopes on the table and asked the child to choose one. The child selected an envelope and took out a set of six pictures in a fold-out accordion-style format. Following standard procedure, the child was informed that only s/he was allowed to see the pictures. The experimenter helped unfolding the pictures, but kept them away from the experimenter’s view. The child was asked to carefully look at the pictures of the whole story, but not to show them to the experimenter. When the child was ready, the experimenter helped to refold the pictures so only the first two pictures remained visible to the child. For narrative elicitation in the telling mode, the experimenter then asked the child to tell the story, using standard prompts, and assisted with unfolding the pictures (first two, then four, then all six), as the child proceeded to tell the story. For elicitation in the retelling mode, the child also chose an envelope out of three, took out the pictures and looked at the pictures of the whole story. When the child was ready, the experimenter told the child the scripted story, and asked the child to tell the story to the experimenter. To avoid repetition effects, different stories were used for each elicitation from the child (MAIN contains four parallel picture-based stories). All oral narratives were audio-recorded, and for some of the participating research groups, video-recorded as well. All narratives were transcribed verbatim by trained native speakers of the respective languages. The transcripts were then checked by the authors for the respective languages. The procedure included having independent transcribers transcribe a minimum of 5% of the audio files per language a second time, and calculating a Cohen kappa reliability score with Computerized Language Analysis as an intertranscriber reliability measure. In the present study we only included data where the agreement on the transcribed words was at least 80%. Coding and analyses of story components were performed by trained native speakers of the respective languages according to the MAIN protocol, which includes detailed guidelines on scoring. Analysis Regression models were implemented in a Bayesian framework in R using the brms package 60 . We used default priors implemented in the package for all parameters. To compare models, we used Bayesian approximate leave-one-out cross-validation 61 based on differences in expected log posterior density (ELPD) estimates and the associated standard error (se). Based on suggestions in the literature 62 , we considered a model to provide a substantially better fit when the difference in ELPD in favor of the model exceeded two times the standard error of the difference. To evaluate individual predictors in a model, we investigated its posterior mean and the associated 95% Credible Interval (CrI). We classified a predictor to be substantial when its 95% CrI did not overlap with 0. The data and code to reproduce the analysis can be found in the following repository: https://github.com/manuelbohn/main RQ1 The first model used to analyze overall narrative skills across languages predicted the probability of producing a narrative component by age (scaled), language status (mono- or bilingualism), the story mode (telling vs. retelling) and an interaction between age and language status. The model included random intercepts for child (some individuals were tested twice, in different languages) and test language. Furthermore, we included random slopes for age within test language. Model formula (in brms notation): produce ~ age * language status + mode + (1 | child) + (age | test language). RQ 2 The second model only used the data from the bilinguals and predicted the probability of producing a narrative component by age, test language status (L1 vs. L2) and story mode (telling vs. retelling). Including an additional interaction between age and test language status decreases the model fit and was therefore omitted (ELPD difference in favor of model without interaction = 1.61, se of difference = 0.90). The model included random intercepts for child and test language and a random slope for age within test language. Including a random slope for test language status within test language to model differences in L1 and L2 for each language was not possible because there was only L1 data for some languages. Model formula: produce ~ age + test language status + mode + (1|child) + (age | test language). RQ 3 The third model focused on the structural components of narratives. To do so, we added a random intercept for component along with random slopes for age and test language within component. This enabled us to estimate the probability of producing each component both overall and separately for each test language. Furthermore, we were able to estimate the effect of age for each component. Model formula: produce ~ age * language status + mode + (1|child) + (age | test language) + (age + test language | component). RQ 4 The fourth model focused on the production of complete narratives. For this, we re-coded the data such that we looked at each of the three episodes for each child and asked if they produced all three structural elements (attempt, goal or internal state, and outcome) in any of the episodes. If yes, this was scored as 1, if not it was scored as 0. Thus, children could get a maximum score of 3. The model predicted the probability of producing a complete narrative by age, test language, and mode. Model formula: narrative ~ age + language status + mode + (1 | child) + (age | test language). A model including an interaction between test language and age did not perform substantially better (ELPD difference in favor of model with interaction = 1.90, se of difference = 2.00). When inspecting the estimates of the interaction model, we see a strong interaction between age and language status (β = -0.76, 95% CrI = -1.44 to -0.10) so that with increasing age, bilinguals are predicted to outperform older monolinguals. As noted in the results section, this interaction is most likely the consequence of the absence of data for older monolinguals. The fifth model was the equivalent of the second model for complete narratives. Including only the data from bilinguals, we predicted the probability of producing a complete narrative by age, test language status and mode. Model formula: narrative ~ age + test language status + mode + (1|child) + (age | test language). For the sixth and final model, we changed the coding of complete narratives to only count as a complete narrative when the child produced an attempt a goal, and an outcome (ignoring internal states). The model formula was the same as for the fourth model. Declarations The Ethics Committee of the German Linguistic Society has reviewed your application on the basis of the descriptions and information you have submitted. From the commission's point of view, there are currently no ethical concerns about conducting the study, i.e. you have received a positive vote. References Reese, E., Suggate, S., Long, J. & Schaughency, E. Children’s oral narrative and reading skills in the first 3 years of reading instruction. Reading and Writing , 23, 627–644 (2010). Bliss, L., McCabe, A. & Miranda, E. Narrative assessment profile: Discourse analysis for school-age children. Journal of Communication Disorders , 31, 347–363 (1998). Berman, R.A. & Slobin, D.I. Relating events in narrative: A crosslinguistic developmental study . Hillsdale, NJ: Erlbaum (1994). Nicolopoulou, A. & Richner, E. S. From actors to agents to persons: The development of character representation in young children's narratives. Child development , 78(2), 412–429 (2007). Bohnacker, U. Tell me a story in English or Swedish: Narrative production and comprehension in bilingual preschoolers and first graders. Applied Psycholinguistics , 37(1), 19–48 (2016). Peterson, C. & McCabe, A. Developmental psycholinguistics: Three ways of looking at a child's narrative. New York: Plenum (1983). Trabasso, T. & Nickels, M. The development of goal plans of action in the narration of a picture story. Discourse Processes , 15, 249–275 (1992). Hadley, P. A. Language sampling protocols for eliciting text-level discourse. Language, Speech, and Hearing Services in Schools , 29, 132–147 (1998). Catts, H. W., Fey, M. E., Zhang, X. & Tomblin, J. B. Language basis of reading and reading disabilities: Evidence from a longitudinal investigation. Scientific studies of reading , 3(4), 331–361 (1999). Sénéchal, M. & LeFevre, J. A. Parental involvement in the development of children’s reading skill: A five-year longitudinal study. Child development , 73(2), 445–460 (2002). Tabors, P. O., Snow, C. E. & Dickinson, D. K. Homes and schools together: Supporting language and literacy development (2001). Bishop, D. V. M. & Edmundson, A. Language impaired 4-year -olds: Distinguisting transient from persistent impairment. Journ al of Speech and Hearing Disorders, 52, 156–173 (1987). Gutiérrez-Clellen, V. F. Narratives in two languages: Assessing performance of bilingual children. Linguistics and Education, 13(2), 175–197 (2002). Stein, N. L. & Glenn, C. G. An analysis of story comprehension in elementary school children. In R. Freedle (Ed.), Discourse processing: Multidisciplinary perspectives. Norwood, NJ: Ablex, 53–120 (1979). Schneider, P., Hayward, D. & Dubé, R. V. Storytelling from pictures using the Edmonton Narrative Norms Instrument. Journal of Speech Pathology and Audiology , 30(4), 224–238 (2006). Hayward, D., Schneider, P. & Gillam, R. B. Age and task-related effects on young children’s understanding of a complex picture story. Alberta Journal of Educational Research, 55(1) (2009). Khan, K. S., Gugiu, M. R., Justice L. M., Bowles, R. P., Skibbe, L. E. & Piasta, S. B. Age-related progressions in story structure in young children’s narratives. Journal of Speech, Language, and Hearing Research 59 (December 2016), 1395–1408 (2016). Trabasso, T., Stein, N. L., Rodkin, P. C., Munger, M. P. & Baughn, C. R. Knowledge of goals and plans in the on-line narration of events. Cognitive development, 7(2), 133–170 (1992). Stein, N. L. & Albro, E. R. Building complexity and coherence: Children’s use of goalstructured knowledge in telling good stories. In M. Bamberg (Ed.), Learning how to narrate: New directions in child development (pp. 5–44). Mahwah, NJ: Lawrence Erlbaum Associates (1997). Bitetti, D., Hammer, C. S. & López, L. M. The narrative macrostructure production of Spanish–English bilingual preschoolers: Within-and cross-language relations. Applied Psycholinguistics, 41(1), 79–106 (2020). Labov, W. Some principles of linguistic methodology. Language in society, 1(1), 97–120 (1972). Mandler, J. M. & Johnson, N. S. Remembrance of things parsed: Story structure and recall. Cognitive psychology, 9(1), 111–151 (1977). Stein, N. L. & Policastro, M. The concept of a story: a comparison between children’s and teachers’ viewpoints. In H. Mandl, N. L. Stein & T. Trabasso (Eds.), Learning and comprehension of text. Hillsdale, NJ: Lawrence Erlbaum, 113–155 (1984). Westby, C. E. Assessing and facilitating text comprehension problems. In H. Catts & A. Kamhi (Eds.), Language and reading disabilities (pp. 157–232). Boston, MA: Allyn & Bacon (2005). Pearson, B. Z. Bilingual Infants: What we know, what we need to know. In M. Suarez-Orozco & M. Paez (Eds.), Latinos: Remarking America (pp. 306–320). Berkeley, CA: University of California Press (2002). Montanari, S. The development of narrative competence in the L1 and L2 of Spanish-English bilingual children. International Journal of Bilingualism, 8(4), 449–497 (2004). Fiestas, C. E. & Peña, E. D. Narrative discourse in bilingual children: Language and task effects. Language, Speech, and Hearing Services in Schools, 35(2), 155–168 (2004). Mayer, M. Frog, Where Are You? New York: Dial Books (1969). Renfrew, C. The Bus Story: A test of continuous speech. North Place, Old Headington: Oxford (1969). Gillam, R. B. & Pearson, N. A. Test of Narrative Language (TNL). PRO-ED (2004). Gagarina, N. V., Klop, D., Kunnari, S., Tantele, K., Välimaa, T., Balčiūnienė, I.