Cleft palate is one of the most common maxillofacial congenital malformations, which results in severe speech disorders. Compensatory articulation disorder, also known as non-oral articulation disorder (NOA), is considered as the major pathological change among these patients. However, the outcome of speech therapy, an important treatment method, for NOA is often unsatisfactory. This is attributed to the erroneous articulation patterns and entrenched habits in patients with NOA, which require considerable training intensity and time. According to preliminary results from the investigators' own study, as well as studies by others, structural and functional changes have been clearly identified in some brain regions of patients with NOA, suggesting that abnormal neural networks are involved in the progression of NOA. Thus, the investigators proposed the hypothesis that speech therapy effectively corrects articulation disorders through reconfiguration of pathological neural function and reorganization of the abnormal neural network involved in NOA. In this study, multimodal brain imaging techniques will be applied to investigate differences in brain functional connectivity and structural connectivity networks among groups with oral articulation (OA), varying degrees of NOA in postoperative cleft palate patients, and healthy controls. The relationship between improvement in speech intelligibility and alterations in brain networks before and after intervention will be compared. This study aims to reveal the neural network substrates associated with NOA and speech therapy. Overall, through this comprehensive study, the investigators aim not only to provide new insight into the underlying neural mechanism of NOA but also to accumulate evidence for improving the efficacy of speech therapy and discovering new therapeutic strategies in clinical practice.

Cleft palate often leads to compensatory articulation disorder (Non-oral Articulation, NOA), which is resistant to conventional speech therapy. This study is grounded in the hypothesis that the efficacy of speech training is mediated by the reorganization of pathological neural networks and structures associated with NOA.

This non-randomized, longitudinal study employs a parallel-group design involving three cohorts: patients with NOA receiving speech training, postoperative cleft palate patients with normal oral articulation (OA) as a clinical control, and healthy controls. The primary objective is to identify the distinctive neural signatures (in both functional connectivity and brain structure) linked to NOA and its remediation through training.

All participants undergo multimodal magnetic resonance imaging (including high-resolution T1-weighted and resting-state functional scans) and standardized speech assessments at baseline. Only the NOA group then receives a structured speech training intervention, followed by post-intervention reevaluation. By comparing changes within and between groups, this study aims to disentangle neural alterations specific to NOA from those related to cleft palate in general.

The integrative analysis of brain-wide changes with a focus on speech-related regions is expected to provide a systems-level understanding of the neural mechanisms underlying NOA and treatment-induced recovery. The findings may contribute to the development of more effective, neuroscience-informed therapeutic strategies.

Inclusion Criteria:

• Aged between 4 and 16 years.
• For patient groups: Diagnosed with non-syndromic cleft palate and having undergone cleft palate repair surgery at least 1 year prior.
• For the NOA group: Diagnosed with compensatory articulation disorder (Non-oral Articulation, NOA) through standardized speech assessment.
• For the OA group: Confirmed to have no compensatory articulation disorder (Oral Articulation, OA) through standardized speech assessment.
• For the healthy control group: No history of any craniofacial malformations, speech disorders, or neurological diseases.
• Participants and their legal guardians are able to understand and willing to provide written informed consent.
• Able to cooperate with and complete magnetic resonance imaging (MRI) examinations (e.g., no claustrophobia, no contraindicated metal implants).

Exclusion Criteria:

• Presence of other known syndromes or cognitive impairments (e.g., autism spectrum disorder, intellectual disability).
• History of other neurological diseases that may affect speech function (e.g., cerebral palsy, hearing loss).
• Presence of severe other orofacial anomalies, or an unrepaired palatal fistula.
• Contraindications for MRI examination (e.g., non-removable metal implants, severe claustrophobia).
• Having received systematic speech therapy within the past 6 months.
• Any other condition deemed by the investigator to be unsuitable for participation in the study.

A final decision regarding the sharing of individual participant data (IPD) has not been made at the time of this registration. This study involves sensitive neuroimaging and speech data. Data sharing considerations, including the feasibility of complete de-identification, compliance with local regulations and ethical approvals, and the establishment of appropriate data access protocols, require further review by the study investigators and sponsoring institution. A formal data sharing plan may be developed in the future and will be made available upon reasonable request or via a controlled-access repository if deemed feasible and appropriate.