Building upon the established moderate effect of interactive reading on narrative development, this investigation extends to examine age-specific manifestations across three ontogenetic stages (3–4 years, 4–5 years, 5–6 years). As delineated in Table 2, significant enhancements in narrative competence emerge universally (aggregate g > 0, p < 0.01), confirming developmental pervasiveness. Crucially, the 4–5 years cohort exhibits maximal effect magnitude (g = 0.635, p < 0.001), signifying a critical ontogenetic window for intervention efficacy.

Table 3 demonstrates significant facilitative effects of all four pedagogical agents on children’s narrative competence (g > 0, p < 0.01). Effect magnitude stratification reveals peer interaction as the most potent predictor [g = 0.675, 95% CI (0.519, 0.830)], exceeding Cohen’s large-effect threshold (0.5–1.0). Subsequent predictors include parental engagement (g = 0.597) and researcher facilitation (g = 0.459), with teacher-guided interaction demonstrating modest efficacy (g = 0.164). This hierarchy suggests that peer-mediated linguistic scaffolding—characterized by reciprocal discourse patterns, observational learning mechanisms, and low-affect communication contexts—optimally catalyzes narrative development. While parental, researcher, and pedagogical inputs remain valuable, their differential efficacy necessitates strategically diversified interventions prioritizing peer dyads and family participation.

To investigate temporal dynamics, this study examined duration-dependent effects of interactive reading interventions. Table 4 reveals a monotonic dose–response relationship: effect magnitudes escalate with intervention length, peaking at ≥17 weeks [g = 0.644, 95% CI (0.485, 0.803)], followed by 9–16 weeks [g = 0.484, 95% CI (0.369, 0.599)], and ≤8 weeks [g = 0.267, 95% CI (0.120, 0.413)]. Critically: (1) All durations exceed Cohen’s small-effect threshold (g > 0.20), confirming universal efficacy. (2) ≥17-week interventions reach large-effect magnitude (g > 0.50). (3) Significant between-group gradients exist (Δg max-min = 0.377, p < 0.001). These findings establish temporal accumulation as a key efficacy modulator, where extended implementation generates cascading gains in narrative competence through sustained linguistic scaffolding.

The results of this study show that there are significant differences in the influence of interactive reading on the development of narrative ability of children at different ages. Compared with 3–4 years old and 5–6 years old, in 4–5 years old, interactive reading has a significant effect on the development of narrative ability of children, and the greatest impact (g = 0.635, p < 0.01). This research result is consistent with the research results (Wu, 2014), interactive sharing reading intervention significantly improves the reading interest and narrative ability of 4–6 year old children. The same results were obtained in the study of Huang (2020), who found that after interactive picture book sharing reading activities, children aged 4–5 have a rich vocabulary and their vocabulary utilization level has been greatly improved, which indicates that picture book reading activities, as one of the interactive reading methods, can significantly improve the narrative ability of children aged 4–5. The above results can be explained in the following aspects:

First, Senechal et al. (2008) showed that interactive reading can promote children’s expressive vocabulary, morphology and syntax comprehension. The study of Huang (2020) found that the key stage of children’s language development, especially the 4–5 years old period, is regarded as the active period of rapid growth of vocabulary. However, vocabulary is the material of narration for children. Children with higher vocabulary level are more likely to use “building materials” to express their ideas with more words and confidence to express boldly, so they are more likely to carry out longer narration with richer content (Li C., 2014; Li J., 2014). Morgan and Meier (2008) proposed that at this critical stage, interactive reading builds a dynamic language environment through two-way dialogue between adults and young children, and has become an effective way to improve children’s narrative ability. During this process, adults use diverse vocabulary and abstract expressions. Through interactive communication and listening to feedback, they not only enhance children’s understanding and memory of vocabulary but also significantly increase their vocabulary. With the improvement of language skills, young children gradually construct narrative frameworks that are more complex in structure and more rigorous in logic, making the story content more rich and three-dimensional, and the expression of emotions more delicate and profound. In their research, German and Simon (1991) analyzed children’s word-finding skills in discourse and found that although children with word-finding disorder had no difference with normal children in language productivity, they showed more word-finding features in their narration. This indicates that the improvement of vocabulary ability can reduce the stuttering phenomenon in narration, making the narrative more smooth and the expression of emotions more delicate. Therefore, interactive reading, through its unique interactive mode, significantly promotes the rapid growth of vocabulary and the improvement of application ability of children aged 4–5, providing a solid language foundation and rich expression resources for the leap of children’s narrative ability.