… Walters, J. MAIN: Multilingual assessment instrument for narratives. ZAS papers in linguistics, 56, 155–155 (2012). Gagarina, N., Klop, D., Kunnari, S., Tantele, K., Välimaa, T., Bohnacker, U. & Walters, J. MAIN: Multilingual assessment instrument for narratives–Revised. ZAS Papers in Linguistics, 63, 20–20 (2019). Bialystok, E., Luk, G., Peets, K. F. & Sujin, Y. A. N. G. Receptive vocabulary differences in monolingual and bilingual children. Bilingualism: Language and cognition, 13(4), 525–531 (2010). Haman, E., Wodniecka, Z., Marecka, M., Szewczyk, J., Białecka-Pikul, M., Otwinowska,A. … Foryś-Nogala, M. How does L1 and L2 exposure impact L1 performance in bilingual children? Evidence from Polish-English migrants to the United Kingdom. Frontiers in psychology, 8, 1444 (2017). Rhys, M. & Thomas, E. M. Bilingual Welsh–English children's acquisition of vocabulary and reading: implications for bilingual education. International Journal of Bilingual Education and Bilingualism, 16(6), 633–656 (2013). Vermeer, A. Breadth and depth of vocabulary in relation to L1/L2 acquisition and frequency of input. Applied psycholinguistics, 22(2), 217–234 (2001). Berman, R. A. “Narrative development in multilingual contexts: A cross-linguistic perspective”. In Ludo Verhoeven & Sven Strömqvist (eds.), Narrative Development in a Multilingual Context (pp 419–428), Amsterdam/Philadelphia: John Benjamins, (2001). Govindarajan, K. & Paradis, J. Narrative abilities of bilingual children with and without developmental language disorder (SLI): Differentiation and the role of age and input factors. Journal of Communication Disorders, 77, 1–16 (2019). Iluz-Cohen, P., & Walters, J. Telling stories in two languages: Narratives of bilingual preschool children with typical and impaired language. Bilingualism: language and cognition, 15(1), 58–74 (2012). Paradis, J., Soto-Corominas, A., Daskalaki, E., Chen, X. & Gottardo, A. Morphosyntactic development in first generation Arabic—English children: The effect of cognitive, age, and input factors over time and across languages. Languages , 6(1), 51 (2021). Bitetti, D. & Hammer, C. S. The home literacy environment and the English narrative development of Spanish–English bilingual children. Journal of Speech, Language, and Hearing Research, 59(5), 1159–1171 (2016). Bohnacker, U., Lindgren, J. & Öztekin, B. Storytelling in bilingual Turkish-Swedish children: Effects of language, age and exposure on narrative macrostructure. Linguistic Approaches to Bilingualism, 12(4), 413–445 (2022). Lindgren, J. & Bohnacker, U. How do age, language, narrative task, language proficiency and exposure affect narrative macrostructure in German-Swedish bilingual children aged 4 to 6?. Linguistic Approaches to Bilingualism, 12(4), 479–508 (2022). Heilmann, J., Miller, J. F., Nockerts, A. & Dunaway, C. Properties of the narrative scoring scheme using narrative retells in young school-age children. American Journal of Speech-Language Pathology, 19, 154–16 (2010). Uccelli, P. & Páez, M. Narrative and vocabulary development of bilingual children from kindergarten to first grade: Developmental changes and associations among English and Spanish skills. Language, Speech, and Hearing Services in Schools, 38, 225–236 (2007). Oller, D. K., Cobo-Lewis, A. B. & Pearson, B. Z. Profiles in early bilingual learning: Vocabulary acquisition and the distributed characteristic. International Clinical Phonetics and Linguistic Association, LA (2004). Oller, D. K., Pearson, B. Z. & Cobo-Lewis, A. B. Profile effects in early bilingual language and literacy. Applied psycholinguistics, 28(2), 191–230 (2007). Gagarina, N. & Klassert, A. Input dominance and development of home language in Russian-German bilinguals. Frontiers in Communication , 3, 40 (2018). Paradis, J. Individual differences in child English second language acquisition: Comparing child-internal and child-external factors. Linguistic approaches to bilingualism, 1(3), 213–237 (2011). Yip, V. & Matthews, S. Relative clauses in Cantonese-English bilingual children: Typological challenges and processing motivations. Studies in Second Language Acquisition, 29(2), 277–300 (2007). Gagarina, N. Narratives of Russian–German preschool and primary school bilinguals: Rasskaz and Erzaehlung. Applied Psycholinguistics, 37(1), 91–122 (2016). Gagarina, N. & Milano, A. Grammatical Development Within a Context of Early Bilingual Education and Language Support Programs. In Handbook of Early Language Education (pp. 1–34). Cham: Springer International Publishing (2021). Cohn, N. & Magliano, J. P. Editors’ introduction and review: Visual narrative research: An emerging field in cognitive science. Topics in Cognitive Science, 12(1), 197–223 (2020). Van den Broek, P., Rapp, D. N. & Kendeou, P. Integrating memory-based and constructionist processes in accounts of reading comprehension. Discourse processes, 39(2–3), 299–316 (2005). Bohnacker, U. & Gagarina, N. Introduction to MAIN–Revised, how to use the instrument and adapt it to further languages. ZAS papers in linguistics, 64, xiii-xxi (2020). Soodla, P. (2011). Picture-elicited narratives of Estonian children at the kindergarten-school transition as a measure of language competence (Doctoral dissertation). Trabasso, T., Secco, T. & van den Broek, P. W. Causal cohesion and story coherence. In Heinz Mandl, Nancy L.Stein & Tom Trabasso (eds.), Learning and comprehension of text (pp. 83–111). Hillsdale, NJ: Lawrence Erlbaum Associates, (1984). Mandler, J. M., Scribner, S., Cole, M. & DeForest, M. Cross-cultural invariance in story recall. Child development, 19–26 (1980). Benson, M. S. Psychological causation and goal-based episodes: Low-income children's emerging narrative skills. Early Childhood Research Quarterly , 12(4), 439–457 (1997). Bürkner, P. C. Advanced Bayesian multilevel modeling with the R package brms. arXiv preprint arXiv:1705.11123 (2017). Vehtari, A., Gelman, A. & Gabry, J. Practical Bayesian model evaluation using leave-one-out cross-validation and WAIC. Statistics and computing, 27, 1413–1432 (2017). Sivula, T., Magnusson, M., Matamoros, A. A. & Vehtari, A. Uncertainty in Bayesian leave-one-out cross-validation based model comparison. arXiv preprint arXiv:2008.10296 (2020). Additional Declarations There is NO Competing Interest. Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-5672219","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":397267425,"identity":"efa997a8-2d7f-44eb-9dba-646b48935f2e","order_by":0,"name":"Natalia Gagarina","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABBklEQVRIiWNgGAWjYNACAwYGfnYGhsMMDBJQETYitEg2k6YFpAuonhnBxaNF3v34xc8VBXfkjQ8zHzxcuMPCnoH/8OMPDGU2OLUYnskpljxj8Mxw22G2hMMzz0gAbTpmJsFwLg23loacBMkGg8OM2w7zGBzmbZNgsz/YYMbA2HYYt5b+N8k/gVrsNzfzfwBp4WFgZv/8gbHtP26/SKQfA9mSuIGZhwGkRYKBjcdAgrHtAE4tBhJv2CyBWpJnHGYzODyzTcKAgYenTCLhXDJuW/rTH99s+HPYtr+9+fHnwrY6YIgd3/zhQ5kdblsO8BhgEU7AqQFoSwP7AzzSo2AUjIJRMAqAAADANlGrQGESwQAAAABJRU5ErkJggg==","orcid":"https://orcid.org/0000-0002-5136-1071","institution":"Leibniz-Zentrum Allgemeine Sprachwissenschaft","correspondingAuthor":true,"prefix":"","firstName":"Natalia","middleName":"","lastName":"Gagarina","suffix":""},{"id":397267426,"identity":"3daf1ee3-4fef-430e-8503-c691961ffb65","order_by":1,"name":"Manuel Bohn","email":"","orcid":"https://orcid.org/0000-0001-6006-1348","institution":"Leuphana University Lü","correspondingAuthor":false,"prefix":"","firstName":"Manuel","middleName":"","lastName":"Bohn","suffix":""},{"id":397267427,"identity":"c1d47987-8fe9-44d0-b15f-0e12c52681d4","order_by":2,"name":"Ute Bohnacker","email":"","orcid":"","institution":"Uppsala University","correspondingAuthor":false,"prefix":"","firstName":"Ute","middleName":"","lastName":"Bohnacker","suffix":""},{"id":397267454,"identity":"ae537b0d-16fd-414d-b8e1-f254f325daad","order_by":3,"name":"Sharon Armon-Lotem","email":"","orcid":"https://orcid.org/0000-0003-3360-9472","institution":"Bar-Ilan University","correspondingAuthor":false,"prefix":"","firstName":"Sharon","middleName":"","lastName":"Armon-Lotem","suffix":""},{"id":415431000,"identity":"6ebaf8f6-34bf-423e-930e-54598781290c","order_by":4,"name":"Dato Abashidze","email":"","orcid":"","institution":"Leibniz-Centre General Linguistics, Germany","correspondingAuthor":false,"prefix":"","firstName":"Dato","middleName":"","lastName":"Abashidze","suffix":""},{"id":397267453,"identity":"2a11590e-5d2f-4acd-928a-8834dd38a625","order_by":5,"name":"Kurbonidin Alamshoev","email":"","orcid":"","institution":"Institute of Humanities, National Academy","correspondingAuthor":false,"prefix":"","firstName":"Kurbonidin","middleName":"","lastName":"Alamshoev","suffix":""},{"id":415431001,"identity":"ad466093-2f6a-42b3-b288-59343baee73e","order_by":6,"name":"Carmit Altman","email":"","orcid":"","institution":"Bar-Ilan University, Israel","correspondingAuthor":false,"prefix":"","firstName":"Carmit","middleName":"","lastName":"Altman","suffix":""},{"id":397267444,"identity":"ce8803ba-ac82-4a75-8464-382afdf254a1","order_by":7,"name":"Amirkul' Amanov","email":"","orcid":"","institution":"Satbayev University","correspondingAuthor":false,"prefix":"","firstName":"Amirkul'","middleName":"","lastName":"Amanov","suffix":""},{"id":397267452,"identity":"dd9aa5c9-b872-4a96-a673-30afb37ab511","order_by":8,"name":"Kathleen Kay Amora","email":"","orcid":"","institution":"Research Centre for Natural Sciences, Budapest","correspondingAuthor":false,"prefix":"","firstName":"Kathleen","middleName":"Kay","lastName":"Amora","suffix":""},{"id":397267436,"identity":"4c0cfea7-f84d-4641-a31a-6f46de160aaa","order_by":9,"name":"Stanislava Antonijevic","email":"","orcid":"","institution":"University of Galway","correspondingAuthor":false,"prefix":"","firstName":"Stanislava","middleName":"","lastName":"Antonijevic","suffix":""},{"id":397267433,"identity":"a11f0a41-d6d7-4eb9-9208-d02aed7efc3d","order_by":10,"name":"Reili Argus","email":"","orcid":"","institution":"Tallinn University","correspondingAuthor":false,"prefix":"","firstName":"Reili","middleName":"","lastName":"Argus","suffix":""},{"id":415431003,"identity":"85eae00e-9225-46ad-821a-dee9539df397","order_by":11,"name":"Marta Białecka","email":"","orcid":"","institution":"Nicolaus Copernicus University, Poland","correspondingAuthor":false,"prefix":"","firstName":"Marta","middleName":"","lastName":"Białecka","suffix":""},{"id":397267429,"identity":"80d71dd3-5d98-4f15-8a2e-bec77e509f3a","order_by":12,"name":"Elma Blom","email":"","orcid":"","institution":"Utrecht University","correspondingAuthor":false,"prefix":"","firstName":"Elma","middleName":"","lastName":"Blom","suffix":""},{"id":397267428,"identity":"a9a897d3-44b4-43b9-9361-9f51ee71eeed","order_by":13,"name":"Tessel Boerma","email":"","orcid":"","institution":"Utrecht University","correspondingAuthor":false,"prefix":"","firstName":"Tessel","middleName":"","lastName":"Boerma","suffix":""},{"id":397267430,"identity":"58452b98-47fd-4fa9-ad2b-e9af8a14d447","order_by":14,"name":"Angel Chan","email":"","orcid":"","institution":"The Hong Kong Polytechnic University","correspondingAuthor":false,"prefix":"","firstName":"Angel","middleName":"","lastName":"Chan","suffix":""},{"id":397267438,"identity":"79b2824b-8444-4bd7-8e40-534fc3af9455","order_by":15,"name":"Uma Chimirala","email":"","orcid":"","institution":"NALSAR University of Law","correspondingAuthor":false,"prefix":"","firstName":"Uma","middleName":"","lastName":"Chimirala","suffix":""},{"id":397267440,"identity":"cf2d50c4-d02f-43cc-bee1-fe2ca323124a","order_by":16,"name":"Vicky Chondrogianni","email":"","orcid":"","institution":"The University of Edinburgh","correspondingAuthor":false,"prefix":"","firstName":"Vicky","middleName":"","lastName":"Chondrogianni","suffix":""},{"id":397267445,"identity":"27870c3a-3a89-4ac4-9b82-975989a82c55","order_by":17,"name":"Ineta Dabašinskienė","email":"","orcid":"","institution":"Vytautas Magnus University","correspondingAuthor":false,"prefix":"","firstName":"Ineta","middleName":"","lastName":"Dabašinskienė","suffix":""},{"id":397267439,"identity":"e45b61e8-4389-4e8c-ad3d-a7e09d25c627","order_by":18,"name":"Sveta