1. Scene: the filter that silences the home language

A rainy Monday, a group of four-year-olds. A child comes through the door with a story in the language of his home: his grandmother saw the water in the yard, put an umbrella in his hand and asked him to tell how the wet earth smelled. The teacher receives him calmly. In the corner the tablet is still open. It does not show the grandmother. It shows a dashboard of “target-language minutes,” a streak of correct answers and a filter notice: the weekend audio was marked as speech outside the language of the activity.

The teacher follows the notice, asks for the school language and brings the microphone closer. Machine translation returns a flat sentence: the grandmother, the errand and the smell disappear. The drill offers a drawing of an umbrella and a blank for the target-language word. The child repeats it. The cell turns green, the “L2 progress” score rises by one point and the filter archives the home audio. Whoever looks only at the screen will see support. Whoever heard the door will see the story outside the record.

Each artifact cuts differently, and the score sums those cuts into a number that no one saw being built between the child and the teacher. The vignette is not a sample: it has no group size and it produces no coefficient. What follows asks what would have to happen for the full repertoire to remain a resource for play, care and conversation.

2. Problem: what counts as language support in preschool

The problem is one of category. Products and routines promise language support because they produce a visible output: a translated sentence, a correct item, an accumulated minute, a discarded audio file, a score that rises. Those outputs are software records. Plurilingual development is something else: a child uses the repertoire already brought —home languages, the school language, gesture, object, play— to build a meaning with someone. Confusing the record with development authorizes closing the conversation when the dashboard turns green.

The object of this text is translanguaging pedagogy together with digital mediation in early childhood education: the full repertoire set against automatic translation, the monolingual drill, the dashboard of target-language minutes and the filter that silences the home language. Reading an album, a center’s language policy and emergent literacy enter only when a source in this corpus makes them part of that mediation. This essay does not reopen a manual of reading questions, a general map of multilingual education or a technique of marks on paper.

The protocol counts shared meaning-making, even when the path passes through more than one language; the multimodal repertoire —word, gesture, touch, object, song, drawing and play—; family practices; mediation that designs the play, watches the turn and decides whether to take up, extend or stay quiet; and, if there is a screen, a story that keeps the chosen voice and returns it to the group. Without that, the app’s output stays at the ceiling of the opening box. A target-language minute is easier to display than a conversation in which one child regulates another by shifting footing, tone or language. The question is what recent research shows and what may be inferred, without inventing trials of apps.

3. State of the art: translanguaging versus automatic translation

In the band this corpus can actually be read for, translanguaging is not permission to use the other language and then return to the correct one. It is the repertoire as a practical unit. Ridley and Rowe (2024) state that children translanguage naturally and mobilize the whole semiotic repertoire: words, gesture, touch, song, play and art. The title announces themes and possibilities for the theory, pedagogy and policy of early childhood literacy. Framework. That formulation and that object are what is used here. No list of themes is attributed that the available abstract does not unfold, and no software recommendation is attributed either.

Olğun-Baytaş, Lyu and Yoon (2023) reconstruct how, in early childhood language education in the United States, monoglossic and raciolinguistic ideologies associate one language with one group and label children as learners of the dominant language. Entry assessments, they argue, have been criticized for turning bilingualism into a deficit to be remediated. Framework. The filter in the vignette performs that operation in small. Machine translation aligns with the same norm if its success consists in erasing the first language until only the second remains.

Oakley, Steele, Robinson, Dobinson, Dovchin and Cumming-Potvin (2025) propose, in a conceptual article, Playworld Translanguaging: pedagogical translanguaging and playworlds, as an intentional, inclusive and rights-based pedagogy. Both approaches, they hold, may favor language, literacy, conceptual understanding, problem solving, identity, agency and wellbeing, but there is little theorization and little research on the combination. Framework. It does not authorize measured gains. Robinson, Steele, Dobinson, Oakley, Manado and Grace (2026) report the enactment in the remote Kimberley: kindergarten and early-years educators, a mostly Aboriginal group of children, Bardi Jawi as an ancestral language and place-linked play. They followed Indigenist research principles and critical participatory action research, and treated the Aboriginal child as a producer of knowledge. The pathway was playful literacy anchored in cultural, linguistic and identity funds. Empirical finding of a situated collaboration, not a universal program, not an effect size and not a trial of automatic translation.

