Does Language Training Reshape Children’s Spontaneous Narratives?

This study examined whether a structured language training program leads to measurable changes in children’s narrative production. Narrative samples were analyzed under quantitative design using linguistic inquiry and word count (LIWC) categories to capture stylistic, cognitive, and content-related features of language use. A total of forty-seven children were assigned to Training (n = 36) and Control (n = 11) groups. All participants produced pre- and post-narratives, elicited through a picture-based storytelling task. Between assessments, the Training group completed a structured computer‑based language program at Glitr application, while the Control group engaged in pencil‑and‑paper tasks and standard educational activities. Analyses were conducted across ninety-eight LIWC categories using two complementary approaches: (a) baseline-adjusted ANCOVA (post ~ group + pre) and (b) permutation tests on individual change scores (Δ = POST − PRE), with false discovery rate (FDR) control for multiple comparisons. No statistically reliable group differences were observed after FDR correction across the analyzed categories. This pattern was consistent across both analytical approaches. Exploratory analyses, however, revealed coherent directional tendencies in selected function-word and cognitive-process categories, including relatively greater use of categorical language and attenuated changes in auxiliary verbs, tentative expressions, and cognitive-process markers in the Training group. The findings suggest that structured language training does not produce large, readily detectable changes in LIWC-derived indicators of children’s narratives under the present conditions. Instead, potential effects may be subtle and distributed across multiple linguistic dimensions. These results highlight the challenges of detecting generalization from structured tasks to spontaneous language use and underscore the importance of sensitive, theory-aligned measurement approaches.