Fichman","email":"","orcid":"","institution":"Bar-Ilan University","correspondingAuthor":false,"prefix":"","firstName":"Sveta","middleName":"","lastName":"Fichman","suffix":""},{"id":397267448,"identity":"d3c99218-3907-4dbd-9643-cb86de51b5ed","order_by":19,"name":"Ewa Haman","email":"","orcid":"https://orcid.org/0000-0003-1615-711X","institution":"University of Warsaw","correspondingAuthor":false,"prefix":"","firstName":"Ewa","middleName":"","lastName":"Haman","suffix":""},{"id":397267431,"identity":"739dfa8a-214e-47de-a139-5db7bd17c633","order_by":20,"name":"Gordana Hrzica","email":"","orcid":"","institution":"University of Zagreb","correspondingAuthor":false,"prefix":"","firstName":"Gordana","middleName":"","lastName":"Hrzica","suffix":""},{"id":397267451,"identity":"5e6a74f1-e35a-4774-b233-cf13b8e7a826","order_by":21,"name":"Monika Janikova","email":"","orcid":"","institution":"Comenius University Bratislava","correspondingAuthor":false,"prefix":"","firstName":"Monika","middleName":"","lastName":"Janikova","suffix":""},{"id":397267450,"identity":"836965f4-0543-4bfe-85c8-70e22a77d89b","order_by":22,"name":"Svetlana Kapalková","email":"","orcid":"","institution":"Comenius University Bratislava","correspondingAuthor":false,"prefix":"","firstName":"Svetlana","middleName":"","lastName":"Kapalková","suffix":""},{"id":415431004,"identity":"300acc37-f72e-4ae0-bddb-f00474465158","order_by":23,"name":"Sviatlana Karpava","email":"","orcid":"","institution":"University of Cyprus, Cyprus","correspondingAuthor":false,"prefix":"","firstName":"Sviatlana","middleName":"","lastName":"Karpava","suffix":""},{"id":415431006,"identity":"b2754926-a612-4498-90b5-eed228619e38","order_by":24,"name":"Mariam Komeili","email":"","orcid":"","institution":"University of Toronto, Canada","correspondingAuthor":false,"prefix":"","firstName":"Mariam","middleName":"","lastName":"Komeili","suffix":""},{"id":397267434,"identity":"6b90d209-4439-4c51-bbfa-6257151c28b4","order_by":25,"name":"Sari Kunnari","email":"","orcid":"https://orcid.org/0000-0001-5290-4851","institution":"Oulu University","correspondingAuthor":false,"prefix":"","firstName":"Sari","middleName":"","lastName":"Kunnari","suffix":""},{"id":415431007,"identity":"72e66ac7-f8fd-4772-90ce-6bb0aed0feb3","order_by":26,"name":"Andra Kütt-Leedis","email":"","orcid":"","institution":"Tallinn University, Estonia","correspondingAuthor":false,"prefix":"","firstName":"Andra","middleName":"","lastName":"Kütt-Leedis","suffix":""},{"id":397267443,"identity":"a81aaacc-5a90-42e0-b397-18ebabb08167","order_by":27,"name":"Chiara Levorato","email":"","orcid":"","institution":"University of Padua","correspondingAuthor":false,"prefix":"","firstName":"Chiara","middleName":"","lastName":"Levorato","suffix":""},{"id":397267432,"identity":"859abe1a-41e7-4635-861c-1f33cef2d251","order_by":28,"name":"Josefin Lindgren","email":"","orcid":"","institution":"Uppsala University, Sweden","correspondingAuthor":false,"prefix":"","firstName":"Josefin","middleName":"","lastName":"Lindgren","suffix":""},{"id":415462080,"identity":"4dad82a4-fe70-4f8d-892e-764008784bd6","order_by":29,"name":"Karolina Muszyńska","email":"","orcid":"","institution":"University of Warsaw, Poland","correspondingAuthor":false,"prefix":"","firstName":"Karolina","middleName":"","lastName":"Muszyńska","suffix":""},{"id":397267437,"identity":"0647eea8-23d3-47e2-8133-1f5cbaa106ea","order_by":30,"name":"Gloria Nyarko","email":"","orcid":"","institution":"Truman State University","correspondingAuthor":false,"prefix":"","firstName":"Gloria","middleName":"","lastName":"Nyarko","suffix":""},{"id":397267447,"identity":"0f8daa38-da5e-4e5d-bdc5-9702b1ddf332","order_by":31,"name":"Celestino Oriikiriza","email":"","orcid":"","institution":"Makerere University","correspondingAuthor":false,"prefix":"","firstName":"Celestino","middleName":"","lastName":"Oriikiriza","suffix":""},{"id":397267449,"identity":"2d046437-6f68-4e94-8960-78c2091f3f4b","order_by":32,"name":"Agnieszka Otwinowska-Kasztelanic","email":"","orcid":"https://orcid.org/0000-0002-1127-3982","institution":"University of Warsaw","correspondingAuthor":false,"prefix":"","firstName":"Agnieszka","middleName":"","lastName":"Otwinowska-Kasztelanic","suffix":""},{"id":397267441,"identity":"972deb6f-4589-464d-9e3d-0a04c8fb07c4","order_by":33,"name":"Gayathri Raman","email":"","orcid":"","institution":"The English and Foreign Languages University, Hyderabad","correspondingAuthor":false,"prefix":"","firstName":"Gayathri","middleName":"","lastName":"Raman","suffix":""},{"id":397267442,"identity":"fea6bf3c-691d-457c-9928-c3d93845bc5a","order_by":34,"name":"Maja Roch","email":"","orcid":"","institution":"University of","correspondingAuthor":false,"prefix":"","firstName":"Maja","middleName":"","lastName":"Roch","suffix":""},{"id":415462081,"identity":"8fa664c0-379d-492a-924b-b2e7e5f08d76","order_by":35,"name":"Ali Talebi","email":"","orcid":"","institution":"University of Alberta, Canada","correspondingAuthor":false,"prefix":"","firstName":"Ali","middleName":"","lastName":"Talebi","suffix":""},{"id":415462082,"identity":"041aa257-d56b-4a57-8fe4-8adc3696cff4","order_by":36,"name":"Nathalie Topaj","email":"","orcid":"","institution":"Leibniz-Centre General Linguistics, Germany","correspondingAuthor":false,"prefix":"","firstName":"Nathalie","middleName":"","lastName":"Topaj","suffix":""},{"id":397267446,"identity":"b9eab340-b69b-42ae-844a-682816c2f520","order_by":37,"name":"Sigal Uziel-Karl","email":"","orcid":"","institution":"Achva Academic College","correspondingAuthor":false,"prefix":"","firstName":"Sigal","middleName":"","lastName":"Uziel-Karl","suffix":""},{"id":397267435,"identity":"15887386-231b-408d-b406-8b27152c2980","order_by":38,"name":"Taina Välimaa","email":"","orcid":"","institution":"Oulu University","correspondingAuthor":false,"prefix":"","firstName":"Taina","middleName":"","lastName":"Välimaa","suffix":""},{"id":397267455,"identity":"74abf093-f497-46a2-93f2-6ac2e596e3b8","order_by":39,"name":"Joel Walters","email":"","orcid":"","institution":"Bar-Ilan University","correspondingAuthor":false,"prefix":"","firstName":"Joel","middleName":"","lastName":"Walters","suffix":""}],"badges":[],"createdAt":"2024-12-18 21:40:27","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-5672219/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-5672219/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":75598979,"identity":"a510c0ca-51d8-4d6b-846e-9c09649c5bdb","added_by":"auto","created_at":"2025-02-06 08:36:28","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":927205,"visible":true,"origin":"","legend":"\u003cp\u003eLegend not included with this version.\u003c/p\u003e","description":"","filename":"image1.png","url":"https://assets-eu.researchsquare.com/files/rs-5672219/v1/beefc5ab06fc38647ec4d365.png"},{"id":75598969,"identity":"cb7cfab3-cdfa-4b3b-9350-8839dc333f29","added_by":"auto","created_at":"2025-02-06 08:36:27","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":218328,"visible":true,"origin":"","legend":"\u003cp\u003eModel predictions for developmental trajectories for mono- and bilinguals. The solid lines show trajectories across languages with 95% CrI. The transparent lines show model predictions for the different languages. The length of each line shows the range for which there is data in any of the languages to inform the model estimates. The two facets show trajectories for the two story modes (telling and retelling).\u003c/p\u003e","description":"","filename":"image2.png","url":"https://assets-eu.researchsquare.com/files/rs-5672219/v1/57c1bb90294734f509982553.png"},{"id":75598972,"identity":"816f7273-b614-4311-b3ad-3f6a8cf1e889","added_by":"auto","created_at":"2025-02-06 08:36:27","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":186680,"visible":true,"origin":"","legend":"\u003cp\u003eModel predictions for developmental trajectories for bilingual children either tested in their L1 or L2. The solid lines with ribbons show trajectories across languages with 95% CrI. The transparent lines show model predictions for the different languages. The length of each line shows the range for which there is data in any of the languages to inform the model estimates. The two facets show trajectories for the two story modes.\u003c/p\u003e","description":"","filename":"image3.png","url":"https://assets-eu.researchsquare.com/files/rs-5672219/v1/b54be782e7d248d3e516d711.png"},{"id":75600933,"identity":"61bc326f-d862-44c6-a4c9-01bfe67a437b","added_by":"auto","created_at":"2025-02-06 08:52:27","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":169595,"visible":true,"origin":"","legend":"\u003cp\u003ePredicted probabilities for the production of different structural elements of the narrative in different languages (for bilinguals at an average age). Points show means with 95% CrI for each component and language.\u003c/p\u003e","description":"","filename":"image4.png","url":"https://assets-eu.researchsquare.com/files/rs-5672219/v1/71e789ff6730aca3abfc637b.png"},{"id":75598983,"identity":"e568a213-e6cb-42c6-9b95-10553b1adc6e","added_by":"auto","created_at":"2025-02-06 08:36:28","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":191864,"visible":true,"origin":"","legend":"\u003cp\u003eModel predictions for developmental trajectories for the different items. The solid lines with ribbons show trajectories across languages with 95% CrI. The length of each line shows the range for which there is data in any of the languages to inform the model estimates. The two facets show trajectories for bi- and monolinguals. Trajectories are for the telling mode.\u003c/p\u003e","description":"","filename":"image5.png","url":"https://assets-eu.researchsquare.com/files/rs-5672219/v1/d1dc7e74a006a3f354a1b125.png"},{"id":75600461,"identity":"681e6dca-6b9a-4294-a893-69fb5cad8a61","added_by":"auto","created_at":"2025-02-06 08:44:27","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":129484,"visible":true,"origin":"","legend":"\u003cp\u003eModel predictions for the probability of producing complete narratives for different languages. Colored dots show mean estimates for mono- and bilinguals with 95% CrI. Predictions are for the mean age and the telling mode.\u003c/p\u003e","description":"","filename":"image6.png","url":"https://assets-eu.researchsquare.com/files/rs-5672219/v1/24856998968752d2a9d930f1.png"},{"id":86262904,"identity":"ef6b4d69-dba6-4317-bae1-dc8a3e5fa95c","added_by":"auto","created_at":"2025-07-08 15:06:11","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":2595399,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-5672219/v1/62b6e5ab-4f71-40b9-8939-ddd0eeaa27f8.pdf"}],"financialInterests":"There is \u003cb\u003eNO\u003c/b\u003e Competing Interest.","formattedTitle":"A story of stories: a large-scale cross-linguistic study of young children’s narratives","fulltext":[{"header":"Introduction","content":"\u003cp\u003eThe ability to produce comprehensible oral stories or narratives is an indispensable part of adequate functioning in society. It forms the basis of human communication and has an impact on social well-being and thus, health\u003csup\u003e\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e,\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e\u003c/sup\u003e. Fledgling narrative skills emerge in the early preschool years and have been said to show to strongly develop up to age nine\u003csup\u003e\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e,\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e,\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e,\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e,\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e\u003c/sup\u003e. Narrative skills are crucial for later preschool and primary school development, bridging oral and literate language\u003csup\u003e\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e\u003c/sup\u003e. They are strongly linked to early literacy, reading, and overall school and life success\u003csup\u003e\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e,\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e,\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e,\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e,\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e\u003c/sup\u003e. Given their importance for social communication, it is not surprising that there is a wealth of studies on the acquisition of narratives, but these studies mainly examine narrative development in monolingual children and predominantly for English and, to some extent, Spanish\u003csup\u003e\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e,\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e,\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e,\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e,\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e,\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e,\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e,\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e\u003c/sup\u003e. These studies are built on theories of narrative organization known as story grammars, which are inspired by the structure of Western fairytales and are not always applicable to non-WEIRD (Western Educated Industrialized Rich Democratic) contexts.