Abstract
Previous studies in speech production and acquisition have mainly focused on how feedback vs. goals and feedback vs. prediction regulate learning and speech control. The present study investigated the less studied mechanism–prediction vs. goals in the context of adult Mandarin speakers’ acquisition of non-native sounds, using an auditory feedback masking paradigm. Participants were asked to learn two types of non-native vowels: /ø/ and /ɵ/—the former being less similar than the latter to Mandarin vowels, either in feedback available or feedback masked conditions. The results show that there was no significant improvement in learning the two targets when auditory feedback was masked. This suggests that motor-based prediction could not directly compare with sensory goals for adult second language acquisition. Furthermore, auditory feedback can help achieve learning only if the competition between prediction and goals is minimal, i.e., when target sounds are distinct from existing sounds in one’s native speech. The results suggest motor-based prediction and sensory goals may share a similar neural representational format, which could result in a competing relation in neural recourses in speech learning. The feedback can conditionally overcome such interference between prediction and goals. Hence, the present study further probed the functional relations among key components (prediction, goals and feedback) of sensorimotor integration in speech learning.
Highlights
In speech acquisition and control, three factors collectively contribute to the development and maintenance of accurate speech production: auditory feedback, motor-based prediction and sensory goals[1,2,3,4]
For the feedback unmasked condition (Fig. 3b), the difference between the highest-value trial and lowest-value trial was significant [F (1, 16) = 8.05, p = 0.01, η2p = 0.34]. These results suggest that vocal learning performance on /ø/ was better in the unmasked condition than in the masked condition
For the feedback unmasked condition (Fig. 7b), the difference between the highest-value trial and lowest-value trial was significant [F (1, 19) = 7.71, p = 0.01, η2p = 0.29]. These results suggest that the vocal learning performance on /ø/ was better in the unmasked condition than in the masked condition
Summary
In speech acquisition and control, three factors collectively contribute to the development and maintenance of accurate speech production: auditory feedback, motor-based prediction and sensory goals[1,2,3,4]. Motor-based prediction (prediction ) refers to the internal estimate of the current state of vocal tract dynamics and subsequent auditory results[3,15,16] Such prediction is based on prior knowledge of the causal relation between speech motor commands and sensory output[1,2,3,4,15,16,17,18,19,20]. The causal relation reflects a key dimension of motor learning (e.g., speech acquisition), the primary computation of which requires the formalization of the motor plan based on sensory targets (i.e., sensory-to-motor transformation) The establishment of such computation and verification of the correctness of sensory-to-motor transformation need an online estimation of how good the results will be based on the planned motor movement. Prediction about sensory consequences of the motor system is needed
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
