Dialogic reading in early childhood education

1. Vignette: the story read aloud and the dialogue that never happened

The vignette is not a field case and it is not a datum. It does not demonstrate that reading straight through harms language, nor that the tablet cancels the book. It is a composite pattern: there is a book, there is an adult, there are children, and still there may be no dialogic reading. The rest of the article reads that pattern in studies from 2021 to 2026. The question is not whether it is advisable to read more titles. It is what distinguishes sharing the book from doing dialogic reading, and under what conditions that distinction matters for vocabulary, narration and emergent language between the ages of zero and six.

2. Problem: sharing the book is not doing dialogic reading

Shared book reading names any episode in which an adult and a child, or an adult and a group, look at a book together and the adult reads or retells. Dialogic reading names the subset in which the adult opens turns, evaluates, expands and waits, and the child is an interlocutor, not only an addressee. The first practice can be routine and daily. The second requires talk that is not on the page: extra-textual talk. Confusing the two is the problem of this article.

The distinction is not semantic. If a preschool counts minutes of story time, or a home declares that it reads every night, the indicator may be full and the dialogue empty. Cutler et al. (2026) observe that higher- and lower-language-gain classrooms can resemble one another in frequency, duration and books: the shared object is not enough. Read et al. (2025) remind us that what counts is the spontaneous conversation alongside the text. Dillon and Newman (2023) add that, even when the practice is called dialogic, its effect depends on its being delivered as it was designed. Özen-Uyar and Yigit-Gencten (2026) locate the home as a context of literacy practices with family participation. None of them authorizes treating the open book as dialogic reading.

The question takes the form population–concept–context. The population is girls and boys from zero to six years of age and the adults who mediate: mothers, fathers, teachers, speech-language pathologists, and caregivers in a neonatal unit. The concept is the quality of dialogue around the book—extra-textual talk—and the conditions that sustain it: continuity across contexts and fidelity. The context covers the home, the early childhood classroom and clinical settings. The question is not how many picture books it is advisable to own, but what the literature shows when the book is shared and the conversation is documented—or omitted. Each claim is marked as an empirical finding, a review finding, a conceptual framework or a design inference. Where the abstract does not carry a figure, none is invented. Four sources are retained only for the object of the title.

3. Framework: extra-textual talk, fidelity and continuity

This section sets the conceptual framework. Three notions organize the corpus: extra-textual talk, intervention fidelity and home–classroom continuity. They are not a causal model tested here. They are lenses so as not to confuse the object—the shared book—with the practice—sustained dialogue.

Read et al. (2025) review forty-five studies, from 1977 to 2022, on caregivers’ extra-textual talk in dyadic reading. Review finding: that spontaneous conversation alongside the text is a common and critical feature for cultivating interactions and supporting language. Systems are grouped by content, by the reader’s global style, and by quantity and complexity. Influences come from readers, children and books: a triangular model. Conceptual framework: extra-textual talk is not an ornament of the story; it is the material that turns the episode into dialogue. Open questions, expansion and wait time are used here as descriptors of quality, not as a protocol validated in this review.

Dillon and Newman (2023) shift the focus to delivery. Dialogic reading is, in their account, an intervention with evidence on oral language, vocabulary and print concepts, but that evidence presupposes fidelity: that educators do what the design asks. Conceptual framework: fidelity is not a formality; it is the condition for what is dialogic not to dissolve into straight-through reading. Naming the practice is not sustaining it. The supports they test are discussed under training.

Özen-Uyar and Yigit-Gencten (2026) supply the third lens with a scoping review of twenty-four studies in Türkiye (2007–2024) on interventions that involve parents in shared reading. Review finding: even relatively brief interventions positively affect child language and parents’ reading behaviors, particularly those of mothers. The underrepresentation of fathers, of children with special needs and of families of low socioeconomic status, and the scant attention to infants and toddlers, are marked as gaps. Conceptual framework: home–classroom continuity is not sending the same title home; it is a literacy practice with family participation, with biases as to who enters the evidence. Design inference: a preschool that reads on Friday and a home that plays an audiobook do not, by juxtaposition, fulfill that continuity. Extra-textual talk describes the turn; fidelity describes whether what was trained occurs; continuity describes whether the dialogue has a second context.