Ghim (2025) titles a qualitative case study of Chinese teachers who accompany Chinese children and families in the United States: transcending translation. Title object. Without an open abstract, no scenes, samples or results are extracted. The object is enough: translating does not exhaust support. An app can produce an equivalent and is not studied here as a substitute for mediation. Yazıcı and Keleş (2026) offer the wider map, with a review distinct from this essay’s. They follow PRISMA and include 43 full peer-reviewed articles, from 2014 to 2023, retrieved in Web of Science, ERIC and Scopus. They read them on five dimensions: trends, methodology, focus, implementations and overall effects on educators and emergent bilingual preschoolers. Their synthesis holds that translanguaging facilitated changes in teachers’ perceptions and practices and favored cognitive, cultural and emotional development by offering an inclusive and interactive environment. Review finding of those 43 texts and of that window, which begins before 2021. This essay does not reanalyze those studies, does not inherit their effects as its own figures and does not turn them into an endorsement of minute dashboards.

4. Method: a PRISMA-lite review, 2021–2026

The PRISMA-lite review is a short procedure, not a systematic review that claims exhaustiveness. The window is 2021–2026. It started from a curated list of 24 records with a DOI verifiable in Crossref. It does not claim to have screened a complete database again. Eligibility, inside that list, asked for early childhood education or early childhood teacher education, translanguaging as the central object and a year inside the window. Of the 24, the corpus keeps 22. Left out were a conference-series communication, whose abstract advances programmatic claims without a design this essay could audit, and an article in a venue of doubtful editorial standard. Their claims are not used.

Inside the 22, kept with the boundary stated, are literacy practices of teacher candidates in primary school (Beauchemin, Shen and Zhang, 2024) and a community writing program that is not a preschool (Abraham, 2023). They illuminate discursive shifts and the materiality of the repertoire. They are not read as preschool trials. Each reading carries a mark. Empirical finding: the open abstract reports a setting, participants or practices, and this text stays there. Framework: a conceptual or ideological text, without its own classroom result. Title object: there was no usable open abstract; the title is named and findings are not manufactured. Design inference: a hypothesis of this essay, made by placing the corpus next to the artifacts. It is not a datum from the sources. Yazıcı and Keleş (2026) remain the systematic PRISMA of the neighborhood, with another window and another 43 texts. Naming them here does not merge the procedures. A number is attributed to the study that reports it and is not combined with others. No new N, d, r, AUC, percentage or DOI is invented. If an abstract reports a correlation and does not publish the coefficient, no value is assigned. The vignette in section 7 is composite and anonymized. It is not fieldwork of this essay.

5. Findings: translanguaging pedagogy in the classroom

Robinson et al. (2026) leave the enactment already situated: the Kimberley, Bardi Jawi, place-linked play, children’s funds of knowledge as the matter of the play. Empirical finding. Language is not added after a neutral activity; the activity is co-designed so that language and place are the play. A target-language drill reverses that direction. Oakley et al. (2025) had warned that the combination was little researched. The 2026 report does not close that warning with an effect.

Dikilitaş, Bahrami and Erbakan (2023) study a bilingual preschool in predominantly monolingual Turkey. At one level, teachers assign Turkish or English to each task as languages of instruction and of interaction. At the other, children alternate without coercion. Six co-teachers were interviewed online and there were diaries with the children’s utterances. Empirical finding. The unforced back-and-forth is the level a purity filter punishes. No coefficient of bilingual gain is reported.

Akmalia and Faizin (2026) examine a translanguaging workshop in the education of Arabic teachers for early childhood. Semi-structured interviews, reflective diaries and lesson plans, read through thematic analysis, report a shift from target-language-only orientations toward more flexible, culturally responsive and inclusive perspectives: Arabic as meaning-making with several resources. Empirical finding. The abstract does not give the number of participants. It is not invented.

Palojärvi, Koivula, Mård-Miettinen and Rutanen (2023) reconstruct the practical theory of an expert teacher when home languages enter. With nexus analysis, pedagogical tact changes form according to the situation, and the use of children’s linguistic expertise holds, even in brief activities. Translanguaging is initiated by the teacher and by the children: a dual locus of power. Practices coincided in the main with the theory already verbalized. Empirical finding of one case.

