Does over-reliance on auditory feedback cause disfluency? An fMRI study of induced fluency in people who stutter.

This study tested the idea that stuttering is caused by over-reliance on auditory feedback. The theory is motivated by the observation that many fluency-inducing situations, such as synchronised speech and masked speech, alter or obscure the talkers feedback. Typical speakers show speaking-induced suppression of neural activation in superior temporal gyrus (STG) during self-produced vocalisation, compared to listening to recorded speech. If people who stutter over-attend to auditory feedback, they may lack this suppression response. In a 1.5T fMRI scanner, people who stutter spoke in synchrony with an experimenter, in synchrony with a recording, on their own, in noise, listened to the experimenter speaking and read silently. Behavioural testing outside the scanner demonstrated that synchronising with another talker resulted in a marked increase in fluency regardless of baseline stuttering severity. In the scanner, participants stuttered most when they spoke alone, and least when they synchronised with a live talker. There was no reduction in STG activity in the Speak Alone condition, when participants stuttered most. There was also strong activity in STG in response to the two synchronised speech conditions, when participants stuttered least, suggesting that either stuttering does not result from over-reliance on feedback, or that the STG activation seen here does not reflect speech feedback monitoring. We discuss this result with reference to neural responses seen in the typical population.