4. Method: narrative review focused on contexts

The design is a critical narrative review, not a meta-analysis. The studies do not share a metric, an age, a language or an outcome, and they are not combined here into one effect size. The purpose is to articulate a category argument—sharing the book does not by itself constitute dialogic reading—with verified sources, organized by context: home, classroom, interventions, training, and clinical or bilingual conditions. The unit of reading is the abstract deposited in Crossref or, if it is missing, the title. Figures that the abstract does not contain are not reconstructed.

The search was documented on 16 October 2026 in the Crossref REST API, without Scopus, Web of Science or ERIC as databases in their own right. Sixteen queries of sixty rows: 960 nominal records. After automatic eligibility, the seed file contains 261 records. Manual screening, by title and abstract, was neither double-blind nor conducted with two reviewers. Priority was given to dialogic reading and shared book reading in early childhood education, with coverage of the home, the classroom, intervention, review, dual language learners and developmental language disorder. Generative artificial intelligence as an axis, and feedback literacy in higher education, were excluded. Twenty-four articles entered, all from the seed. Twenty have a usable abstract. Four are retained only by the title: Ghosh et al. (2026), Ergenekon and Işıkoğlu (2026), Gonzalez et al. (2024) and Kennedy and McLoughlin (2023). Table 1 gathers those figures, which are not effect sizes.

An empirical finding designates a result that the abstract of a primary study reports; a review finding, someone else’s synthesis; a conceptual framework, definitions not presented as proof of efficacy; a design inference, a consequence for the preschool, the home or training that this review draws and that the sources did not measure as such. The corpus does not include direct Mexican evidence: that absence is declared in the limitations; it is not filled in.

Table 1. Identification and screening flow in Crossref, documented on 16 October 2026. Figures from the flow file, not effect sizes.

5. Home findings: caregivers, screens and languages

The home is the first context in which the book can be shared without dialogue. Rosslund et al. (2025) examine, in a pre-registered study, shared reading, daily screen time and vocabulary in 1,442 Norwegian infants at twelve and twenty-four months. Empirical finding: a positive association between shared reading and vocabulary at both ages, and a negative association between screen time and vocabulary at twenty-four months. Exploratory analyses: the positive relationship with expressive vocabulary at twelve months was stronger in lower socioeconomic-status groups, read as a possible compensatory mechanism. The abstract does not authorize treating the vignette’s tablet as a single cause. It does prevent equating “there is a book at home” with “there is conversation around the book.”

Yang et al. (2025) examine an understudied feature of maternal talk: idioms. Thirty-three Chinese children in Norway (3;0–5;5) and their mothers were observed in shared reading at the outset; vocabulary in Chinese was assessed three times over one year. Empirical finding: mothers used, on average, 1.8 idioms and explained one third. Use was correlated with amount of talk and lexical diversity, and predicted growth in receptive vocabulary in Chinese. The authors speculate that it could mark parental linguistic sophistication. Design inference: extra-textual quality includes resources that the text does not supply. An audiobook in the kitchen produces neither those 1.8 idioms nor their explanation.

Conica et al. (2023) look at fathers—underrepresented, as Özen-Uyar and Yigit-Gencten (2026) noted—and at book type. Forty-six father–child dyads (twenty girls); fathers’ mean age, 38.74 years (SD = 5.74); children’s, 37.89 months (SD = 2.71). Empirical finding: with a book that has text, fathers produced higher language quality, extended the topic more and issued more conversational turns, and described and labeled images less than with the wordless picture book. Paternal talk was differentially linked to child language (r = .45 to .78); the authors suggest an adjustment to the child’s abilities. Design inference: the book is not a neutral support; written text and a wordless picture book elicit distinct functions of extra-textual talk.

