You are viewing the SNL 2017 Archive Website. For the latest information, see the Current Website.

 
Poster B58, Wednesday, November 8, 3:00 – 4:15 pm, Harborview and Loch Raven Ballrooms

Enhancing Speech Motor Learning With Noninvasive Brain Stimulation

Adam Buchwald1, Mara Steinberg Lowe1, Holly Calhoun1, Rebecca Wellner1, Stacey Rimikis1;1New York University

Introduction. The potential to enhance human performance using non-invasive brain stimulation has been the topic of active research and controversy. Motor learning has been the domain that has most consistently revealed enhancement from combining training with noninvasive neuromodulation provided by transcranial direct current stimulation (tDCS; Brunoni et al., 2012), with some evidence that the stimulation should be applied prior to the training task (Giacobbe et al., 2013). While tDCS has been studied as a tool to facilitate word production in aphasia and unimpaired speakers with mixed success (Marongolo et al., 2016; Westwood et al., 2017)), it has not been previously tested as a tool to facilitate speech motor learning. We report a study successfully using tDCS as an adjunct to a speech motor learning protocol to enhance speech motor learning, although increased learning is only significant when stimulation is applied prior to the learning task. Methods. 80 participants with no history of impairment completed a two-day protocol. Speech Motor learning task. L1 English participants produced nonwords with 8 different nonnative onset clusters (e.g., /fm/, /gd/). Following brief pre-practice with detailed feedback, participants performed 18 minutes of structured practice producing nonwords with auditory and orthographic presentation (e.g., audio: /fmiku/, written: FMEEKOO). Short-term retention (R1; 30 mins after practice) and long-term retention (R2; 2 days after practice) tested both trained and novel nonwords. Neuromodulation. For active tDCS, a 1x1 Soterix battery-driven current stimulator delivered 20 minutes of 1mA current. The anode was placed over the left motor cortex (C3) and the cathode over the right supraorbital area. Sham tDCS used the same the electrode montage with current ramped up to 1mA over 30 seconds and then immediately decreased. Stimulation was administered before or during the practice session, with participants randomly assigned to one of four groups based on timing (before-during) and type (active-sham) of stimulation. Results. The primary outcome measures were change in word and cluster accuracy (assessed with perception and acoustics) between the first half of practice (P1) and: P2 (second half of practice); R1; and R2. Logistic regression mixed effects models were used to examine changes in whole-word accuracy and cluster accuracy. All groups improved on the task. There was a significant 3-way interaction for both measures, in which participants receiving active stimulation prior to the task improved more from P1 to R1 (word: 10%; cluster: 8%) than other groups (all <4%; word: β=.47, z=3.20, p < .002; cluster: β=.39, z=2.57, p < .01). There was no effect of whether a stimulus item had been trained. Summary and Conclusion. This research represents a novel approach of using tDCS to facilitate learning of novel sound structure sequences in a motor learning paradigm. This work also highlights the importance of considering the various parameters involved in using tDCS, and suggests a link between settings that facilitate motor learning in speech and non-speech domains. Possible extensions to impaired speakers will be discussed.

Topic Area: Speech Motor Control and Sensorimotor Integration

Back to Poster Schedule