Speech production in stuttering: Impact of syllable frequency and word length on accuracy and fluency

Developmental stuttering is a neurodevelopmental speech fluency disorder but its causes and underlying mechanisms have not yet been fully understood. Psycholinguistic studies have found lower non-word repetition accuracy in individuals who stutter, which has been proposed to reflect deficits in translating phonological plans into motor programs. At the interface between phonological and phonetic encoding, fluent speakers are thought to benefit from the retrieval of stored preassembled motor programs as evidenced by syllable frequency effects. However, it is unclear whether AWS also benefit from high-frequency syllables during speech production. To systematically test phonetic encoding procedures in AWS, we designed a novel pseudoword repetition task with pseudowords of different lengths (2, 3 or 4 syllables) containing either high- or low-frequency first syllables to test for response accuracy and speech fluency in 19 AWS and 19 fluently speaking controls (FC). Results showed an increase in errors in both speaker groups and an increase in stuttering frequency in AWS with longer pseudowords. There was no significant effect of syllable frequency on speech fluency in AWS. However, for response accuracy there was a significant interaction between group and syllable frequency: Whereas FC showed the expected syllable frequency effect on response accuracy with more errors on pseudowords with LF first syllables, AWS produced more errors on pseudowords with high frequency first syllables. Based on Levelt and colleagues' psycholinguistic model, we interpreted the inverse syllable frequency effect as an indication of a disrupted interplay between the phonetic encoding routines of retrieving and assembling syllable motor programs in AWS whose stored motor programs might be more fragile.