Sia et al. (2026) observe sixty-six dyads at home at 6, 12, 18 and 24 months, and the predictive value of prompt, evaluate, expand and repeat. Empirical finding: there is progression, with variability across parents. Prompt appears at six months and increases; evaluate, toward twelve months; expand and repeat, from eighteen months. Repeat at twelve and at eighteen months predicted language six months later; evaluate at eighteen months predicted outcomes at twenty-four and at thirty-six months; expand at eighteen and at twenty-four months also predicted language six months later. The abstract underscores feasibility before the age of two and the importance of the parental baseline. The child at the edge of the circle does not arrive blank: the child arrives with a history of turns that did, or did not, occur.

Bergström et al. (2024) test the feasibility of interactive shared reading inspired by dialogic reading, with techniques adapted to the preverbal level. Eleven parent–infant dyads took part in four modeling sessions every two months; between sessions, they read with those techniques. Bayley-III and CSBS DP ITC were used. Empirical finding: feasibility and linguistic development were favorable; not significant for cognition. Classical dialogic reading is designed for children older than two years; shared reading is recommended from the first year. Design inference: lowering the age is not bringing the same protocol forward; it is redesigning extra-textual talk for someone who does not yet produce sentences. Table 2 summarizes the home without comparing effects.

Table 2. Shared reading in the home. The contribution paraphrases the abstract, with no coefficients added.

6. Classroom findings: talk quality and books used

In the classroom the pattern is collective: one adult, one group, one picture book, a weekly rhythm. Cutler et al. (2026) observe thirty United States classrooms with higher language gains across the year and thirty with lower gains. They examine frequency, duration and grouping, and genre, sophisticated vocabulary, amount of input and syntactic complexity of the books. Empirical finding: there are similarities between the two groups, both in the activities and in the books. The abstract does not report that the lower-gain classrooms read less, or worse titles. Design inference: if what is observable in the episode does not distinguish those who gain more language, the difference has to be sought in extra-textual talk and in how it is sustained. Counting picture books or minutes is not enough.

Cao et al. (2026) enter through the educator’s interpretive stance. Thirty preschool educators in China took part in a professional-learning program centered on the contradictory relations between text and image in This Is Not My Hat, by Jon Klassen. Interviews and recorded readings were analyzed, before and after. Empirical finding: there were shifts in stance, more inferential language and a more openly dialogic construction of meaning, with more elaboration of what is left implicit and of alternative interpretations. The authors propose that the way educators read shapes the way they talk. Design inference: preparing dialogic reading is not only training questions; it is also preparing a literary reading of the picture book, because stance filters the turns.

Cipolletti et al. (2025) test a structured intervention on vocabulary, with a multiple probe across books. Children from two classrooms of three- and four-year-olds: twelve in the whole group and five, selected by the PELI language index and by an expressive measure of target vocabulary, with three weekly lessons in a small group (explicit instruction, representational gestures and practice). Empirical finding: the five in the small group increased vocabulary across four picture books, from baseline to intervention. The abstract concludes that preschoolers benefit from explicit instruction in a small group within story reading. It is not claimed that the whole group was sufficient, and no magnitude is invented. Design inference: the child who hardly speaks in the circle is not served by the same collective episode; grouping and explicitness are part of the quality of talk.

7. Interventions: vocabulary, narration and needs

Interventions turn the framework into a device. Özcan (2026) evaluates, in a systematic review, the effect of dialogic reading on narration in preschool. After searching Web of Science, EBSCO and ERIC, sixteen experimental and quasi-experimental studies entered (N > 1,800) with PICO-S criteria. Review finding, random-effects: a significant and positive overall effect, g = 0.62, 95% CI (0.44, 0.80). Larger for macrostructure (g = 0.82) and for causal language (g = 0.78). Digital and robotic modalities, g = 0.88, with caution because of the limited number of studies. This review does not meta-analyze those sixteen studies again. It takes the external finding: when there is a dialogic intervention and narration is measured, effects are documented. It does not authorize reading Friday straight through as if that were the intervention.

