Search Abstracts | Symposia | Slide Sessions | Poster Sessions
Brain areas involved in speech motor learning in altered auditory feedback in Mandarin speakers
Poster Session B, Friday, September 12, 4:30 - 6:00 pm, Field House
Qiming Yuan1, Haijun Yao2, Guosheng Ding2, Kate E. Watkins1; 1University of Oxford, 2Beijing Normal University
In speech communication, auditory perception plays a crucial role in encoding, monitoring, and adjusting speech to ensure accuracy and effectiveness. When speaking, the brain continuously compares the planned speech outcome with auditory feedback. If there is a mismatch between the intended auditory target and the actual perceived speech output, the brain initiates corrective adjustments to minimise errors. One effective method to investigate speech motor control is using altered auditory feedback tasks. In these tasks, participants are exposed to real-time perturbations in pitch, loudness, or formant frequencies. Participants generally respond by adjusting their speech production in response to feedback perturbation. Such adaptive responses occur implicitly and involuntarily. Previous neuroimaging studies using unpredictable altered auditory feedback tasks revealed increased activity in right speech motor cortex and auditory cortex bilaterally. However, unpredictable altered auditory feedback does not involve the feedforward control system as feedback returns to normal after each perturbation. In the current MRI study, we adopted a whole-brain analysis to reveal the brain areas involved in speech adaptation during sustained altered auditory feedback in native Mandarin speakers. A behavioural study showed that Mandarin speakers adapt to formant perturbations by offsetting the shift in the first vowel formant. The extent of adaptation is similar to that seen in speakers of non-tonal languages. We scanned 48 native Mandarin Chinese speakers while they read the syllable “dē” out loud. Speech was recorded and fed back to them either normally (baseline) or with a 110-Mel increase in the frequency of the first vowel formant (adaptation). Speech production and feedback occurred during a short (1200 ms) silent interval between echo-planar volume acquisitions (800 ms, voxel size = 2.4 mm isotropic). We compared brain activity between baseline and adaptation phases of the feedback paradigm. Brain activity was greater during adaptation compared with baseline in the right inferior frontal, supramarginal and angular gyri, and posterior lobe of the left cerebellum. Learning-related linear decreases in brain activity were seen in the primary motor cortex and somatosensory cortex, at the level of the lip and tongue representation, bilaterally. Two 10-minute resting-state MRI scans were collected before baseline and after adaptation. We then looked for changes in resting-state networks that reflected acquisition of the adapted state. We found increased connectivity between right Heschl’s gyrus and the rest of the bilateral auditory network. For the resting-state network involving the parietal cortex bilaterally, we saw increased connectivity with primary motor and supplementary motor cortex in the adapted resting state. These results are consistent with a role for a right lateralised frontal and parietal brain network of feedback control (task data) and a feedforward control/learning system within the motor, auditory and somatosensory cortex (resting-state data).
Topic Areas: Speech Motor Control, Language Production