Second, according to Piaget’s cognitive development theory, children aged 4 to 5 are at the end of the preoperational stage, and their cognitive development and social and emotional needs show a double leap: at this stage, children’s social needs surge, they start to build complex social skills, and they exhibit a strong desire for self-expression and interpersonal understanding. Interactive reading provides an excellent carrier for social learning during this critical period by creating in-depth dialogue scenarios between adults and children, and promotes the systematic development of children’s social interaction abilities in language interaction. For example, through story telling, young children can practice and learn how to communicate and collaborate with others in simulated social situations (Koivula et al., 2020; Kohm et al., 2016). In addition, interactive reading provides a platform for young children to express their thoughts and feelings through storytelling methods that can effectively support young children’s social–emotional reasoning, to meet their needs for social interaction and emotional development (Koivula et al., 2020). Specifically, on the one hand, social interaction provides a rich language and emotional communication environment for young children, which is crucial for the development of narrative skills. In social interaction, children learn how to organize language, express emotion and construct story framework through imitation, interaction and cooperation (Hebert-Myers et al., 2006). On the other hand, the satisfaction of emotional needs contributes to the development of children’s emotional regulation ability, which is an integral part of narrative skills. Good emotional regulation enables young children to better understand and express complex emotional states, which is essential for emotional depth and richness in narrative (Brinton and Fujiki, 2017). For example, through storytelling activities, young children learn not only how to express their own emotions, but also how to understand those of others, and this ability is an important foundation for the development of narrative skills (Betawi, 2014).

Third, the brains of 4 to 5 year olds are undergoing rapid development and reorganization, especially in brain regions associated with language processing. According to functional magnetic resonance imaging (fMRI) and other neuroimaging studies, developmental changes occur in the support networks of brain regions during the development of language skills in children from infancy to adulthood (Vannest et al., 2009). In particular, the brain’s ability to process language is developing rapidly during 4–6 years of age. Ríos-López et al. (2020) in their study recorded the development of neural oscillatory activity in response to language in children aged 4–5 years using functional magnetic resonance imaging (fMRI). This study found that from 4–6 years of age, children’s brains begin to process verbal information more efficiently, especially in processing slower-temporal components of speech, such as syllabic and prosodic information, and that delta and theta band activity in the right hemisphere is indirectly correlated with intelligibility of speech. This finding is also supported by Weiss et al. (2018) study, which examined brain processing of speech in children aged 5–6 years by functional magnetic resonance imaging (fMRI) and found that the brains of these children have shown specialization in both speech and semantic processing. This suggests that at 4–5 years of age, children’s brain language regions have begun to develop rapidly and specialize. Neural network maturation in language areas is shown by enhanced functional connectivity in children aged 4–5 years, while functional connectivity in infancy and early childhood is prospectively associated with language and basic literacy skills at 6.5 years of age (Yu et al., 2022). This suggests that the neural network connections formed in the early stage may lay a neural foundation for the subsequent development of language processing capabilities. Interactive reading may support the development of narrative comprehension by promoting effective connections between brain regions. Schmithorst et al. (2007) found that feedback networks in the brain involve efficient connections from Broca’s area and the medial side of the upper frontal lobe to the posterior part of the bilateral superior temporal gyrus when performing narrative processing tasks, and that this connectivity strengthens with age. This suggests that through interactive reading, connections between key areas of the brain responsible for language and narrative processing can be promoted, thereby supporting the development of narrative comprehension in children. Paranawithana et al. (2023) used functional near infrared spectroscopy (fNIRS) in their study to find that normal hearing infants had significantly enhanced functional connections between primary language areas during their first year of life. This enhanced functional connectivity helps improve the efficiency and complexity of language processing, which in turn supports more complex language tasks such as narrative.

