Neural mechanisms of speech motor learning in persons who stutter

Abstract

Fluent speech production requires rapid coordination among respiratory, laryngeal, and articulatory processes and is mediated by multiple neural systems (Bohland & Guenther, 2006). Stuttering is a fluency disorder characterized by core deficits in speech motor planning. Previous research indicates people who stutter (PWS) exhibit deficits in speech motor sequence learning and are slower and less accurate over practice relative to fluent speakers (Ludlow, Siren, & Zikira, 2004; Namasivayam & VanLieshout, 2004; Smits-Bandstra & De Nil, 2007; Smits-Bandstra, De Nil, & Saint-Cyr, 2006). Furthermore, the neural bases of impaired speech motor sequence learning in PWS are not well understood. We present a study in which PWS (n=18) and persons with fluent speech (PFS) (n=17) were taught phonotactically illegal (e.g. gbesb) and phonotactically legal (e.g. blerk) speech motor sequences over two practice sessions. Functional magnetic resonance imaging (fMRI) was used to investigate brain regions underlying the production of learned illegal syllables and novel illegal syllables. With practice, subjects produced syllables more accurately, which is indicative of motor sequence learning. Our findings suggest a speech motor performance deficit in PWS. Furthermore, these findings indicate speech motor sequence learning relies on a speech motor sequence learning network.