\u003c/p\u003e \u003cp\u003eOne the most prominent theories continues to be Stein and Glenn\u0026rsquo;s (1979)\u003csup\u003e\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e\u003c/sup\u003e story grammar, according to which a well-formed narrative includes a number of components: a setting (which introduces the characters and describes the non-linguistic context of a story in time and space), an initiating event, internal response, internal plan, attempt, direct consequence and a reaction. Empirical studies of narrative acquisition have however grappled with identifying these proposed components in authentic narrative data, yielding non-comparable results. Other approaches have used different numbers and types of components\u003csup\u003e\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e,\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e,\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e,\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e,\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e,\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e\u003c/sup\u003e. These theories have in common that they cannot always be verified against empirical narrative material.\u003c/p\u003e \u003cp\u003eOther challenges are methodological. Whilst there are many publications on the higher-order organization of narratives, no study has provided results based on transcribed quantified data for more than 240 children\u003csup\u003e\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e,\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e,\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e,\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e,\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e,\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e,\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e\u003c/sup\u003e, in fact, frequently sample size is only 3 to 50 children, across varying age ranges\u003csup\u003e\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e,\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e,\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e\u003c/sup\u003e. Some studies report results for telling, i.e. story generation from the pictorial stimuli, others for the retelling of a story that an adult had first told to a child, and some also compare these two different modes, but not on same elicitation material\u003csup\u003e\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e\u003c/sup\u003e. Moreover, most picture-based acquisition studies have used elicitation materials that are convenience-based and not linguistically motivated, such as random wordless picture books, e.g., \u003cem\u003eFrog where are you?\u003c/em\u003e\u003csup\u003e\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e,\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e\u003c/sup\u003e about a boy and his run-away pet frog. Alternatively, single pictures, cartoons, or other picture sequences are used that are often eurocentric and culture-specific, e.g., the \u003cem\u003eBus Story Test\u003c/em\u003e featuring a \u0026lsquo;bobby\u0026rsquo; policeman, a red double-decker bus and bucolic scenery \u0026ndash; all typical for the UK\u003csup\u003e\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e\u003c/sup\u003e, or the \u003cem\u003eTest of Narrative Language\u003c/em\u003e, depicting a meal at an American fast-food chain or aliens in a spaceship landing in a park\u003csup\u003e\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e\u003c/sup\u003e. Further, such elicitation materials do not include sufficiently parallel pictorial stimuli to assess the two languages of bilingual children in a comparable manner.\u003c/p\u003e \u003cp\u003eIn order to overcome these theoretical and methodological shortcomings, as part of the COST Action IS0804 \u003cem\u003eLanguage Impairment in a Multilingual Society: Linguistic Patterns and the Road to Assessment\u003c/em\u003e, a research group of linguists, developmental psychologists and speech-language pathologists from over twenty countries (from Northern Europe to South Africa) developed a multidimensional theoretical model for use with diverse monolingual and bilingual populations, different languages, and cultures.\u003c/p\u003e \u003cp\u003eThis model has been translated into an open-source narrative elicitation tool \u003cem\u003eMultilingual Assessment Instrument for Narratives\u003c/em\u003e\u003csup\u003e\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e,\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e\u003c/sup\u003e (MAIN) which is now successfully being used by more than 3200 registered users in 97 languages in 67 countries, primarily with monolingual and bilingual children aged three to twelve. MAIN includes picture-based stories, instructions for use, scoring sheets, short questionnaires, examples of well-formed narratives, etc. (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://main.leibniz-zas.de\u003c/span\u003e\u003cspan address=\"https://main.leibniz-zas.de\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eThis linguistic model operates within and across several dimensions of causally-temporally organized minimal constituents of events, methodically translated into episodes. Episodes are centered around protagonists (in this model these are typically animals that are familiar to children across a variety of cultures, see Fig.\u0026nbsp;1). An episode consists of five integral components: an initiating internal state, a goal, an attempt, an outcome, and a resulting internal state. As stories typically consist of several episodes, our model includes three episodes to provide multiple opportunities for a storyteller to verbalize these components. Their quantification is the first dimension of the model.\u003c/p\u003e \u003cp\u003eThe categorization of these components into two types, factual and inferred, is the second dimension of the model. Factual components are directly observable events, such as a protagonist\u0026rsquo;s physical actions (Attempts) and their immediate consequences (Outcomes). For example, in an episode, one might observe a cat climbing a tree (Attempt) and catching a chick by the leg (Outcome). As opposed to factual components, inferred components encode less obvious aspects of the story since they are not directly visible. This component type focuses on the underlying intentions (Goals), emotions and mental states (Internal States) that drive the protagonists\u0026rsquo; actions, and requires a deeper understanding and interpretation of the protagonists\u0026rsquo; motivations and reactions. For instance, in the episode above, the cat\u0026rsquo;s wish to pounce (Goal), its happiness to catch its prey, and the chick\u0026rsquo;s fear (Internal States) need to be inferred.\u003c/p\u003e \u003cp\u003eThe combination of inferred and factual components elements ensures that an episode is not just a series of random events but a causally connected sequence. Also not random is their combination: in our model, one of the two inferred initial components (Internal State as initiating event or Goal) that trigger an episode must be combined with the ensuing actions (encoded by Attempt and Outcome). When a storyteller is able to combine inferred and factual components into an episodic sequence in this order one can say that s/he can verbalize a complete episode, and by extension \u0026ndash; tell a well-formed story. This is the third dimension.\u003c/p\u003e \u003cp\u003eIn Fig.\u0026nbsp;1a pictorial \u0026lsquo;translation\u0026rsquo; of the linguistic model is exemplified by the stories Baby Birds (top) and Baby Goats (bottom).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eIn sum, the linguistic model operates across multiple dimensions and emphasizes the causal-temporal organization of narrative events by combining factual and inferred components. By dissecting an episode into its five key components, it offers a systematic approach to understanding and analyzing stories. This structured framework reveals the underlying elements that make storytelling an integral part of human culture and cognition. It has been adapted to and used in over 90 languages in different countries all over the world, which attests to the versatility of the model.\u003c/p\u003e \u003cp\u003eWhile stories are told and listened to by children around the globe, the number of children growing up with stories in more than one language is steadily increasing. It is therefore astonishing that not very much is known about how bilingual children develop narrative skills \u0026ndash; how bilingual children tell a picture-based narrative to a na\u0026iuml;ve listener \u0026ndash; in their two languages (home language, L1, and second/societal language, L2), but also in comparison with monolinguals. For the acquisition of a range of language domains, such as the lexicon and morphosyntax, monolingual children have been generally claimed to outperform bilingual age peers. While the empirical evidence for the acquisition of vocabulary strongly points in this direction\u003csup\u003e\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e,\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e,\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e,\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e\u003c/sup\u003e, results concerning monolinguals vs bilinguals are far more mixed for other language domains, and for narrative skills there has been too little empirical data so far for robust generalizations.\u003c/p\u003e \u003cp\u003eAlso mixed is the evidence for monolingual vs bilingual \u003cem\u003enarrative\u003c/em\u003e development. Some researchers have put forward the idea that bilingual children should not lag behind monolinguals in how they structure narratives because this can be considered a more general cognitive language skill that can be transferred between languages\u003csup\u003e\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e,\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e,\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e,\u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e,\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e\u003c/sup\u003e. This cognitive skill has been downplayed by researchers who instead put lexical skills at the forefront of narrative ability. They argue that a threshold in expressive vocabulary must be reached, firstly, for well-formed narratives to be produced, and secondly, for cross-linguistic carry-over of hierarchical structure to occur in a bilingual child\u003csup\u003e\u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e,\u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e,\u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e\u003c/sup\u003e. For the monolingual and bilingual acquisition of narratives then, there are conflicting theoretical predictions, and the empirical evidence is scarce as yet.