Noninvasive Neurostimulation to the Inferior Frontal Sulcus or Cerebellum Improves Accuracy But Not Timing in a Speech Motor Learning Task.

Abstract

Speech motor learning depends on phonological working memory mechanisms in left posterior inferior frontal sulcus (LpIFS) and motor learning mechanisms in right cerebellum (RCbm). This study examines how transcranial direct current stimulation (tDCS), applied focally to these regions, affects speech production speed and accuracy when practicing novel syllables containing non-native consonant clusters. Healthy young adults (N = 105) practiced producing four pseudowords containing non-native clusters (e.g., "vzutp") with either anodal stimulation to LpIFS, anodal stimulation to RCbm, or no stimulation (i.e., sham; n = 35 participants per condition). Trained stimuli were produced more quickly and accurately than novel stimuli across all stimulation conditions. Overall accuracy (novel and trained combined) was higher after LpIFS stimulation compared to sham but not significantly different between LpIFS and RCbm stimulation, nor after RCbm stimulation compared to sham. The difference in accuracy between novel and trained stimuli (i.e., trained - novel accuracy) was higher after either LpIFS or RCbm stimulation compared to sham, indicating better learning of syllabic motor programs with excitation of these brain regions. No significant difference in trained - novel accuracy was found between LpIFS and RCbm stimulation. Our measure of speed (i.e., stimulus duration) was not differentially affected by stimulation condition. Taken together, these findings highlight the distinct contributions of phonological and motor learning mechanisms to non-native syllable acquisition and suggest that neural stimulation can enhance learning outcomes. Future research should explore the long-term retention of these effects and examine whether real-time feedback enhances learning outcomes.