Learning and adaptation in speech production without a vocal tract

• Published in: Scientific reports Vol. 9; no. 1; pp. 13582 - 11
• Main Authors: Thompson, Megan M C, Houde, John F, Nagarajan, Srikantan S
• Format: Journal Article
• Language: English
• Published: England Nature Publishing Group 19.09.2019; Nature Publishing Group UK
• Subjects: Adult; Auditory Perception - physiology; Auditory stimuli; Feedback; Feedback, Sensory; Female; Humans; Learning - physiology; Male; Middle Aged; Motor task performance; Observational learning; Sensorimotor system; Speech; Speech - physiology; Speech Perception - physiology; User-Computer Interface; Visual perception; Young Adult
• ISSN: 2045-2322; 2045-2322
• DOI: 10.1038/s41598-019-49074-4

How is the complex audiomotor skill of speaking learned? To what extent does it depend on the specific characteristics of the vocal tract? Here, we developed a touchscreen-based speech synthesizer to examine learning of speech production independent of the vocal tract. Participants were trained to reproduce heard vowel targets by reaching to locations on the screen without visual feedback and receiving endpoint vowel sound auditory feedback that depended continuously on touch location. Participants demonstrated learning as evidenced by rapid increases in accuracy and consistency in the production of trained targets. This learning generalized to productions of novel vowel targets. Subsequent to learning, sensorimotor adaptation was observed in response to changes in the location-sound mapping. These findings suggest that participants learned adaptable sensorimotor maps allowing them to produce desired vowel sounds. These results have broad implications for understanding the acquisition of speech motor control.