\u003c/p\u003e \u003cp\u003eThis is where the present study comes in. We investigate 2608 narratives told or retold by 1189 children age 3.41 to 13.83 years (mean age\u0026thinsp;=\u0026thinsp;7.47, SD\u0026thinsp;=\u0026thinsp;2.50), speaking 33 languages. We report on monolinguals and bilinguals, and for bilinguals, we compare their L1 and L2. In doing this, we apply the above-mentioned multidimensional model and investigate the quantification of narrative components, compare the distribution of factual and inferred types, and the combination of these components into well-formed narratives.\u003c/p\u003e \u003cp\u003eThus, the overarching aim of the study is to advance our understanding of how storytelling (assessed via single components and complete episodes) develops in monolingual and bilingual children from a multidimensional perspective. We derive four research questions:\u003c/p\u003e \u003cp\u003eRQ1: What are the differences in children\u0026rsquo;s overall narrative skills across languages?\u003c/p\u003e \u003cp\u003eRQ2: For bilingual children, how does the development of narrative skills in L1 compare to that in L2?\u003c/p\u003e \u003cp\u003eRQ3: What is the general pattern in children\u0026rsquo;s use of factual and inferred components across the investigated languages?\u003c/p\u003e \u003cp\u003eRQ4: How do complete episodes evolve across languages and in monolingual vs bilingual children?\u003c/p\u003e"},{"header":"Results","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e\n \u003ch2\u003eWhat are the differences in children\u0026rsquo;s overall narrative skills across languages?\u003c/h2\u003e\n \u003cp\u003eFirst, we focused on the developmental trajectories of mono- and bilinguals in overall narrative skills. We found strong effects for age, language status (mono- vs. bilingual) and story mode (telling vs. retelling the story). Across languages, children improved with age (fixed effect in Bayesian regression model: \u0026beta;\u0026thinsp;=\u0026thinsp;0.68, 95% CrI\u0026thinsp;=\u0026thinsp;0.47 to 0.91). An interaction between age and language status (\u0026beta; = -0.49, 95% CrI = -0.78 to -0.20) points to different developmental trajectories for mono- and bilinguals. At the lower end of the age range, monolinguals outperform bilinguals. The slope for bilinguals, however, is much steeper, such that they eventually come to outperform the monolinguals (see Fig. \u003cspan class=\"InternalRef\"\u003e2\u003c/span\u003e). Overall, children perform better when retelling the story, compared to when telling it (\u0026beta; = -0.36, 95% CrI = -0.41 to -0.32).\u003c/p\u003e\n \u003cp\u003ePlease note that the data coverage is different for mono- and bilinguals. There is no data for older monolinguals, but there is data for older bilinguals. The high level of performance of older bilinguals might be responsible for the interaction effect and if older monolinguals had been tested, the interaction effect might have been reduced or even absent. Thus, we also fitted a model with the interaction excluded, in which case, there was a slight overall advantage for monolinguals (\u0026beta;\u0026thinsp;=\u0026thinsp;0.14, 95% CrI\u0026thinsp;=\u0026thinsp;0.04 to 0.23).\u003c/p\u003e\n\u003c/div\u003e\n\u003ch3\u003eHow does the development of bilinguals\u0026rsquo; narrative skills in L1 compare to that in L2?\u003c/h3\u003e\n\u003cp\u003eNext, we analyzed only the data from bilingual children and focused on differences in overall narrative skills between L1 and L2. The results again showed a strong effect of age (\u0026beta;\u0026thinsp;=\u0026thinsp;0.58, 95% CrI\u0026thinsp;=\u0026thinsp;0.36 to 0.82) and story mode (\u0026beta; = -0.36, 95% CrI = -0.41 to -0.31). Most importantly, children perform better when tested in their L2 compared to L1 (\u0026beta;\u0026thinsp;=\u0026thinsp;0.22, 95% CrI\u0026thinsp;=\u0026thinsp;0.09 to 0.34).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eWhat is the general pattern in children\u0026rsquo;s use of factual and inferred components across the investigated languages?\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe third question asked about the different elements of the narrative. Instead of looking at the overall narrative skills, we analyzed whether there are differences between children speaking different languages in how likely they are to produce structural elements of the narrative \u0026ndash; internal states, goals, attempts, and outcomes. Attempts and outcomes are considered to be factual components of the narrative while goals and internal states are inferred components.\u003c/p\u003e\n\u003cp\u003eAcross languages, children were more likely to produce factual compared to inferred components: outcomes were the most likely to be produced (\u0026beta;\u0026thinsp;=\u0026thinsp;0.86, 95% CrI = -0.25 to 1.96), closely followed by attempts (\u0026beta;\u0026thinsp;=\u0026thinsp;0.81, 95% CrI = -0.29 to 1.90). Internal states were substantially less likely than the previous two components (\u0026beta; = -0.51, 95% CrI = -1.61 to 0.56) and goals were the least likely to be produced (\u0026beta; = -0.88, 95% CrI = -1.99 to 0.20). Even though there was substantial variation across languages in how likely each component was produced, the overall pattern was largely the same (see Fig. \u003cspan class=\"InternalRef\"\u003e4\u003c/span\u003e).\u003c/p\u003e\n\u003cp\u003eWhen analyzing the developmental trajectory for the different components across languages, we see that attempts and outcomes are more likely to be produced already in younger children. Goals and internal states are very unlikely to be produced at the same age. Internal states show a steeper slope compared to goals and are quite likely to be produced by older children (see Fig. \u003cspan class=\"InternalRef\"\u003e5\u003c/span\u003e).\u003c/p\u003e\n\u003ch3\u003eHow do complete episodes evolve across languages and in monolingual vs bilingual children?\u003c/h3\u003e\n\u003cp\u003eThe final question focused on differences between mono- and bilinguals\u0026rsquo; production of complete narratives. That is, we asked if children produced at least one narrative including an attempt, a goal or an internal state, and an outcome. The results showed that the probability to produce a complete narrative increased with age (\u0026beta;\u0026thinsp;=\u0026thinsp;0.90, 95% CrI\u0026thinsp;=\u0026thinsp;0.57 to 1.25) and was higher when retelling the story (\u0026beta;\u0026thinsp;=\u0026thinsp;0.48, 95% CrI\u0026thinsp;=\u0026thinsp;0.36 to 0.60). Furthermore, monolinguals outperformed bilinguals (\u0026beta;\u0026thinsp;=\u0026thinsp;0.28, 95% CrI\u0026thinsp;=\u0026thinsp;0.05 to 0.50, see Fig. \u003cspan class=\"InternalRef\"\u003e6\u003c/span\u003e). These results did not change on a stricter coding of a complete narrative, i.e. when leaving out internal states and only coding an episode as complete when the child produced a goal, an attempt, and an outcome (age: \u0026beta;\u0026thinsp;=\u0026thinsp;0.88, 95% CrI\u0026thinsp;=\u0026thinsp;0.44 to 1.21; retelling: \u0026beta;\u0026thinsp;=\u0026thinsp;0.60, 95% CrI\u0026thinsp;=\u0026thinsp;0.45 to 0.76; monolingual: \u0026beta;\u0026thinsp;=\u0026thinsp;0.31, 95% CrI\u0026thinsp;=\u0026thinsp;0.00 to 0.61).\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eThe present study enhances our understanding of children’s storytelling development by applying a multidimensional model of narrative organization to a large pool of monolingual and bilingual data. Specifically, the study identifies similarities and differences in children’s overall narrative skills across languages, compares narrative skill development in L1 and L2 for bilingual children, examines the pattern in children’s use of factual and inferred components across languages, and analyzes the evolution of complete episodes. A dataset of over 2600 oral texts told or retold by 1189 monolingual and bilingual children from age three to thirteen, speaking 33 languages has been used for this purpose.\u003c/p\u003e \u003cp\u003eFirstly, as regards \u003cb\u003eRQ1\u003c/b\u003e, we could document a substantial development of overall narrative skills across languages for children, irrespective of language status (monolingual, bilingual). This is not very surprising, and it corroborates the results of previous studies using much smaller samples and a range of different elicitation materials\u003csup\u003e\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e,\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e,\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e,\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e,\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eMore surprising is our finding concerning \u003cb\u003eRQ2\u003c/b\u003e, namely that there are different developmental trajectories for the two groups (monolingual and bilingual) with an interaction between age and language status or ‘lingualism’ of a child. At younger ages, monolingual children exhibit better narrative abilities compared to their bilingual peers (we cannot relate these results to previous studies as there are no studies with such comparisons for linguistically motivated picture sequences). The bilingual children in our sample demonstrate a steeper developmental slope and surpass age-matched monolinguals in narrative performance. However, as noted in the results, it is unclear if the steeper slope for bilinguals would persist if there was data for older monolinguals. Future research should close this gap. As presented here, the findings could be explained by the strong dependence of the early narrative skills on the lexicon. On the one hand, there is evidence that early narrative structure is associated with lexical skills\u003csup\u003e\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e,\u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e,\u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e44\u003c/span\u003e,\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e,\u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e45\u003c/span\u003e\u003c/sup\u003e. On the other hand, a wealth of studies has found that expressive vocabularies of young bilinguals in one language are on average smaller than for age-matched monolinguals\u003csup\u003e\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e,\u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e46\u003c/span\u003e,\u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e47\u003c/span\u003e,\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e\u003c/sup\u003e. Although the present study does not investigate the direct effect of expressive skills on narration, based on the existing evidence we suggest the following: As the linguistic encoding of the narrative structures relies on words, the smaller expressive vocabularies of young bilinguals may more strongly impede the verbalization of narrative hierarchical structures, leading to a perceived lag of bilinguals in this domain. With increased exposure, bilingual vocabulary extends beyond the threshold and allows for the carry-over of knowledge of narrative hierarchical structures.\u003c/p\u003e \u003cp\u003eWe also found that children, regardless of their lingualism (language status, i.e. monolingual, bilingual), perform better when retelling than when telling a picture-based narrative. A likely explanation lies in the priming effect of the ‘model’ story that the children listen to prior to retelling, which can bring out a better performance in the child. While this result of better performance for retelling than telling (i.e., story-generation from scratch) might not be mind-boggling, it has been never shown, first, for picture-based methodology grounded in a theoretical linguistic model, second, with structurally exactly parallel elicitation material for telling and retelling, and, third, for such a large number of children across languages and ages.