Somatosensory feedback and auditory feedback have both been found to be important for speech production. Using a sensorimotor learning paradigm, 60 native English speakers’ speech was manipulated to produce a gradual adaptation in the production of vowel formants. Participants were randomly assigned to one of three conditions: a silent waiting condition, and two conditions that involved the silent production of words. Participants’ de-adaptation was measured after this manipulation. It was predicted that those who produced words silently would have stronger de-adaptation, meaning they would return to their regular speech faster than those in the silent waiting condition. Contrary to the hypothesis, all conditions showed the same level of de-adaptation, which suggests that the production of silent speech had no effect. The lack of statistically significant difference between conditions could be explained by the experimental set-up or the training model used to elicit de-adaptation. Despite the lack of difference found between conditions, the current study provides further support for the idea that auditory feedback is used for accurate speech adaptation.
Keywords: somatosensory feedback, speech production, sensorimotor learning, adaptation