Vocal Fold Vibratory Kinematics and Acoustic Correlates in Pressed and Breathy Phonation

The overall aims of this study were to (a) examine how vocal fold kinematics differ across typical, pressed, and breathy phonation in vocally healthy adults and (b) investigate the relationships between high-speed videoendoscopic-derived kinematic measures and acoustic measures of cepstral peak prominence (CPP) and the amplitude difference between the first two spectral harmonics (H₁-H₂) and whether the relationships vary by phonation type.

Forty vocally healthy adults (32 female, eight male, with a mean age of 26 years) underwent simultaneous transoral rigid high-speed videoendoscopy (HSV; 4,000 frames per second) and acoustic recording during sustained /i:/ in three phonation types: typical, pressed, and breathy. Primary HSV parameters included closing quotient (ClQ), speed index (SI), amplitude-to-length ratio (ALR), stiffness index (STI), and normalized maximum area declination rate (MADRn). Primary acoustic measures were CPP and H₁-H₂. Mixed analyses of variance were conducted for phonation type differences in HSV parameters with main effects of phonation type, sex, and their interaction. Then, multiple regression models with phonation type interactions were conducted to assess the relationships between HSV and acoustic measures.

Relative to typical phonation, simulated pressed phonation showed lower values of ClQ, higher MADRn, and higher STI with large effects, whereas simulated breathy phonation demonstrated higher ClQ and lower MADRn with medium effects. CPP was significantly negatively correlated with ClQ and positively correlated with MADRn, SI, and STI. H₁-H₂ was significantly positively correlated with ClQ and ALR and negatively correlated with MADRn, SI, and STI. There was a significant phonation type interaction with the correlations between H₁-H₂ and MADRn, SI, and STI; in each, breathy phonation had a strong, negative relationship and pressed phonation had a small or negligible relationship. ClQ consistently correlated with both acoustic measures across all phonation types.

Vibratory patterns in pressed phonation were suggestive of increased vocal fold contact stress, as lower ClQ and higher MADRn values suggest more abrupt, faster glottal closure. CPP and H₁-H₂ can reflect underlying glottal physiology, but their predictive value depends on phonation type in most cases. However, findings suggest that ClQ could be a robust physiological parameter with stable acoustic correlates regardless of phonation type.