Purpose: This study investigates the hypothesis that younger speakers and speakers with more severe speech sound disorders are more likely to use simpler (undifferentiated) tongue gestures due to difficulties with, or immaturity of, lingual motor control. Method: The hypothesis is tested using cross-sectional secondary data analysis of synchronous audio and high-speed ultrasound recordings from children with idiopathic speech sound disorders ( n = 30, aged 5;0–12;11 [years;months]) and typically developing children ( n = 29, aged 5;8–12;10), producing /a/, /t/, /ɹ/, /l/, /s/, and /ʃ/ in an intervocalic /aCa/ environment. Tongue shape complexity is measured using NINFL (Number of INFLections) and modified curvature index (MCI) from splines fitted to ultrasound images at the point of maximal lingual gesture. Age, perceived accuracy, and consonant are used as predictors. Results: The results suggest that as age increases, children with speech sound disorders have lower MCI compared to typically developing children. Increase in age also led to decrease of MCI for the typically developing group. In the group of children with speech sound disorders, perceptually incorrect /ɹ/ productions have lower MCI than correct productions, relative to /a/. Conclusions: There is some evidence of systematic tongue shape complexity differences between typically developing children and children with speech sound disorders when accounting for increase in age. Among children with speech sound disorders, increase in age and perceptually incorrect consonant realizations are associated with decreasing tongue shape complexity.

Purpose: This study investigates the hypothesis that younger speakers and speakers with more severe speech sound disorders (SSD) are more likely to use simpler (undifferentiated) tongue gestures due to difficulties with, or immaturity of, lingual motor control. Method: The hypothesis is tested using cross-sectional secondary data analysis of synchronous audio and high-speed ultrasound recordings from children with idiopathic SSD (n=30, aged 5;0 to 12;11) and typically developing (TD) children (n=29, aged 5;8-12;10), producing /a, t, ɹ, l, s, ʃ/ in an intervocalic /aCa/ environment. Tongue shape complexity is measured using “number of inflections” (NINFL) and “modified curvature index” (MCI) from splines fitted to ultrasound images at the point of maximal lingual gesture. Age, accuracy, and consonant are used as predictors. Results: The results suggest that as age increases children with SSD have lower MCI compared to TD children. Increase in age also led to decrease of MCI for the TD group. In the group of children with SSD incorrect /ɹ/ productions have lower MCI than correct productions, relative to /a/. Conclusion: There is some evidence of systematic tongue shape complexity differences between TD children and children with SSD when accounting for increase in age. Among children with SSD, increase in age and incorrect consonant realisations are associated with decreasing tongue shape complexity.