\u003c/p\u003e \u003cp\u003eThe next result related to RQ2 showed better narrative performance in the L2 as compared to the L1 of bilingual children and a strong effect of age and story mode for both languages. The finding that bilingual children exhibit stronger development of narrative skills in the societal language compared to their home language underscores the influence of environmental exposure and social context on language development. Bilinguals have broader opportunities to use the societal language in various settings, such as schools, friends and others outside home interactions, as well as via media consumption. These frequent and diverse L” exposures lead to increased and diverse input and a richer set of communicative situations where bilinguals actively use their L2, typically at the expense of the L1. This leads to dominance in L2 (for example, for the lexicon\u003csup\u003e\u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e48\u003c/span\u003e,\u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e47\u003c/span\u003e,\u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e49\u003c/span\u003e,\u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e50\u003c/span\u003e\u003c/sup\u003e. Additionally, formal training in the educational institutions in L2 also supports narrative development in the L2\u003csup\u003e51,52\u003c/sup\u003e. In contrast, the home language (L1) is used in more limited and context-specific situations, providing fewer opportunities for practice and development. Since narrative skills can be transferred between languages, an effect of age is seen in both, L1 and L2 languages\u003csup\u003e\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e,\u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e,\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e,\u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e,\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eThe general patterns in children’s use of factual and inferred components across the investigated languages are addressed by \u003cb\u003eRQ3\u003c/b\u003e. Children were found to be far more likely to include factual components, such as outcomes and attempts, in their narrative productions compared to inferred components like internal states and goals. Specifically, outcomes were the most frequently produced component, closely followed by attempts, while internal states and goals were substantially less likely to be included in children’s narratives, with goals being the least produced. This consistent pattern across languages suggests that factual components are more accessible or salient to all children, also younger ones, possibly due to their concrete nature and direct (straightforward) observability in stories\u003csup\u003e\u003cspan citationid=\"CR53\" class=\"CitationRef\"\u003e53\u003c/span\u003e\u003c/sup\u003e. In contrast, inferred components, which require understanding and expression of abstract concepts like intentions and emotions, appear to be more challenging for young children. Whilst even very young children are able to identify and infer meaningful relations between events to some degree and create mental network representations of these events, their inferential processes are generally not as developed as those of older children and adults. As their knowledge of social interaction and facts of the world increases, they cannot only connect concrete, physical events that occur closely together (i.e., attempts and outcomes) but also increasingly connect events that are more distant and, in particular, abstract, internal events (i.e. goals, mental and other internal states, such as emotional reactions)\u003csup\u003e\u003cspan citationid=\"CR54\" class=\"CitationRef\"\u003e54\u003c/span\u003e,\u003cspan citationid=\"CR55\" class=\"CitationRef\"\u003e55\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eIn our dataset, the probability of mentioning internal states at the youngest age is low, but the developmental slope is steeper compared to goals, making them more common in older children’s narratives. Outcomes may be more important since they trigger the next action, function as the initiating force for the episode to come and are more crucial to move the story on. Our results are in concordance with previous studies which showed that outcomes are both a very commonly produced (e.g.,\u003csup\u003e\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e,\u003cspan citationid=\"CR56\" class=\"CitationRef\"\u003e56\u003c/span\u003e,\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e,\u003cspan citationid=\"CR60\" class=\"CitationRef\"\u003e60\u003c/span\u003e\u003c/sup\u003e) and a very commonly recalled component (in story recall, e.g.,\u003csup\u003e\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e,\u003cspan citationid=\"CR57\" class=\"CitationRef\"\u003e57\u003c/span\u003e,\u003cspan citationid=\"CR58\" class=\"CitationRef\"\u003e58\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eEarlier studies of the development of narrative structure have also found that attempts and outcomes are some of the earliest and most frequent components produced in children’s storytelling, whilst goals and internal states lag behind. However, unlike the present study, older work involved much smaller samples, did not include both monolinguals and bilinguals and often covered English only. Our findings show this pattern for theoretically-based pictorial stimuli, for a much larger dataset and across 33 languages. Generally, despite variation across languages in the likelihood of producing each narrative component, the overall developmental pattern remains largely consistent. This cross-linguistic consistency underscores potential universal aspects of narrative development in children.\u003c/p\u003e \u003cp\u003eThe final finding of our study, which concerns \u003cb\u003eRQ4\u003c/b\u003e, sheds light on the differences between monolingual and bilingual children in the production of complete narratives, defined as those including at least one episode consisting of an attempt, and internal state or a goal, and an outcome. Here we also found an effect of age, which is expected. Again, our findings reveal that monolingual children outperform bilingual children in producing complete narratives. This performance gap could be attributed to several factors. Monolingual children might have more extensive vocabulary and stronger narrative skills in their single language due to consistent and concentrated exposure. In contrast, bilingual children split their linguistic input and practice between two languages, which at first may slow their development of proficiency and narrative abilities in each individual language somewhat.\u003c/p\u003e \u003cp\u003eAn interesting additional outcome was that the likelihood of producing a complete narrative (which increased with age and was higher when children were retelling a story rather than telling one) remained consistent even when we applied stricter criteria for coding a complete narrative, i.e. when an episode was only considered complete if the child included a goal, an attempt, and an outcome, excluding internal states. This result shows that the original methodology, which suggested goal-based episodes as the core of a successful and comprehensible narrative, stands (cf.\u003csup\u003e\u003cspan citationid=\"CR59\" class=\"CitationRef\"\u003e59\u003c/span\u003e\u003c/sup\u003e).\u003c/p\u003e \u003cp\u003eThe present study thus lays a novel theoretical and methodological foundation for the field of developmental narrative research and brings it up to a new level: It a) allows for the use of unified theory-driven pictorial stimuli that have been widely tested – also on non-WEIRD languages, b) confirms the goal-based approach to the episode as the fundamental unit of a well-formed narrative and as a basic dimension to evaluate children’s narrative skills, and c) shows how narrative theory can be developed, tested and applied both within and across languages with different populations to enhance our understanding of a main human behavior ability – the skill to tell stories.\u003c/p\u003e\n\n\n\n \n\n "},{"header":"Methods","content":"\u003ch2\u003eEthical compliance\u003c/h2\u003e\u003cp\u003e The research reported here complies with the ethical regulations of the respective countries and was carried out in accordance with the principles of the revised Helsinki Declaration.\u003c/p\u003e\u003ch3\u003eParticipants\u003c/h3\u003e\u003cp\u003e2608 narratives told or retold by 1189 monolingual and bilingual children age 3.41 to 13.83 years (mean age = 7.47, SD = 2.50), speaking 33 languages were examined. All children attended preschools or primary schools where the societal language is the language of schooling – this holds for all datasets.\u003c/p\u003e\u003cp\u003e Prior to testing, all parents provided written consent agreeing to participate in the study. In addition, parents completed a questionnaire providing background information about the child, developmental language milestones, social status, medical history, and so forth.\u003c/p\u003e\u003cp\u003eOn the basis of the parental questionnaire, and conversations with the teachers, children with deficits in hearing, neurological development, cognitive development, with a diagnosis of language disorder or with articulation problems were excluded from the present study.\u003c/p\u003e\u003cp\u003eThe children came from a variety of social backgrounds. As socioeconomic status is determined in different ways in different countries it was not included in the analyses here.\u003c/p\u003e\u003ch3\u003eNarrative elicitation, transcription, and coding\u003c/h3\u003e\u003cp\u003eTelling and retelling were elicited with the MAIN instrument, which allows narrative elicitation in multiple languages from the same child. The structurally parallel, comparable picture-based Baby Goats and Baby Birds stories from MAIN\u003csup\u003e\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e\u003c/sup\u003e were used. The full set of guidelines for administration and narrative elicitation, story scripts and picture stimuli, scoring protocols, comprehension questions, and background questionnaire is available from \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://www.zas.gwz-berlin.de/zaspil.html\u003c/span\u003e\u003cspan address=\"http://www.zas.gwz-berlin.de/zaspil.html\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. Trained native speakers of the language in question administered the task. For bilingual children the task was administered twice, in a single-language setting (monolingual mode) in the two respective languages, with an interval of at least two days in between. At the beginning of the testing, a set of warm-up questions was asked. After that, the experimenter placed three colored envelopes on the table and asked the child to choose one. The child selected an envelope and took out a set of six pictures in a fold-out accordion-style format. Following standard procedure, the child was informed that only s/he was allowed to see the pictures. The experimenter helped unfolding the pictures, but kept them away from the experimenter’s view. The child was asked to carefully look at the pictures of the whole story, but not to show them to the experimenter. When the child was ready, the experimenter helped to refold the pictures so only the first two pictures remained visible to the child. For narrative elicitation in the telling mode, the experimenter then asked the child to tell the story, using standard prompts, and assisted with unfolding the pictures (first two, then four, then all six), as the child proceeded to tell the story. For elicitation in the retelling mode, the child also chose an envelope out of three, took out the pictures and looked at the pictures of the whole story. When the child was ready, the experimenter told the child the scripted story, and asked the child to tell the story to the experimenter. To avoid repetition effects, different stories were used for each elicitation from the child (MAIN contains four parallel picture-based stories). All oral narratives were audio-recorded, and for some of the participating research groups, video-recorded as well. All narratives were transcribed verbatim by trained native speakers of the respective languages. The transcripts were then checked by the authors for the respective languages. The procedure included having independent transcribers transcribe a minimum of 5% of the audio files per language a second time, and calculating a Cohen kappa reliability score with Computerized Language Analysis as an intertranscriber reliability measure. In the present study we only included data where the agreement on the transcribed words was at least 80%. Coding and analyses of story components were performed by trained native speakers of the respective languages according to the MAIN protocol, which includes detailed guidelines on scoring.