Riad et al. (2025) contribute a pragmatic randomized trial, with a switching-replications design, in Swedish preschools. Eighty-five children from diverse linguistic, socioeconomic and cultural backgrounds, in nine preschools in three districts. Ten teachers, trained in dialogic reading and coached by three special-education teachers, led the small group. Empirical finding: the intervention was feasible and the children progressed in oral language. The abstract does not report an effect size and does not break results down by home language. It does locate learners of an additional language as a population at risk in the majority language, with differences already early. Design inference: feasibility in an ordinary classroom, with coaching, is part of fidelity. A protocol that does not fit the school day is not dialogic reading.

Sundström et al. (2024) combine shared reading and the Lexicon Pirate method in a small group, with multilingual children aged six to seven, in preschool class in low-socioeconomic-status areas in Sweden, with speech, language and communication needs. Non-randomized pretest–posttest: eighteen children, nine in the intervention and nine in the control; twelve sessions in six weeks. Empirical finding: significant gains in trained and untrained words relative to the control. There were no effects on semantic fluency or on comprehension of semantic relations. Design inference: the small group can move vocabulary without moving every semantic domain.

Three studies are retained only for the object of the title. Ghosh et al. (2026) study the impact of a dialogic-reading intervention on young dual language learners; neither magnitude nor direction is attributed to them. Ergenekon and Işıkoğlu (2026) examine dialogic-reading interventions with mathematics picture books; an effect is not asserted. Gonzalez et al. (2024) investigate a shared reading rich in science and social-studies content and its bearing on the vocabulary of preschool dual language learners; no numerical result is extracted either. Inference bounded to the object: the field does not reduce the practice to the literary story or to monolingualism. Those lines are not evaluated here for lack of an abstract; nor are they erased from the map.

8. Training and fidelity: what sustains practice

The vignette does not fail for lack of a picture book: it fails because nothing sustains the dialogue once the book is open. Dillon and Newman (2023) formulate this as a problem of fidelity. In early childhood education they conducted two single-case studies. The first, an alternating-treatments design, examined a script and a checklist with four educators. Empirical finding: both supports increased fidelity equivalently; the educators preferred the checklist. The second, a multiple-baseline design, examined those supports and the increased fidelity in relation to emergent literacy in four children. Empirical finding: book-based vocabulary increased when fidelity was higher. The authors conclude that fidelity is best delivered with supports such as checklists, and that delivering the practice as it was designed can strengthen reading skills. Without fidelity, the name does not guarantee extra-textual talk. Design inference: preferring the checklist is choosing the support that the adult actually uses.

Vrinda et al. (2026) build a shared-reading module for Malayalam-speaking children aged three to five with language disorders, by means of a three-round Delphi with twenty speech-language pathologists. Empirical finding of consensus: in-person training; twelve sessions, twice a week; logs and video for fidelity; culturally relevant stories, with rich illustrations and minimal text. Target areas: print conventions, word and letter discrimination, rhyme and word awareness. The module integrates dialogic reading, print referencing and phonological awareness. Design inference: fidelity is designed with language, culture, frequency and a means of verification. A preschool that already does story time has not, for that reason, solved the module.

Kennedy and McLoughlin (2023) are retained only for the object of the title: a systematic review of the emergent literacy of English learners through dialogic reading. Neither effect sizes nor a number of studies are attributed to them. They signal that the synthesis for English language learners is an object of the field. This review does not turn that object into a thesis on bilingualism, and it does not fill in the missing abstract.