Sitorus, Sahiruddin and Emaliana (2026) observe and interview two teachers in an international preschool, in an English-medium classroom, with distinct trajectories. The abstract suggests inclusion, a deeper linguistic understanding and cultural awareness, and mentions challenges without detailing them. It calls for interviews with students and with families. Empirical finding bounded to those two teachers. It is not generalized to preschools in Indonesia.

Hamman-Ortiz and Romero (2025) title a comparative case study: translanguaging as mediated praxis, with bilingual and monolingual teachers experimenting with that pedagogy. Title object. Without an open abstract it is not said which group shifted more, in which tasks or with what intensity. The pedagogy is not a switch that works the same for any linguistic biography. An app does not mediate that praxis.

Beauchemin, Shen and Zhang (2024) analyze shifts of teacher candidates in read-alouds with translanguaging. The abstract places them in primary school, not with three- to five-year-olds: the transfer is analogical and is marked as such. Some position multilingual pupils as experts more firmly than others; some open a knowledge the candidate does not have; others open the defense of linguistic, cultural and textual rights. Empirical finding. There is no sample size. The unit is the decision of one moment, not the average of minutes.

Harvey and Wong (2024) study a Mandarin immersion preschool in the United States, mostly with learners who have no family heritage of that language and under a Mandarin-only policy. Despite the rule, the teachers practiced covert translanguaging to extend discourse. They identify thirteen strategies in two zones: socializing children into school norms, not only into the language, and focusing on content learning. Empirical finding. Thirteen is a categorization, not an efficacy indicator. Extended discourse crossed the purity rule. The rule made it clandestine.

Yoon and Dong (2024) follow two five-year-old children, immigrants and emergent bilinguals, in toy-unboxing play, with English and Korean repertoires. The play creates a translanguaging space in the service of meaning. Flexibility appears in three domains: play with the toy and with the self; emotion with family and other intimates; transcultural relation with peers and with a virtual audience. Empirical finding of two cases. The video welcomes the repertoire. It does not translate it into a single output language.

Samuelsson (2022) examines touch in a multilingual preschool with two-year-olds, where translanguaging pedagogies do not yet circulate widely. Touch creates a floor of shared experience on which languages can be learned and allows a sensory exploration of multicultural experiences. Empirical finding. At that age the condition passes through one body next to another body. Pointing at a card on a screen does not reproduce that floor.

Valencia Mazzanti (2024) describes, in a three-year qualitative study, linguistic practices of multilingual children learning mathematics: multimodal representations drawn from productions, interactions, a diary and audio. The argument is that one must leave a prior idea of language and respond to diverse practices, and that this understanding can favor meaningful mathematical learning by dismantling power hierarchies. Empirical finding. There are no scores. A score in a single language does not see the representation with which sense is being built.

Choi (2024) analyzes, with microethnography and multimodal analysis, two literacy events of a family in a science museum. A child and his mother organize the performance, joint attention and the narrative. Pointing brings them into the reading of the signage and allows the mother to translate it. Choi treats that translation as repertoire, not as a defect. Empirical finding of two events. Human translation occurs inside shared attention.

Yin (2026) ethnographs, over one year, a rural Chinese preschool. Longlinghua-speaking children develop bilingual competence while learning Putonghua: they shift footing, adjust tone contour and revocalize roles in order to gain uptake, regulate peers and belong. The work challenges deficit and calls for a pedagogy of recognition of local dialects alongside the national norm. Empirical finding. The summary used does not report group size. A filter on the local language would erase those resources.

Abraham (2023) is cited with the boundary in view. In a community writing center for a Latinx community in Philadelphia, with Saturday workshops and emergent bilingual children, a ball of sticky putty opened a literacy moment outside the sociocultural frame that had been planned. The text returns to bodies, laughter and material. Empirical finding of a community program, not of a preschool. There is meaning stuck to an object and to a laugh.

McClain, Mancilla-Martinez, Flores and Buckley (2021) conduct an explanatory sequential mixed-method study in English-dominant preschools. In microphone recordings of emergent bilinguals from Spanish-speaking homes (N = 12) they report a correlation between adults’ use of Spanish and open questions. The coefficient is not published and is not invented here. The qualitative strand suggests fuller participation when there is teacher translanguaging. Empirical finding. The open question traveled with the home language. A closed target-language item does not see that pair.