\u003c/p\u003e\u003ch2\u003eAnalysis\u003c/h2\u003e\u003cp\u003eRegression models were implemented in a Bayesian framework in R using the brms package\u003csup\u003e\u003cspan citationid=\"CR60\" class=\"CitationRef\"\u003e60\u003c/span\u003e\u003c/sup\u003e. We used default priors implemented in the package for all parameters. To compare models, we used Bayesian approximate leave-one-out cross-validation\u003csup\u003e\u003cspan citationid=\"CR61\" class=\"CitationRef\"\u003e61\u003c/span\u003e\u003c/sup\u003e based on differences in expected log posterior density (ELPD) estimates and the associated standard error (se). Based on suggestions in the literature\u003csup\u003e\u003cspan citationid=\"CR62\" class=\"CitationRef\"\u003e62\u003c/span\u003e\u003c/sup\u003e, we considered a model to provide a substantially better fit when the difference in ELPD in favor of the model exceeded two times the standard error of the difference. To evaluate individual predictors in a model, we investigated its posterior mean and the associated 95% Credible Interval (CrI). We classified a predictor to be substantial when its 95% CrI did not overlap with 0. The data and code to reproduce the analysis can be found in the following repository: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://github.com/manuelbohn/main\u003c/span\u003e\u003cspan address=\"https://github.com/manuelbohn/main\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/p\u003e\u003ch2\u003eRQ1\u003c/h2\u003e\u003cp\u003eThe first model used to analyze overall narrative skills across languages predicted the probability of producing a narrative component by age (scaled), language status (mono- or bilingualism), the story mode (telling vs. retelling) and an interaction between age and language status. The model included random intercepts for child (some individuals were tested twice, in different languages) and test language. Furthermore, we included random slopes for age within test language. Model formula (in brms notation): produce ~ age * language status + mode + (1 | child) + (age | test language).\u003c/p\u003e\u003ch2\u003eRQ 2\u003c/h2\u003e\u003cp\u003eThe second model only used the data from the bilinguals and predicted the probability of producing a narrative component by age, test language status (L1 vs. L2) and story mode (telling vs. retelling). Including an additional interaction between age and test language status decreases the model fit and was therefore omitted (ELPD difference in favor of model without interaction = 1.61, se of difference = 0.90). The model included random intercepts for child and test language and a random slope for age within test language. Including a random slope for test language status within test language to model differences in L1 and L2 for each language was not possible because there was only L1 data for some languages. Model formula: produce ~ age + test language status + mode + (1|child) + (age | test language).\u003c/p\u003e\u003ch2\u003eRQ 3\u003c/h2\u003e\u003cp\u003eThe third model focused on the structural components of narratives. To do so, we added a random intercept for component along with random slopes for age and test language within component. This enabled us to estimate the probability of producing each component both overall and separately for each test language. Furthermore, we were able to estimate the effect of age for each component. Model formula: produce ~ age * language status + mode + (1|child) + (age | test language) + (age + test language | component).\u003c/p\u003e\u003ch2\u003eRQ 4\u003c/h2\u003e\u003cp\u003eThe fourth model focused on the production of complete narratives. For this, we re-coded the data such that we looked at each of the three episodes for each child and asked if they produced all three structural elements (attempt, goal or internal state, and outcome) in any of the episodes. If yes, this was scored as 1, if not it was scored as 0. Thus, children could get a maximum score of 3. The model predicted the probability of producing a complete narrative by age, test language, and mode. Model formula: narrative ~ age + language status + mode + (1 | child) + (age | test language). A model including an interaction between test language and age did not perform substantially better (ELPD difference in favor of model with interaction = 1.90, se of difference = 2.00). When inspecting the estimates of the interaction model, we see a strong interaction between age and language status (β = -0.76, 95% CrI = -1.44 to -0.10) so that with increasing age, bilinguals are predicted to outperform older monolinguals. As noted in the results section, this interaction is most likely the consequence of the absence of data for older monolinguals.\u003c/p\u003e\u003cp\u003eThe fifth model was the equivalent of the second model for complete narratives. Including only the data from bilinguals, we predicted the probability of producing a complete narrative by age, test language status and mode. Model formula: narrative ~ age + test language status + mode + (1|child) + (age | test language).\u003c/p\u003e\u003cp\u003eFor the sixth and final model, we changed the coding of complete narratives to only count as a complete narrative when the child produced an attempt a goal, and an outcome (ignoring internal states). The model formula was the same as for the fourth model.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003eThe Ethics Committee of the German Linguistic Society has reviewed your application on the basis of the descriptions and information you have submitted. From the commission's point of view, there are currently no ethical concerns about conducting the study, i.e. you have received a positive vote.\u003c/p\u003e\n\u003cp\u003e\u0026nbsp;\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eReese, E., Suggate, S., Long, J. \u0026amp; Schaughency, E. Children\u0026rsquo;s oral narrative and reading skills in the first 3 years of reading instruction. \u003cem\u003eReading and Writing\u003c/em\u003e, 23, 627\u0026ndash;644 (2010).\u003c/li\u003e\n\u003cli\u003eBliss, L., McCabe, A. \u0026amp; Miranda, E. Narrative assessment profile: Discourse analysis for school-age children. \u003cem\u003eJournal of Communication Disorders\u003c/em\u003e, 31, 347\u0026ndash;363 (1998).\u003c/li\u003e\n\u003cli\u003eBerman, R.A. \u0026amp; Slobin, D.I. \u003cem\u003eRelating events in narrative: A crosslinguistic developmental study\u003c/em\u003e. Hillsdale, NJ: Erlbaum (1994).\u003c/li\u003e\n\u003cli\u003eNicolopoulou, A. \u0026amp; Richner, E. S. From actors to agents to persons: The development of character representation in young children's narratives. \u003cem\u003eChild development\u003c/em\u003e, 78(2), 412\u0026ndash;429 (2007).\u003c/li\u003e\n\u003cli\u003eBohnacker, U. Tell me a story in English or Swedish: Narrative production and comprehension in bilingual preschoolers and first graders. \u003cem\u003eApplied Psycholinguistics\u003c/em\u003e, 37(1), 19\u0026ndash;48 (2016).\u003c/li\u003e\n\u003cli\u003ePeterson, C. \u0026amp; McCabe, A. Developmental psycholinguistics: Three ways of looking at a child's narrative. New York: Plenum (1983).\u003c/li\u003e\n\u003cli\u003eTrabasso, T. \u0026amp; Nickels, M. The development of goal plans of action in the narration of a picture story. \u003cem\u003eDiscourse Processes\u003c/em\u003e, 15, 249\u0026ndash;275 (1992).\u003c/li\u003e\n\u003cli\u003eHadley, P. A. Language sampling protocols for eliciting text-level discourse. \u003cem\u003eLanguage, Speech, and Hearing Services in Schools\u003c/em\u003e, 29, 132\u0026ndash;147 (1998).\u003c/li\u003e\n\u003cli\u003eCatts, H. W., Fey, M. E., Zhang, X. \u0026amp; Tomblin, J. B. Language basis of reading and reading disabilities: Evidence from a longitudinal investigation. \u003cem\u003eScientific studies of reading\u003c/em\u003e, \u0026lt;background-color:#FFCC66;ivertical-align:baseline;\u0026gt;3\u0026lt;/background-color:#FFCC66;ivertical-align:baseline;\u0026gt;(4), 331\u0026ndash;361 (1999).\u003c/li\u003e\n\u003cli\u003eS\u0026eacute;n\u0026eacute;chal, M. \u0026amp; LeFevre, J. A. Parental involvement in the development of children\u0026rsquo;s reading skill: A five-year longitudinal study. \u003cem\u003eChild development\u003c/em\u003e, 73(2), 445\u0026ndash;460 (2002).\u003c/li\u003e\n\u003cli\u003eTabors, P. O., Snow, C. E. \u0026amp; Dickinson, D. K. \u003cem\u003eHomes and schools together: Supporting language and literacy development\u003c/em\u003e (2001).\u003c/li\u003e\n\u003cli\u003eBishop, D. V. M. \u0026amp; Edmundson, A. Language impaired 4-year -olds: Distinguisting transient from persistent impairment. Journ al of Speech and Hearing Disorders, 52, 156\u0026ndash;173 (1987).\u003c/li\u003e\n\u003cli\u003eGuti\u0026eacute;rrez-Clellen, V. F. Narratives in two languages: Assessing performance of bilingual children. Linguistics and Education, 13(2), 175\u0026ndash;197 (2002).\u003c/li\u003e\n\u003cli\u003eStein, N. L. \u0026amp; Glenn, C. G. An analysis of story comprehension in elementary school children. In R. Freedle (Ed.), Discourse processing: Multidisciplinary perspectives. Norwood, NJ: Ablex, 53\u0026ndash;120 (1979).\u003c/li\u003e\n\u003cli\u003eSchneider, P., Hayward, D. \u0026amp; Dub\u0026eacute;, R. V. Storytelling from pictures using the Edmonton Narrative Norms Instrument. \u003cem\u003eJournal of Speech Pathology and Audiology\u003c/em\u003e, 30(4), 224\u0026ndash;238 (2006).\u003c/li\u003e\n\u003cli\u003eHayward, D., Schneider, P. \u0026amp; Gillam, R. B. Age and task-related effects on young children\u0026rsquo;s understanding of a complex picture story. Alberta Journal of Educational Research, 55(1) (2009).\u003c/li\u003e\n\u003cli\u003eKhan, K. S., Gugiu, M. R., Justice L. M., Bowles, R. P., Skibbe, L. E. \u0026amp; Piasta, S. B. Age-related progressions in story structure in young children\u0026rsquo;s narratives. \u003cem\u003eJournal of Speech, Language, and Hearing Research\u003c/em\u003e 59 (December 2016), 1395\u0026ndash;1408 (2016).\u003c/li\u003e\n\u003cli\u003eTrabasso, T., Stein, N. L., Rodkin, P. C., Munger, M. P. \u0026amp; Baughn, C. R. Knowledge of goals and plans in the on-line narration of events. Cognitive development, 7(2), 133\u0026ndash;170 (1992).\u003c/li\u003e\n\u003cli\u003eStein, N. L. \u0026amp; Albro, E. R. Building complexity and coherence: Children\u0026rsquo;s use of goalstructured knowledge in telling good stories. In M. Bamberg (Ed.), Learning how to narrate: New directions in child development (pp. 5\u0026ndash;44). Mahwah, NJ: Lawrence Erlbaum Associates (1997).\u003c/li\u003e\n\u003cli\u003eBitetti, D., Hammer, C. S. \u0026amp; L\u0026oacute;pez, L. M. The narrative macrostructure production of Spanish\u0026ndash;English bilingual preschoolers: Within-and cross-language relations. Applied Psycholinguistics, 41(1), 79\u0026ndash;106 (2020).\u003c/li\u003e\n\u003cli\u003eLabov, W. Some principles of linguistic methodology. Language in society, 1(1), 97\u0026ndash;120 (1972).\u003c/li\u003e\n\u003cli\u003eMandler, J. M. \u0026amp; Johnson, N. S. Remembrance of things parsed: Story structure and recall. Cognitive psychology, 9(1), 111\u0026ndash;151 (1977).\u003c/li\u003e\n\u003cli\u003eStein, N. L. \u0026amp; Policastro, M. The concept of a story: a comparison between children\u0026rsquo;s and teachers\u0026rsquo; viewpoints. In H. Mandl, N. L. Stein \u0026amp; T. Trabasso (Eds.), Learning and comprehension of text. Hillsdale, NJ: Lawrence Erlbaum, 113\u0026ndash;155 (1984).