9. Conditions: bilingualism, DLD and clinical contexts

Conditions modulate who speaks, who waits and what counts as a gain. Karahan Tığrak and Özcebe (2026) integrate dialogic reading into Responsive Teaching, a caregiver-mediated intervention, with thirty children aged 48 to 72 months with developmental language disorder, randomly assigned to Responsive Teaching alone or to the combination. Receptive language, phonological working memory and emotional regulation were assessed at pretest, posttest and twelve-week follow-up. Empirical finding: the combined group improved more in all three domains. Working memory gained immediately; language and regulation, more so at follow-up. Design inference: for developmental language disorder, dialogic reading can be a multi-domain complement, not only a vocabulary workshop. The child at the edge of the circle may require a design that is not exhausted by the collective Friday.

Souto et al. (2025) insert retrieval practice. Thirteen children with developmental language disorder (mean age 59.13 months) and fourteen with typical development (57.07 months) learned eight novel words in two picture books; half under repeated spaced retrieval and half under repeated study. Empirical finding: both groups remembered forms and meanings better under retrieval. Recognition did not differ between conditions. The typical group outperformed the developmental-language-disorder group in recall and recognition of the form. Mild decline at one week; no interactions. The authors conclude that retrieval practice can benefit word learning. Design inference: extra-textual quality is also the way the word is asked for again. The monologue of the story omits that mechanism.

Sariyer Temelli and Beyza Irkli (2026) examine interactive shared reading and screens in the homes of children aged three to six with language delay (thirty-two) and with typical development (thirty). Sixty-two parents answered a forty-seven-item questionnaire, including the child–parent shared-reading activities scale. Empirical finding: parents of children with typical development reported more interactive shared reading. Higher parental schooling was associated with greater involvement. Daily screen exposure did not differ between groups; more than half of the children with delay were exposed for three or more hours a day. The authors read interactive reading as an important component of the environment, particularly for delay. Inference, with Rosslund et al. (2025): the screen does not replace the turn, and “we read at home” can coexist with high exposure and low interactivity.

Farran et al. (2025) open the neonatal intensive care unit. Of 338 articles, eight met eligibility (peer-reviewed, in English, infants in the neonatal intensive care unit, 2003–2023). Review finding: despite the modest number and the limited methodologies, there are benefits of shared reading for infants and caregivers during the stay. They recommend inserting it as standard practice. This review does not invent outcomes from those eight studies. The shared book can begin before preschool; even there, the object is the interaction, not the title.

Yuan et al. (2026) are read as a brief contrast, not as a thesis. One hundred forty-four typical readers and one hundred forty-two with difficulties, from Chinese kindergartens, were assigned to an autonomy style, a controlling style, or control, in shared reading mediated by a chatbot. Empirical finding: instructional style had a different impact according to competence; typical readers improved more with autonomy, and those with difficulties, with a controlling style. Bounded inference: the style of the episode matters and is not the same for everyone. The contrast is not converted into a thesis on generative artificial intelligence, and chatbots are not recommended in circle time. It is retained because the human corpus already shows the same point: without quality of the turn, the book does not decide the outcome.

10. Discussion: from shared book to sustained dialogue

The corpus does not prove a single mechanism. It converges on a category distinction. There is a shared book when there is co-presence and a text or a picture book. There is dialogic reading when, in addition, there is extra-textual talk of quality, continuity across contexts and fidelity to the practice one claims to be doing. Figure 1 is not a datum: it is the scheme with which the twenty-four studies are read.

