Participants were randomized one-to-one into a cognitive-poetics-based narrative intervention or an active literary-reading control, with both conditions running three 40-minute sessions per week over 12 weeks, for a total of 36 sessions. Assessments took place at baseline, at week 6, at week 12 immediately after the program ended, and again four weeks later. The design deliberately avoided the classic weakness of reading studies: instead of comparing reading with nothing, both groups received identical exposure to British and American coming-of-age fiction, matched instructor contact, and matched session length.

The critical difference lay in the processing structure. In the intervention condition, children followed a fixed session sequence: four minutes of orientation to the passage, ten minutes of guided reading, six minutes identifying characters' goals, beliefs, and emotional shifts, eight minutes comparing alternative character perspectives, eight minutes generating alternative interpretations or cognitive reappraisals of difficult events, and four minutes of closing reflection. Prompts included questions such as what a character believed was happening, how the same event might look from another character's perspective, what new information could change an initial interpretation, and how a setback might be understood differently.

The active control, by contrast, spent its sessions on conventional reading comprehension, vocabulary, plot sequencing, and factual discussion, without any perspective-shifting or reappraisal exercises. The literary corpus was selected with unusual care. A five-member panel of literature-education specialists, a developmental psychologist, a literacy teacher, and a school counselor screened 18 candidate British and American novels before allocation, excluding fourteen for unsuitable reading levels, overly mature themes, structural problems with excerpt-based teaching, or excessive cultural background demands.

Four works survived: Wonder by R. J. Palacio, The Boy at the Back of the Class by Onjali Q. Raúf, Matilda by Roald Dahl, and Holes by Louis Sachar. Each was mapped to one of four thematic units covering self-discovery and identity, relationships and perspective understanding, setbacks and emotional adjustment, and growth and meaning reconstruction. Because participants read standardized simplified-Chinese translations of selected excerpts rather than English originals, variation in English proficiency could not determine comprehension, and both groups read the same translated passage in a given session whenever possible.

Delivery fidelity was tightly monitored. Four instructors with postgraduate training and at least three years of experience with school-age children completed an eight-hour standardized preparation program and two supervised mock sessions, each required to reach at least 85 percent adherence on a fidelity checklist before delivery. Independent observers rated 18 of the 72 condition-sessions, finding average intervention adherence of 92.8 percent and average control-condition differentiation of 95.2 percent, with inter-rater agreement of ICC (2, k) = 0.91.

Attendance was high and nearly identical across arms: 91.9 percent of sessions in the intervention group and 91.1 percent in the control. No serious adverse events were reported, and children's responses to the material were monitored for discomfort throughout. The outcome measures were correspondingly rigorous.

Executive function was assessed with computerized tasks targeting its three core components: an arrow Flanker task for inhibitory control, a backward digit-span task for working memory, and a color-shape task-switching paradigm for cognitive flexibility, each with split-half reliabilities around 0.81 to 0.86. Psychological adjustment was measured with the 47-item Revised Child Anxiety and Depression Scale, the parent-and-teacher Strengths and Difficulties Questionnaire, and the Emotion Regulation Questionnaire for Children and Adolescents. Narrative transportation, the cognitive, emotional, and imagery-based absorption in a story world, was measured at weeks 6 and 12 as a candidate mechanism, alongside a scenario-based performance task in which children generated alternative interpretations of age-relevant vignettes about academic setbacks, peer exclusion, and criticism.

The results showed clear group-by-time interactions. The executive-function composite differed across groups over time with F = 18.64, p < 0.001, and a partial eta-squared of 0.095. The intervention group's composite mean rose from 0.02 at baseline to 0.51 at post-intervention, while the control group moved from -0.01 to only 0.18.

The unadjusted between-group standardized mean difference at week 12 was d = 0.38 (95 percent CI 0.07 to 0.68). Psychological adjustment moved in the same favorable direction: the RCADS total score showed a group-by-time F = 15.72 with a post-intervention difference of d = -0.82, and the SDQ difficulties score showed F = 12.48 with d = -0.65, both favoring the intervention group. Cognitive reappraisal showed the largest contrast of all, at d = 0.81.