Rahmadani (2025) titles translanguaging strategies for students with autism spectrum disorder in early childhood English learning. Title object. Without an open abstract, strategies are not listed and efficacy is not claimed. Rahmadani (2026), in a qualitative case in Indonesia, crosses family language policy and translanguaging at home and at school, with interviews with families and the teacher and with observation of both scenes. Families’ ideologies and everyday practices shape bilingual development and complement classroom strategies. The authors suggest that a strong alignment appeared associated with greater participation, vocabulary acquisition and social communication. Qualitative case finding. There is no magnitude. “Appeared associated” is not converted here into a measured effect.

6. Findings: digital mediation and its limits

None of the 22 works trials an automatic translation app, a drill package or a dashboard of target-language minutes. That trial is not manufactured. There are practices in which the digital, the media form or a language rule occupies a precise place. From there come design inferences, not product verdicts.

Yoon and Dong (2024) offer the closest analogue: an unboxing video that keeps English and Korean, also in family emotion and before a virtual audience. Design inference: a screen can be the frame of the repertoire if it does not translate the child toward a single output language. Recording the chosen voice, returning it to the group and letting the family recognize the intonation goes in that direction. Replacing the voice with a flat sentence goes the other way. Choi (2024) shows the other limit. The mother translates the signage because a finger put the text in common. Design inference: translating can be mediation when the person who shares the scene decides the fragment and holds it with gesture. Autotranslation omits that decision. The corpus does not say that all translation is an error. It says that the translation observed served joint attention.

Harvey and Wong (2024) give the institutional analogue of the filter. Under a Mandarin-only policy, extended discourse was sustained with covert translanguaging and thirteen strategies. Design inference: penalizing the home language reproduces that condition. A software brand is not measured; the logic is named. Samuelsson (2022) adds the body at age two: touch is the floor on which languages are learned. A carousel that counts taps occupies the hands and displaces that touch. There is no trial that compares the two.

Oakley et al. (2025) and Robinson et al. (2026) prevent calling any screen game a Playworld. The framework asks for intentionality and states that the evidence is scarce; the Kimberley shows co-design, place and an ancestral language, not a downloadable package. Design inference: a digital game would come close only if educators co-design it, if the language of the place enters the plot and if children bring their funds. A drill meets none of the three, and its dashboard of minutes does not add the co-design. Dikilitaş et al. (2023) separate the language assigned to the task from the back-and-forth without coercion: machine translation collapses the two. McClain et al. (2021) add the kind of question. In twelve emergent bilinguals, adult Spanish correlated with open questions, with no coefficient published. Design inference: what traveled with the home language was the open question; the drill rewards the closed item.

The digital story this essay can defend, inside what the corpus shows, resembles the unboxing that keeps the repertoire, the pointing that asks for human translation and the discourse that crosses a purity rule. It can welcome the repertoire if teacher mediation remains in charge of meaning. Autotranslation, the monolingual drill, the minutes dashboard and the filter operate as substitutes. The corpus does not exonerate them for being digital, nor condemn them with a number no one measured. They remain at the ceiling of the box.

7. Anonymized classroom case (ages 3–5, plurilingual)

The vignette is composite and anonymized. It names no center, it is not a data set and it does not authorize N, d or r. A group of three- to five-year-olds brings distinct family repertoires: a school language and home languages, without saying how many or which, because fixing them would invent an enrollment. The cycle lasts six weeks of an ordinary day. That duration is the frame of the composition, not the length of a fieldwork study.

In play, a girl narrates a pretend kitchen in the language of her home. A peer answers in the school language and passes her a spoon. The teacher stays in the play, takes up who is cooking and for whom, and returns a sentence that links the two languages without stopping the first. She does not open the tablet or ask for an item. The kinship with Robinson et al. (2026) is one of design: language enters as a resource of the play, not as a test after the play. In the reading of an album, she asks first about what is seen. She accepts the home language and the school language on the same page, and only afterward offers a school-language word for a detail the group already wants to name. There is no drill at the end. No vocabulary gain is claimed.