Cutler et al. (2026) support the first row: what is visible in the episode may not distinguish classrooms. Read et al. (2025) and Sia et al. (2026) support the second: extra-textual talk is measured and, before the age of two, already predicts. Rosslund et al. (2025), Yang et al. (2025), Conica et al. (2023), Özen-Uyar and Yigit-Gencten (2026) and Sariyer Temelli and Beyza Irkli (2026) support the third: the home is not an echo of the classroom. Dillon and Newman (2023) and Vrinda et al. (2026) support the fourth: fidelity rises with supports and is designed by consensus. Özcan (2026), Cipolletti et al. (2025), Riad et al. (2025), Sundström et al. (2024), Karahan Tığrak and Özcebe (2026) and Souto et al. (2025) support the fifth when there is an intervention and language is measured; Bergström et al. (2024) and Farran et al. (2025) extend it to the infant. Cao et al. (2026): the reading stance filters talk. Yuan et al. (2026): instructional style is not neutral.

Design inference, not a combined finding: a preschool that selects titles and does not design the turn, family continuity and fidelity remains inside the monologue. Ghosh et al. (2026), Gonzalez et al. (2024), Ergenekon and Işıkoğlu (2026) and Kennedy and McLoughlin (2023) mark objects—dual language learners, disciplinary content, the mathematics picture book, English language learners—that the scheme must house without invented results. The book matters; it is not enough. Designing dialogic reading is designing conditions for conversation: what is asked, how long one waits, how one expands, who else in the family does it, how one checks that the adult has not returned to reading straight through, and how the cycle is adjusted when there is developmental language disorder, two languages, or not yet any sentences.

11. Limitations

The search is Crossref only. Scopus, Web of Science and ERIC were not consulted as databases in their own right, except when someone else’s abstract—that of Özcan (2026)—declares having done so. Screening was manual, by a single reader, of title and abstract. There is no double-blind procedure, no agreement between reviewers, and no meta-analysis of the effects of this corpus. The figures in Table 1 are not prevalence.

Four sources lack an abstract. Ghosh et al. (2026), Ergenekon and Işıkoğlu (2026), Gonzalez et al. (2024) and Kennedy and McLoughlin (2023) are cited by the title; attributing results to them would be to invent. The geographic bias is evident: Norway, Sweden, China, Türkiye, India, Ireland, the United States, and neonatal intensive care studies in English. There is little direct Mexican evidence; none is fabricated. The vignette is composite, not a field case. Primary samples range from eleven dyads (Bergström et al., 2024) to 1,442 infants (Rosslund et al., 2025); they are not weighted. The g of Özcan (2026) is not the effect of our twenty-four studies. Yuan et al. (2026) were used as a contrast of instructional style, not as an evaluation of conversational systems in the preschool.

12. Conclusions

Sharing a book with a child or with a group is not, by itself, dialogic reading. The 2021–2026 corpus allows that claim without rejecting the story. There is an association between shared reading and vocabulary in infants (Rosslund et al., 2025), early trajectories that predict language (Sia et al., 2026), preverbal feasibility (Bergström et al., 2024) and benefits reported in the neonatal intensive care unit (Farran et al., 2025). There are classrooms that resemble one another in books and minutes and not in gains (Cutler et al., 2026), homes that report reading and accumulate screen time (Sariyer Temelli and Beyza Irkli, 2026), and practices that move vocabulary only when fidelity rises (Dillon and Newman, 2023). The object and the practice do not coincide.

The documented gains—vocabulary in a small group (Cipolletti et al., 2025; Sundström et al., 2024), narration (Özcan, 2026), oral language (Riad et al., 2025), multiple domains in developmental language disorder (Karahan Tığrak and Özcebe, 2026), recall with retrieval (Souto et al., 2025), receptive vocabulary linked to idioms (Yang et al., 2025), paternal talk according to book type (Conica et al., 2023)—appear when there is extra-textual talk of quality, or when an intervention installs it and measures it. Cao et al. (2026) remind us that such talk depends on how the adult interprets the picture book. Read et al. (2025) offer the analytical vocabulary so that measurement is not improvised.