All three component executive-function domains, inhibitory control, working memory, and flexibility, moved in the same broad direction as the composite. Using a temporally ordered structural equation model estimated in Mplus with 5,000 bootstrap resamples, the researchers tested the chain intervention assignment to week-6 narrative transportation to week-12 cognitive reappraisal to week-16 outcomes, controlling baseline reappraisal and the corresponding baseline outcome. Intervention assignment predicted higher transportation (standardized beta = 0.42), transportation predicted higher reappraisal (beta = 0.36), and reappraisal in turn predicted better week-16 executive function (beta = 0.28) and lower psychological difficulties (beta = -0.31).

The total indirect effects through the combined transportation and reappraisal routes were 0.14 for executive function and -0.18 for psychological difficulties, with confidence intervals excluding zero. A residual direct intervention effect remained for both outcomes, meaning the proposed mechanisms accounted for only part of the overall relation. Model fit indices, including CFI of 0.979 and RMSEA of 0.043, indicated acceptable-to-good global fit.

The authors are careful about what these numbers do and do not show. The four-week follow-up captured partial short-term persistence: the executive-function gap narrowed by week 16, and its unadjusted confidence interval included zero, so the researchers describe the result as short-term persistence rather than a sustained developmental benefit. The psychological-adjustment differences remained visible at follow-up, with RCADS and SDQ contrasts of d = -0.60 and d = -0.48 respectively.

Because the mediators were not randomized, the indirect-effect model remains an associational test rather than proof of causal mediation, and unmeasured processes such as greater engagement, richer peer discussion, or expectancy effects could have contributed. Participants and instructors knew which activities they received, leaving performance and expectancy effects possible even though outcome assessors were blinded. The study was also not preregistered, so its reporting hierarchy reflects the analysis actually conducted rather than a prespecified confirmatory plan.

Several further limitations frame the findings. The sample came from specific recruitment settings, and replication across regions, socioeconomic backgrounds, languages, and educational systems is needed before broader conclusions. The use of translated British and American texts, while controlling for English proficiency, may have altered character voice and emotional nuance in ways that matter for perspective work.

Parent-teacher agreement on the SDQ was moderate, informant-specific perspectives remain relevant, and repeated executive testing may have produced practice effects in both groups, although the active control makes a simple retest explanation less likely. Instructor and session clustering was not modeled. The authors emphasize that the intervention was educational and developmental, not a clinical treatment, and that the RCADS scores in this non-clinical sample indicate symptom severity rather than diagnosis.

Even with those caveats, the practical implication is concrete and narrower than turning literature lessons into therapy. In ordinary reading instruction, children can be asked not only what happened in a story but to defend competing interpretations, compare viewpoints, trace shifts in a character's goals or beliefs, and propose more than one response to a setback. Those activities remain squarely within literacy teaching while engaging the cognitive-control and emotion-regulation processes that are still maturing between ages 10 and 12.

The study's own comparison with a 2026 review of school-based creative bibliotherapy, which found encouraging but methodologically uneven results across 23 studies, underscores the point: favorable programs repeatedly used suitable text selection, structured interaction, and trained delivery. Whether the cognitive-poetics approach produces durable developmental gains, and which of its bundled components carries the effect, remain open questions that longer follow-up and component-manipulation designs will need to answer.

Subject of Research: Effects of a cognitive-poetics-based narrative reading intervention on executive function and psychological adjustment in children aged 10-12

Article Title: Effects of a cognitive-poetics-based narrative intervention using British and American coming-of-age novels on children's executive function and psychological adjustment: a randomized controlled study

Article References: Li, S., & Xie, S. (2026). Effects of a cognitive-poetics-based narrative intervention using British and American coming-of-age novels on children's executive function and psychological adjustment: a randomized controlled study. Frontiers in Psychology, 17, Article 1952178. https://doi.org/10.3389/fpsyg.2026.1952178

Image Credits: AI Generated

DOI: 10.3389/fpsyg.2026.1952178

Keywords: executive function, cognitive poetics, narrative transportation, cognitive reappraisal, coming-of-age fiction, randomized controlled trial, children, psychological adjustment, bibliotherapy, developmental psychology, reading intervention, emotion regulation