In the digital story, two children record thirty seconds about the wet yard, each in the language chosen. There is no machine translation. In the afternoon the group listens to both voices. A family recognizes a home word and adds a detail. The recording stays in the language of origin. The minutes dashboard exists on another screen in the office and no one opens it to decide whether the recording was worth it. The next morning the teacher brings that word, does not correct the pronunciation and writes it next to the school word as a working pair. There is no L2 score. The vignette claims no improvement in vocabulary, literacy or belonging, no number of children, languages or recordings, no balanced bilingual teacher, no written policy and no comparison with a drill room. It is not a replica of Yin (2026), of Choi (2024) or of McClain et al. (2021).

8. Discussion: criteria for genuine mediation

The four criteria are design hypotheses. They are not a validated norm, they have no cutoff and they do not certify a center. They ask whether mediation remains in the hands of the person who shares the scene.

  1. The full repertoire is welcome. It is documented by the shift of footing, tone and role between Longlinghua and Putonghua (Yin, 2026), by English and Korean in play, family and audience (Yoon and Dong, 2024), by touch at age two (Samuelsson, 2022), by an ancestral language made into the activity (Robinson et al., 2026), by the unforced back-and-forth (Dikilitaş et al., 2023) and by covert entry under a Mandarin-only rule (Harvey and Wong, 2024). Hypothesis: genuine mediation leaves that repertoire visible. Filing it in a filter suspends the hypothesis.
  2. Teacher observation is situated. Palojärvi et al. (2023) describe a pedagogical tact that changes form according to the situation and a child’s expertise that holds, with a dual locus of power. Beauchemin et al. (2024), at the primary-school boundary, analyze the decision of each moment in the read-aloud. Akmalia and Faizin (2026) look for the change in diaries and lesson plans. McClain et al. (2021) move from a correlation with no published coefficient to a qualitative reading of Spanish. Hypothesis: situated observation listens to what is being built in that turn. A minutes dashboard and an L2 score observe something else: duration and accuracy in the language declared as the target. They can be archived. They do not replace the person who is in the scene.
  3. There is a link with the family and with the home language. Choi (2024) places the mother translating inside joint attention. Rahmadani (2026) suggests, with no magnitude, that alignment between family policy and the classroom matters for the participation of children with autism. Ghim (2025) puts in the title the accompaniment of families beyond translation. Sitorus et al. (2026) call for interviews with families and with the children. Palojärvi et al. (2023) and McClain et al. (2021) take home languages as matter, not as residue. Hypothesis: that language remains a bond with the person who cares at home. The filter penalizes it and calls the cut target-language quality.
  4. The digital serves shared meaning. Yoon and Dong (2024), Choi (2024) and Harvey and Wong (2024) offer, respectively, a genre that unfolds the repertoire, a human translation of a text pointed to in common and extended discourse that linguistic purity did not know how to house. Oakley et al. (2025) ask for intentionality and remind us that the combination of play and translanguaging is little researched: the hypothesis is not a program with effects already known. Hypothesis: the tablet or the recording is used when it lengthens a meaning the group is already building. Autotranslation, the drill, the minute and the filter reverse the service and replace the person who mediates.

The four hold together. Repertoire without observation becomes a slogan; observation without family stays in the room; a bond that the screen translates alone loses the recognizable voice; a recording inside a filter repeats the clandestinity that Harvey and Wong (2024) documented. Monday is not repaired with a better translation. It is repaired if the grandmother’s sentence returns to the play.

They would be falsified by a single-language program, without back-and-forth and with families inside the decision, that produced the same participation, the same extended discourse and the same belonging, measured with designs this essay does not have. They would also be falsified by a minutes dashboard that predicted that participation better than situated observation. That study is not among the 22. Hamman-Ortiz and Romero (2025), read through the title, remind us that praxis is mediated by the person who experiments: the teacher is not a peripheral of the software.

9. Limits

This text interprets 22 records from a list of 24, window 2021–2026. It is not a meta-analysis and it does not reanalyze the 43 articles of Yazıcı and Keleş (2026), whose window begins in 2014. The figures —43 articles, N = 12, six co-teachers, two teachers, two children, two events, thirteen strategies— belong to each abstract and are not pooled. Where there is no coefficient, none is estimated.

Ghim (2025), Hamman-Ortiz and Romero (2025) and Rahmadani (2025) are cited for the object of the title. Ridley and Rowe (2024) are cited for the formulation of the semiotic repertoire and for the object of the title, without a taxonomy the text at hand does not give. Beauchemin et al. (2024) observe primary school. Abraham (2023) observes a community center in Philadelphia. Valencia Mazzanti (2024) spans three years without a more precise age band. Samuelsson (2022) is at age two. Robinson et al. (2026) include kindergarten and the early years. Using them to think about early childhood education is an inference of relevance, not a single sample.