Continuity and fidelity are not later chapters. Özen-Uyar and Yigit-Gencten (2026) show that involving families moves language and reading behaviors, with gaps concerning fathers, special needs and infants. Vrinda et al. (2026) show that a module for language disorder is built with frequency, language, culture and verification. Kennedy and McLoughlin (2023), Ghosh et al. (2026), Gonzalez et al. (2024) and Ergenekon and Işıkoğlu (2026) remind us, even if only by the title, that English language learners, dual language learners and content picture books are not extras. Yuan et al. (2026) remind us that the style of the episode is not indifferent to the child’s competence.

Designing dialogic reading is designing conditions for conversation around the book. It is not filling the shelf, fulfilling Friday, accepting “we read at home” as evidence, or naming the practice without verifying it. It is opening turns, waiting, expanding, articulating the circle and the home, sustaining fidelity, and adjusting the cycle when the interlocutor does not yet speak, speaks in more than one language, or presents a developmental language disorder. The story may have been read. The dialogue has not. The difference is not one of title: it is one of the architecture of conversation.

NEXTECH.IA Editorial Lab / Ingeniero Mitre

References

1. Bergström, E., Bratt, A. S., & Svensson, I. (2024). An interactive shared reading intervention designed for preverbal infants: A feasibility study exploring early language and cognitive development. Child Language Teaching and Therapy, 40(3), 222–237. https://doi.org/10.1177/02656590241250234
2. Cao, S., Zou, Y., Han, F., Cai, Z., & Hu, J. (2026). Educators’ changing interpretive stance and meaning-making language during shared reading of a postmodern picture book. Language and Education. https://doi.org/10.1080/09500782.2026.2730419
3. Cipolletti, L. B., Murdoch, A., & Keelor, J. (2025). Impact of a Structured Shared Book Reading Intervention on the Vocabulary Knowledge of Preschool-Age Children: An Exploratory Study. Early Childhood Education Journal, 53(8), 3133–3147. https://doi.org/10.1007/s10643-025-01920-z
4. Conica, M., Kelly, L., Nixon, E., & Quigley, J. (2023). Father and Toddler Language During Shared Book Reading with Text‐Based and Wordless Picture Books. Reading Research Quarterly, 58(4), 655–667. https://doi.org/10.1002/rrq.501
5. Cutler, L., Gabas, C., Piasta, S. B., Kim, H., & Schachter, R. E. (2026). Characteristics of Shared Book Reading Activities and Books Used During Shared Reading in Higher- and Lower-Language-Gains Classrooms. Journal of Research in Childhood Education. https://doi.org/10.1080/02568543.2026.2613131
6. Dillon, C., & Newman, D. (2023). Supporting Intervention Fidelity of Dialogic Reading to Support Preschool Children’s Early Literacy Skills. Perspectives on Early Childhood Psychology and Education, 7(2). https://doi.org/10.58948/2834-8257.1029
7. Ergenekon, E., & Işıkoğlu, N. (2026). Examining the impact of dialogic reading interventions with math picture books. Cognitive Development, 78, 101685. https://doi.org/10.1016/j.cogdev.2026.101685
8. Farran, L. K., Leslie, S. L., & Brasher, S. N. (2025). Promoting language and literacy through shared book reading in the NICU: A scoping review. PLOS ONE, 20(3), e0318690. https://doi.org/10.1371/journal.pone.0318690
9. Ghosh, E., Banerjee, R., James, C., & Malala, D. (2026). Transforming Language and Literacy: The Impact of Dialogic Reading Intervention on Young Dual Language Learners. Early Childhood Education Journal. https://doi.org/10.1007/s10643-026-02333-2
10. Gonzalez, J. E., Kim, H., Anderson, J., & Pollard-Durodola, S. (2024). The Effects of a Science and Social Studies Content Rich Shared Reading Intervention on the Vocabulary Learning of Preschool Dual Language Learners. Early Childhood Research Quarterly, 66, 34–47. https://doi.org/10.1016/j.ecresq.2023.08.011