\u003c/li\u003e\n\u003cli\u003eWestby, C. E. Assessing and facilitating text comprehension problems. In H. Catts \u0026amp; A. Kamhi (Eds.), Language and reading disabilities (pp. 157\u0026ndash;232). Boston, MA: Allyn \u0026amp; Bacon (2005).\u003c/li\u003e\n\u003cli\u003ePearson, B. Z. Bilingual Infants: What we know, what we need to know. In M. Suarez-Orozco \u0026amp; M. Paez (Eds.), Latinos: Remarking America (pp. 306\u0026ndash;320). Berkeley, CA: University of California Press (2002).\u003c/li\u003e\n\u003cli\u003eMontanari, S. The development of narrative competence in the L1 and L2 of Spanish-English bilingual children. International Journal of Bilingualism, 8(4), 449\u0026ndash;497 (2004).\u003c/li\u003e\n\u003cli\u003eFiestas, C. E. \u0026amp; Pe\u0026ntilde;a, E. D. Narrative discourse in bilingual children: Language and task effects. Language, Speech, and Hearing Services in Schools, 35(2), 155\u0026ndash;168 (2004).\u003c/li\u003e\n\u003cli\u003eMayer, M. Frog, Where Are You? New York: Dial Books (1969).\u003c/li\u003e\n\u003cli\u003eRenfrew, C. \u003cem\u003eThe Bus Story: A test of continuous speech.\u003c/em\u003e North Place, Old Headington: Oxford (1969).\u003c/li\u003e\n\u003cli\u003eGillam, R. B. \u0026amp; Pearson, N. A. \u003cem\u003eTest of Narrative Language (TNL).\u003c/em\u003e PRO-ED (2004).\u003c/li\u003e\n\u003cli\u003eGagarina, N. V., Klop, D., Kunnari, S., Tantele, K., V\u0026auml;limaa, T., Balčiūnienė, I.\u0026hellip; Walters, J. MAIN: Multilingual assessment instrument for narratives. ZAS papers in linguistics, 56, 155\u0026ndash;155 (2012).\u003c/li\u003e\n\u003cli\u003eGagarina, N., Klop, D., Kunnari, S., Tantele, K., V\u0026auml;limaa, T., Bohnacker, U. \u0026amp; Walters, J. MAIN: Multilingual assessment instrument for narratives\u0026ndash;Revised. ZAS Papers in Linguistics, 63, 20\u0026ndash;20 (2019).\u003c/li\u003e\n\u003cli\u003eBialystok, E., Luk, G., Peets, K. F. \u0026amp; Sujin, Y. A. N. G. Receptive vocabulary differences in monolingual and bilingual children. Bilingualism: Language and cognition, 13(4), 525\u0026ndash;531 (2010).\u003c/li\u003e\n\u003cli\u003eHaman, E., Wodniecka, Z., Marecka, M., Szewczyk, J., Białecka-Pikul, M., Otwinowska,A. \u0026hellip; Foryś-Nogala, M. How does L1 and L2 exposure impact L1 performance in bilingual children? Evidence from Polish-English migrants to the United Kingdom. Frontiers in psychology, 8, 1444 (2017).\u003c/li\u003e\n\u003cli\u003eRhys, M. \u0026amp; Thomas, E. M. Bilingual Welsh\u0026ndash;English children's acquisition of vocabulary and reading: implications for bilingual education. International Journal of Bilingual Education and Bilingualism, 16(6), 633\u0026ndash;656 (2013).\u003c/li\u003e\n\u003cli\u003eVermeer, A. Breadth and depth of vocabulary in relation to L1/L2 acquisition and frequency of input. Applied psycholinguistics, 22(2), 217\u0026ndash;234 (2001).\u003c/li\u003e\n\u003cli\u003eBerman, R. A. \u0026ldquo;Narrative development in multilingual contexts: A cross-linguistic perspective\u0026rdquo;. In Ludo Verhoeven \u0026amp; Sven Str\u0026ouml;mqvist (eds.), \u003cem\u003eNarrative Development in a Multilingual Context\u003c/em\u003e (pp 419\u0026ndash;428), Amsterdam/Philadelphia: John Benjamins, (2001).\u003c/li\u003e\n\u003cli\u003eGovindarajan, K. \u0026amp; Paradis, J. Narrative abilities of bilingual children with and without developmental language disorder (SLI): Differentiation and the role of age and input factors. Journal of Communication Disorders, 77, 1\u0026ndash;16 (2019).\u003c/li\u003e\n\u003cli\u003eIluz-Cohen, P., \u0026amp; Walters, J. Telling stories in two languages: Narratives of bilingual preschool children with typical and impaired language. Bilingualism: language and cognition, 15(1), 58\u0026ndash;74 (2012).\u003c/li\u003e\n\u003cli\u003eParadis, J., Soto-Corominas, A., Daskalaki, E., Chen, X. \u0026amp; Gottardo, A. Morphosyntactic development in first generation Arabic\u0026mdash;English children: The effect of cognitive, age, and input factors over time and across languages. \u003cem\u003eLanguages\u003c/em\u003e, 6(1), 51 (2021).\u003c/li\u003e\n\u003cli\u003eBitetti, D. \u0026amp; Hammer, C. S. The home literacy environment and the English narrative development of Spanish\u0026ndash;English bilingual children. Journal of Speech, Language, and Hearing Research, 59(5), 1159\u0026ndash;1171 (2016).\u003c/li\u003e\n\u003cli\u003eBohnacker, U., Lindgren, J. \u0026amp; \u0026Ouml;ztekin, B. Storytelling in bilingual Turkish-Swedish children: Effects of language, age and exposure on narrative macrostructure. Linguistic Approaches to Bilingualism, 12(4), 413\u0026ndash;445 (2022).\u003c/li\u003e\n\u003cli\u003eLindgren, J. \u0026amp; Bohnacker, U. How do age, language, narrative task, language proficiency and exposure affect narrative macrostructure in German-Swedish bilingual children aged 4 to 6?. Linguistic Approaches to Bilingualism, 12(4), 479\u0026ndash;508 (2022).\u003c/li\u003e\n\u003cli\u003eHeilmann, J., Miller, J. F., Nockerts, A. \u0026amp; Dunaway, C. Properties of the narrative scoring scheme using narrative retells in young school-age children. American Journal of Speech-Language Pathology, 19, 154\u0026ndash;16 (2010).\u003c/li\u003e\n\u003cli\u003eUccelli, P. \u0026amp; P\u0026aacute;ez, M. Narrative and vocabulary development of bilingual children from kindergarten to first grade: Developmental changes and associations among English and Spanish skills. Language, Speech, and Hearing Services in Schools, 38, 225\u0026ndash;236 (2007).\u003c/li\u003e\n\u003cli\u003eOller, D. K., Cobo-Lewis, A. B. \u0026amp; Pearson, B. Z. Profiles in early bilingual learning: Vocabulary acquisition and the distributed characteristic. International Clinical Phonetics and Linguistic Association, LA (2004).\u003c/li\u003e\n\u003cli\u003eOller, D. K., Pearson, B. Z. \u0026amp; Cobo-Lewis, A. B. Profile effects in early bilingual language and literacy. Applied psycholinguistics, 28(2), 191\u0026ndash;230 (2007).\u003c/li\u003e\n\u003cli\u003eGagarina, N. \u0026amp; Klassert, A. Input dominance and development of home language in Russian-German bilinguals. \u003cem\u003eFrontiers in Communication\u003c/em\u003e, 3, 40 (2018).\u003c/li\u003e\n\u003cli\u003eParadis, J. Individual differences in child English second language acquisition: Comparing child-internal and child-external factors. Linguistic approaches to bilingualism, 1(3), 213\u0026ndash;237 (2011).\u003c/li\u003e\n\u003cli\u003eYip, V. \u0026amp; Matthews, S. Relative clauses in Cantonese-English bilingual children: Typological challenges and processing motivations. Studies in Second Language Acquisition, 29(2), 277\u0026ndash;300 (2007).\u003c/li\u003e\n\u003cli\u003eGagarina, N. Narratives of Russian\u0026ndash;German preschool and primary school bilinguals: Rasskaz and Erzaehlung. Applied Psycholinguistics, 37(1), 91\u0026ndash;122 (2016).\u003c/li\u003e\n\u003cli\u003eGagarina, N. \u0026amp; Milano, A. Grammatical Development Within a Context of Early Bilingual Education and Language Support Programs. In Handbook of Early Language Education (pp. 1\u0026ndash;34). Cham: Springer International Publishing (2021).\u003c/li\u003e\n\u003cli\u003eCohn, N. \u0026amp; Magliano, J. P. Editors\u0026rsquo; introduction and review: Visual narrative research: An emerging field in cognitive science. Topics in Cognitive Science, 12(1), 197\u0026ndash;223 (2020).\u003c/li\u003e\n\u003cli\u003eVan den Broek, P., Rapp, D. N. \u0026amp; Kendeou, P. Integrating memory-based and constructionist processes in accounts of reading comprehension. Discourse processes, 39(2\u0026ndash;3), 299\u0026ndash;316 (2005).\u003c/li\u003e\n\u003cli\u003eBohnacker, U. \u0026amp; Gagarina, N. Introduction to MAIN\u0026ndash;Revised, how to use the instrument and adapt it to further languages. ZAS papers in linguistics, 64, xiii-xxi (2020).\u003c/li\u003e\n\u003cli\u003eSoodla, P. (2011). Picture-elicited narratives of Estonian children at the kindergarten-school transition as a measure of language competence (Doctoral dissertation).\u003c/li\u003e\n\u003cli\u003eTrabasso, T., Secco, T. \u0026amp; van den Broek, P. W. Causal cohesion and story coherence. In Heinz Mandl, Nancy L.Stein \u0026amp; Tom Trabasso (eds.), \u003cem\u003eLearning and comprehension of text\u003c/em\u003e (pp. 83\u0026ndash;111). Hillsdale, NJ: Lawrence Erlbaum Associates, (1984).\u003c/li\u003e\n\u003cli\u003eMandler, J. M., Scribner, S., Cole, M. \u0026amp; DeForest, M. Cross-cultural invariance in story recall. Child development, 19\u0026ndash;26 (1980).\u003c/li\u003e\n\u003cli\u003eBenson, M. S. Psychological causation and goal-based episodes: Low-income children's emerging narrative skills. \u003cem\u003eEarly Childhood Research Quarterly\u003c/em\u003e, 12(4), 439\u0026ndash;457 (1997).\u003c/li\u003e\n\u003cli\u003eB\u0026uuml;rkner, P. C. Advanced Bayesian multilevel modeling with the R package brms. arXiv preprint arXiv:1705.11123 (2017).\u003c/li\u003e\n\u003cli\u003eVehtari, A., Gelman, A. \u0026amp; Gabry, J. Practical Bayesian model evaluation using leave-one-out cross-validation and WAIC. Statistics and computing, 27, 1413\u0026ndash;1432 (2017).\u003c/li\u003e\n\u003cli\u003eSivula, T., Magnusson, M., Matamoros, A. A. \u0026amp; Vehtari, A. Uncertainty in Bayesian leave-one-out cross-validation based model comparison. arXiv preprint arXiv:2008.10296 (2020).\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"","lastPublishedDoi":"10.21203/rs.3.rs-5672219/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-5672219/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"This study investigates children’s narrative skills, using the Multilingual Assessment Instrument for Narratives (MAIN), a theory-based tool adapted to 100 languages. How do children’s overall narrative skills, including the use of factual and inferred components, and the development of complete episodes, differ across languages and between monolinguals and bilinguals? To answer this, we examined 2608 comparable fictional narratives of 1189 monolingual and bilingual children aged 3-13 years, speaking 33 languages. Children told or retold stories elicited with parallel picture sequences. Narratives were analyzed for inclusion of five essential story components (internal states as initiating events, goals, attempts, outcomes, internal states as results), overall narrative structure, use of factual and inferred information, and presence of complete episodes. Monolinguals initially outperformed bilinguals, but bilinguals’ narrative skills developed more rapidly and eventually surpassed those of monolinguals, with all children performing better in story retelling than storytelling, especially in their L2. Across languages, children were more likely to produce factual components (outcomes/attempts) than inferred ones (internal states/goals). The likelihood of producing a complete narrative increased with age; monolinguals consistently outperformed bilinguals, even under stricter coding criteria. Patterns of use for factual and inferred components were consistent across languages, suggesting universal strategies in narrative development.","manuscriptTitle":"A story of stories: a large-scale cross-linguistic study of young children’s narratives","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-02-06 08:36:22","doi":"10.21203/rs.3.rs-5672219/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"12b40b94-4bdb-483d-aba3-00387543dfb0","owner":[],"postedDate":"February 6th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":42324280,"name":"Humanities/Language and linguistics"},{"id":42324281,"name":"Social science/Development studies"}],"tags":[],"updatedAt":"2025-07-08T14:58:03+00:00","versionOfRecord":[],"versionCreatedAt":"2025-02-06 08:36:22","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-5672219","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-5672219","identity":"rs-5672219","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}