Rahmadani (2026) reports no magnitude. Oakley et al. (2025) state that research on their own combination is scarce. Robinson et al. (2026) is not an exportable program. Harvey and Wong (2024) document a Mandarin-only policy, not a software filter. There is no trial of a translation brand, of a commercial drill package or of a commercial minutes dashboard. The critique is a design inference from back-and-forth, the open question, extended discourse, gesture, touch, situated play, family and teacher belief. The six-week vignette adds no evidence. The map —Australia, the United States, Turkey, Indonesia, rural China, a museum, a community center— does not authorize a single recipe for language policy.

10. Conclusions

Monday fits in a green cell and, told that way, looks like support: a translated sentence, a repeated umbrella, a minute, an archived audio file and an L2 score that rises. The grandmother’s story stays outside. That difference is the thesis. An automatic translation app, a package of monolingual drills, a dashboard of “target-language minutes” or a filter that penalizes the home language does not constitute support for plurilingual development if the protocol is missing: shared meaning, a multimodal repertoire, family practices, teacher mediation and digital storytelling that keeps the full repertoire.

The corpus shows the craft somewhere else: play co-designed with an ancestral language (Robinson et al., 2026), without the effect Oakley et al. (2025) had already marked as pending; unforced back-and-forth (Dikilitaş et al., 2023); pedagogical tact (Palojärvi et al., 2023); beliefs in motion in education for teaching Arabic (Akmalia and Faizin, 2026); the decision of one moment (Beauchemin et al., 2024); extended discourse despite a purity rule (Harvey and Wong, 2024); unboxing that keeps two languages (Yoon and Dong, 2024); human translation asked for by a finger (Choi, 2024); touch at age two (Samuelsson, 2022); tone and role in order to belong (Yin, 2026); home policy alongside the room’s (Rahmadani, 2026); and the title that asks to transcend translation with families (Ghim, 2025). The same shift, each in its own mark, is held by Sitorus et al. (2026), Valencia Mazzanti (2024), Abraham (2023), McClain et al. (2021), Olğun-Baytaş et al. (2023) and Ridley and Rowe (2024). Rahmadani (2025) and Hamman-Ortiz and Romero (2025) remain title objects: the classroom is not homogeneous and praxis is not a switch.

None of that fits in a target-language minute. The minute measures staying in the declared language. The craft measures whether meaning was built with the repertoire the child brought. The digital has a place when it lengthens that meaning: a voice the family recognizes, a co-designed play, a text translated because a finger put it in common. It loses the place when it substitutes. The following Monday does not require a new platform. The sentence at the door is heard through to the end, the home language is left inside the play and the page, the turn is observed and the voice is returned as it sounded. The full repertoire is the resource. The green cell can wait.