11. Karahan Tığrak, T., & Özcebe, E. (2026). Dialogic Reading as a Complementary Intervention for Children with Developmental Language Disorder: Multi-Domain Outcomes. Reading & Writing Quarterly. https://doi.org/10.1080/10573569.2026.2669638
12. Kennedy, C., & McLoughlin, A. (2023). Developing the Emergent Literacy Skills of English Language Learners Through Dialogic Reading: A Systematic Review. Early Childhood Education Journal, 51(2), 317–332. https://doi.org/10.1007/s10643-021-01291-1
13. Özcan, S. (2026). The effect of dialogic reading interventions on narrative skills in the preschool period: a systematic review. Frontiers in Education, 11, Article 1817388. https://doi.org/10.3389/feduc.2026.1817388
14. Özen-Uyar, R., & Yigit-Gencten, V. (2026). Family-engaged literacy practices in early childhood: a scoping review of shared book reading studies in Türkiye. European Early Childhood Education Research Journal. https://doi.org/10.1080/1350293x.2026.2676843
15. Read, K., Rabinowitz, S., & Harrison, H. (2025). It’s the talk that counts: A review of how the extra-textual talk of caregivers during shared book reading with young children has been categorized and measured. Journal of Early Childhood Literacy, 25(3), 801–837. https://doi.org/10.1177/14687984231202968
16. Riad, R., Westling Allodi, M., Siljehag, E., & Bölte, S. (2025). Dialogic reading in preschool: a pragmatic randomized trial enrolling additional language learners. Scandinavian Journal of Educational Research, 69(4), 772–789. https://doi.org/10.1080/00313831.2024.2348473
17. Rosslund, A., Kartushina, N., & Mayor, J. (2025). Associations between shared book reading, daily screen time and infants’ vocabulary size. Journal of Child Language, 52(4), 971–983. https://doi.org/10.1017/s0305000924000291
18. Sariyer Temelli, M. N., & Beyza Irkli, E. (2026). Interactive shared reading as a protective factor in children with language delay: beyond screen exposure. Early Child Development and Care. https://doi.org/10.1080/03004430.2026.2717248
19. Sia, M. Y., Tzeng, O. J. L., & Wang, S. (2026). Dialogic Reading before Age Two: Developmental Trajectories and Links to Language Skills. Early Childhood Education Journal, 54(7), 5843–5863. https://doi.org/10.1007/s10643-026-02263-z
20. Souto, S., Leonard, L. B., Deevy, P., Christ, S. L., Karpicke, J. D., & Schroeder, M. L. (2025). Word Learning in Children With Developmental Language Disorder: The Use of Retrieval Practice During Shared Book Reading. Journal of Speech, Language, and Hearing Research, 68(7), 3305–3321. https://doi.org/10.1044/2025_jslhr-24-00809
21. Sundström, S., Höglund, M., Sällberg, R., & Walsö, H. (2024). Small-group vocabulary intervention combining elements from shared book Reading and the Lexicon pirate method – A pilot study of multilingual children with speech, language and communication needs. Child Language Teaching and Therapy, 40(2), 197–212. https://doi.org/10.1177/02656590231215955
22. Vrinda, R., S, S., Karuppali, S., Venkatesh, L., & Kunnath, S. K. (2026). Development of a shared book reading intervention module for children with language disorder through the Delphi consensus method. Child Language Teaching and Therapy. https://doi.org/10.1177/02656590261490797
23. Yang, J., Grøver, V., & Lawrence, J. F. (2025). Chinese mothers use idioms in shared book reading: A predictor for children’s Chinese vocabulary growth? Journal of Child Language, 52(4), 918–944. https://doi.org/10.1017/s0305000924000266
24. Yuan, C., Dong, Y., Jin, R., Mo, J., & Zheng, H. (2026). Shared book reading with AI chatbots: comparison of the effects of instructional style on children’s language abilities between typical and poor readers. Language and Education, 40(1), 278–299. https://doi.org/10.1080/09500782.2025.2501081