Development and preliminary evaluation of a ciliary muscle-oriented motor imagery script for primary school children: a school-based cluster randomized controlled trial.

Background: Myopia in children is occurring earlier and progressing faster, while existing interventions such as outdoor activity and device-based visual training are often difficult to implement consistently in school settings. Motor imagery (MI), as a low-burden cognitive approach, may offer a complementary pathway for visual-function support. However, ciliary muscle-oriented MI scripts specifically designed for children remain underdeveloped. Purpose: This study aimed to develop a ciliary muscle-oriented MI script based on the PETTLEP model for primary school children and to compare two training arrangements: G1 (staged imagery group: 8 weeks of general imagery followed by 8 weeks of specific imagery; Methods: A school-based cluster randomized design was used. A total of 156 children aged 10-11 years from one primary school in Suzhou were assigned by intact class to either G1 or G2. Visual outcomes, including uncorrected distance visual acuity (UDVA), kinetic visual acuity (KVA), accommodative amplitude (AMP), and binocular accommodative facility (BAF), were assessed at baseline (T1) and post-intervention (T3). Imagery ability was assessed using the Sport Imagery Questionnaire (SIQ; cognitive-general imagery [CG] and cognitive-specific imagery [CS] dimensions) at T1, T2, and T3. Results: Both groups improved over time on several outcomes. No between-group difference was found for UDVA at T3. After Benjamini-Hochberg correction, KVA was the only visual outcome showing a significant adjusted between-group difference at T3; BAF was not significant at T3, and AMP showed only a borderline trend. Significant Time × Group interactions were found for KVA, BAF, and CG. G1 scored higher than G2 on CG at both T2 and T3. In CS, both groups improved over time; G2 scored higher at T2, but no between-group difference was found at T3. However, because the study included only four classes in total (two per condition) and the class-level ICC for post-intervention KVA was 0.366, the KVA finding should be interpreted as a hypothesis-generating class-level observation rather than a reliable training effect. Conclusion: This study developed and preliminarily evaluated a ciliary muscle-oriented MI script under structured school-based conditions. The clearest between-group evidence was observed for CG. Different imagery arrangements may be associated with different developmental patterns, but visual-function findings, particularly KVA, require cautious interpretation and replication in studies with more clusters and stronger control conditions.