Laboratorio Editorial de NEXTECH.IA / Ingeniero Mitre

References

  1. Abraham, S. (2023). Paradigmatic fronteras: Troubling available design and translanguaging with sticky literacy. Journal of Early Childhood Literacy. https://doi.org/10.1177/14687984211021944
  2. Akmalia, F., y Faizin, K. (2026). Challenging monolingual paradigms in early childhood Arabic teacher education: Can translanguaging reshape pre-service teachers’ beliefs? Journal of Early Childhood Teacher Education. https://doi.org/10.1080/10901027.2026.2702970
  3. Beauchemin, F., Shen, Y., y Zhang, G. (2024). Exploring teacher candidates’ discursive shifts in translanguaging pedagogies during literacy instruction. Journal of Early Childhood Literacy. https://doi.org/10.1177/14687984241277011
  4. Choi, M.-S. (2024). Translanguaging space through pointing gestures: Multilingual family literacy at a science museum. Journal of Early Childhood Literacy. https://doi.org/10.1177/14687984241276290
  5. Dikilitaş, K., Bahrami, V., y Erbakan, N. T. (2023). Bilingual education teachers and learners in a preschool context: Instructional and interactional translanguaging spaces. Learning and Instruction. https://doi.org/10.1016/j.learninstruc.2023.101754
  6. Ghim, H. (2025). Transcending translation: A qualitative case study of Chinese teachers’ translanguaging practices in supporting Chinese children and families in the United States. Early Childhood Education Journal. https://doi.org/10.1007/s10643-025-01901-2
  7. Hamman-Ortiz, L., y Romero, D. (2025). Translanguaging as mediated praxis: A comparative case study of bilingual and monolingual teachers experimenting with translanguaging pedagogy. Teaching and Teacher Education. https://doi.org/10.1016/j.tate.2024.104878
  8. Harvey, R. E., y Wong, K. M. (2024). Promoting emergent literacy in preschool through extended discourse: Covert translanguaging in a Mandarin immersion environment. Journal of Early Childhood Literacy. https://doi.org/10.1177/14687984241276292
  9. McClain, J. B., Mancilla-Martinez, J., Flores, I., y Buckley, L. (2021). Translanguaging to support emergent bilingual students in English dominant preschools: An explanatory sequential mixed-method study. Bilingual Research Journal. https://doi.org/10.1080/15235882.2021.1963014
  10. Oakley, G., Steele, C., Robinson, C., Dobinson, T., Dovchin, S., y Cumming-Potvin, W. (2025). Towards a playworld translanguaging approach in early childhood education. Australian Review of Applied Linguistics. https://doi.org/10.1075/aral.24132.oak
  11. Olğun-Baytaş, M., Lyu, S. R., y Yoon, Y. (2023). Embracing translanguaging pedagogy in early childhood education: Challenges and possibilities for promoting equity and diversity. Journal of Education, Language, and Ideology. https://doi.org/10.67372/xzcr8e29
  12. Palojärvi, A., Koivula, M., Mård-Miettinen, K., y Rutanen, N. (2023). Manifestations of an expert teacher’s practical theory of language pedagogy in translanguaging situations in early childhood education and care. Apples - Journal of Applied Language Studies. https://doi.org/10.47862/apples.129191
  13. Rahmadani, A. (2025). Translanguaging strategies for autism spectrum disorder (ASD) students in early childhood English language learning. International Journal of Early Childhood. https://doi.org/10.1007/s13158-024-00412-2
  14. Rahmadani, A. (2026). Family language policy and translanguaging practices as support for language development in early childhood students with autism spectrum disorder in Indonesia. Discover Education. https://doi.org/10.1007/s44217-026-01149-x
  15. Ridley, J., y Rowe, L. W. (2024). Translanguaging and early childhood literacy: Themes and possibilities for theory, pedagogy, and policy. Journal of Early Childhood Literacy. https://doi.org/10.1177/14687984241276305
  16. Robinson, C., Steele, C., Dobinson, T., Oakley, G., Manado, S., y Grace, K. (2026). Playworld Translanguaging: A place-based approach for translanguaging play pedagogy in early childhood. Journal of Early Childhood Literacy. https://doi.org/10.1177/14687984261488127
  17. Samuelsson, R. (2022). Touch and translanguaging in a multilingual early childhood education setting. Multimodality & Society. https://doi.org/10.1177/26349795221109355
  18. Sitorus, I. E., Sahiruddin, y Emaliana, I. (2026). “Miss, I want minum!” How Indonesian preschool teachers perceive translanguaging practices in the classroom. Journal of Education of English as a Foreign Language. https://doi.org/10.21776/ub.educafl.2026.009.01.05
  19. Valencia Mazzanti, C. (2024). Translanguaging, multilingualism, and multimodality in young children’s mathematics learning. Contemporary Issues in Early Childhood. https://doi.org/10.1177/14639491221130780
  20. Yazıcı, E. A., y Keleş, N. (2026). Translanguaging as a bilingual education pedagogy in early childhood education: A systematic review. Pacific Early Childhood Education Research Association. https://doi.org/10.17206/apjrece.2026.20.1.179
  21. Yin, X. (2026). Translanguaging to learn and play: Young children’s bilingual communicative competence in a Chinese rural preschool. Anthropology & Education Quarterly. https://doi.org/10.1111/aeq.70088
  22. Yoon, Y., y Dong, P. I. (2024). Cultivating translanguaging spaces: Young emergent bilingual children’s toy unboxing play. Journal of Early Childhood Literacy. https://doi.org/10.1177/14687984241276316