Through a systematic review of 24 literatures, the research results show that there are significant differences in the effect of different interactive objects on promoting the development of children’s narrative ability. In interactive reading, teachers, peers, parents and researchers have different effects on children’s narrative ability. Among them, the influence of peers was the most significant (g = 0.675, p < 0.01). Then came parents (g = 0.597, p < 0.01), then researchers (g = 0.459, p < 0.01), and finally teachers (g = 0.164, p < 0.01). A large number of research evidences show that peer influence is more obvious in interactive reading’s impact on children’s narrative ability. Among them, Sang (2018) found in her study that peer sharing reading has a more significant impact on all dimensions of children’s narrative ability than independent reading. The results of this study indicate that peer sharing reading activities can be used as an effective method and means to cultivate the development of children’s narrative ability in kindergartens. Bokus (1992) found that compared with narration alone, children can introduce new reference content and operate on the partner’s text, such as confirmation and supplement, when they narrate with their partner. This suggests that peer interaction can enrich narrative content and structure and promote the development of narrative skills. Wei (2004) found that stories written in groups have advantages in terms of length, richness of detail, use of characters, and the use of higher levels of connectives and character representations compared to stories written individually. This suggests the potential advantages of peer cooperation in promoting young children’s narrative ability. The above results can be explained in the following aspects:

First, peer interaction is not limited to language communication, but also includes the understanding and expansion of the story content (Kim S. J., 2016; Kim Y. S. G., 2016). Research points out that peer relationships and interaction play an important role in bilingual children’s response to picture books, and these interactions affect their literary response. This suggests that peer involvement can enrich young children’s narrative experience and enable them to understand and construct stories from different perspectives.

Second, the naturalness and authenticity of social interaction. The social advantage of peer interaction is reflected in its natural characteristics of the interaction field: Children at the end of the preoperational stage can achieve stress-free self-expression in peer communication. This de-authoritative communication mode is more in line with the critical period needs of the development of their social interaction skills. This free expression helps young children to better develop narrative skills because they can explore and expand their storytelling without adult intervention (Laird et al., 1994).

Third, diversified modes of communication. Peer interaction creates diversified interactive scenarios covering dimensions such as dialogue and consultation, viewpoint confrontation, and role immersion These diverse modes of communication can help children learn how to organize language, construct plots, and express emotions, which are important components of narrative skills (Laird et al., 1994).

Fourthly, role switching and perspective switching in peer interactions help young children better understand and apply different narrative strategies. Bokus (1992) showed that in peer cooperation, children not only create stories together, but also complement and confirm each other’s narratives. This interactive process helps them learn how to view problems from different perspectives and use more complex connective words and role performance in narrative. This multi-perspective narrative training is essential for the development of children’s narrative skills.

Fifth, Peer interaction builds a unique social support system, providing immediate feedback and emotional support for the narrative process. This immediate feedback mechanism prompts young children to continuously optimize narrative expression during interaction, significantly enhancing narrative quality and content depth (Wang and Cassell, 2003). In contrast, teachers and parents, while also able to provide support and feedback, are often more fixed and authoritative in their roles and may not be as natural and varied as feedback in peer interactions.

Furthermore, when interpreting the universality of the conclusions of this study, it is necessary to take into account the specific context in which they are based. The vast majority of the empirical studies included in this review were conducted in school Settings. This distribution feature provides particularly solid evidence to support our finding that peer interaction in interactive reading can effectively drive the development of narrative ability in young children in educational contexts. However, this also means that the direct evidence that this study can provide regarding the role of peer interaction in the family (usually dominated by parents), libraries or other informal educational Settings is relatively limited. Due to the insufficient number of studies in these contexts, we were unable to conduct effective subgroup analyses to examine the moderating effects of the “context” factor. Therefore, the core conclusion of this study is first and most directly applicable to school education practice. We call for future research to be extended to more diverse social and cultural contexts to verify the robustness of current findings and to deeply reveal the possible contextual specificity of interactive reading mechanisms. This will help provide more targeted theoretical guidance for peer interaction reading intervention in different scenarios.

Studies show that the duration of interactive reading is significantly positively correlated with the development of children’s narrative ability. With the extension of the intervention time, the amount and complexity of language input that children are exposed to gradually increase. This progressive stimulation can not only strengthen the language foundation, but also promote their understanding and application of complex narrative structures. This finding is consistent with the findings of Mol et al. (2008), which showed that even short-term interactive reading interventions can have a positive impact on young children’s language and narrative skills. Peterson et al. (1999) also mentioned in their study that long-term interactive reading, such as 1 year of intervention, can significantly improve children’s vocabulary and narrative skills. The above results can be explained in the following aspects.

First, in the early 1980s, Krashen (1981) proposed the language Input Hypothesis as the core theoretical framework in his language monitoring model, advocating that language acquisition requires understanding understandable inputs slightly higher than the existing level (the i + 1 model). This theory emphasizes that effective input needs to meet two conditions: one is sufficient language exposure, and the other is that the input content needs to be understandable and contain an i + 1 gradient structure.

According to Krashen’s language monitoring model and its core input hypothesis, the positive correlation between the duration of interactive reading intervention and the improvement of children’s narrative ability can be reasonably explained. This theory emphasizes that learners need to achieve language acquisition by being exposed to “i + 1” comprehensible inputs that are slightly higher than the existing level. In continuous intervention, young children are repeatedly exposed to texts that integrate known language (i) with new language elements (+1). This progressive input not only consolidly consoles basic abilities but also stimulates their interest in exploring new vocabulary, grammar and expressions. Especially for the development of narrative ability, the “i + 1” input is crucial because it involves the integration of language forms with plot events and emotional evaluations, as well as the ability to tell stories appropriately according to task requirements. With the increase of intervention duration, the narratives spontaneously generated by young children in natural conversations show significant improvements in terms of length, complexity and diversity (Umiker-Sebeok, 1979). Studies show that extending the duration of interactive reading intervention can significantly promote the development of narrative ability in young children, which is highly consistent with the core mechanism of Krashen’s language input hypothesis. According to the “i + 1” theory, continuous intervention provides a progressive language input environment for young children: by extending the exposure time, children constantly obtain compound texts containing known language (i) and new language elements (+1). This moderately challenging input not only consolidly consoles the existing language foundation but also continuously stimulates the desire to explore new vocabulary and grammatical structures. In the intervention practice, the dual guarantee mechanism of input quantity and quality has been effectively operated: With the extension of the intervention period, the amount of children’s language exposure increases exponentially, and by designing strategies such as narrative dialogues involving mothers, open-ended questions, and extensible feedback, it is ensured that the input always maintains an “i + 1” gradient (Sénéchal, 1997). Studies show that continuously strengthening language exposure and supplementing it with systematic narrative guidance can significantly enhance children’s dual abilities in language expression and narrative construction.

Second, educational theory points out that the cognitive development of young children is often based on the dual foundation of repeated practice and systematic training. Long periods of interactive reading also help establish a child’s narrative framework and structure. As stated by Sénéchal (1997), through repeated reading and questioning, young children are able to better understand and remember new vocabulary, which is very helpful for building complex narrative structures. During the continuous process of parent–child reading together, children gradually deepen their cognitive construction of the plot thread, character relationships and narrative logic through repeated exposure to texts on the same theme or various types of story carriers.

This can be argued for in pedagogy’s “progressive learning theory,” which is a training process that allows for continuous learning from input streams and growth over time, while retaining previously acquired knowledge (Karn et al., 2021). This tells us that learning should constantly contact relevant contents to consolidate achievements. Vretudaki et al. (2023) found in their study that through repeated reading and questioning, young children are able to better understand story structure and make more in-depth comments based on it. Schetz et al. (2000) also found in their study that interactive book reading and story retelling methods in small groups had positive effects on story comprehension and narrative skill development compared to large groups. Horst et al. (2011) found that 3-year-olds who read the same storybook three times in a week showed that children in the repeat reading group performed very accurately on both immediate recall and long-term memory tasks, whereas children who listened to different stories performed accurately only on immediate recall during the last two sessions. This means that repeated reading can help young children remember new vocabulary better and help them form complex narrative structures.

Third, interactive reading stimulates brain development through continuous language interaction, with a focus on promoting the functional improvement of the Broca area (responsible for language generation) and the Wernicke area (responsible for language comprehension). These two brain regions are connected through neural pathways such as the arboid bundle/superior longitudinal bundle, and there are significant differences in the maturity of their connections between the children and adult groups. And the efficient connection of these two regions is essential for language processing (Schmithorst et al., 2007). Scheinost et al. (2022) in his study showed that functional connectivity between Broca’s area and Wernicke’s area develops with age, which supports developmental changes in the brain in language processing. Schmithorst and Holland (2006) also confirmed in their study that effective connections between Broca’s area and Wernicke’s area increase with age, and Romeo et al. (2018) also found that children who had more conversation rounds with adults had stronger activation in Broca’s area. This enhanced connection contributes to improved narrative comprehension and expression (Schmithorst and Holland, 2006).