Auditory Activity Evoked by Self-Produced Foreign Phonemes Changes as Pronunciation Improves
Varjonen, Anni (2021-05-25)
Auditory Activity Evoked by Self-Produced Foreign Phonemes Changes as Pronunciation Improves
Varjonen, Anni
(25.05.2021)
Julkaisu on tekijänoikeussäännösten alainen. Teosta voi lukea ja tulostaa henkilökohtaista käyttöä varten. Käyttö kaupallisiin tarkoituksiin on kielletty.
avoin
Julkaisun pysyvä osoite on:
https://urn.fi/URN:NBN:fi-fe2021060132664
https://urn.fi/URN:NBN:fi-fe2021060132664
Tiivistelmä
Phoneme learning is a complex process that involves the integration of auditory perception and motor activity, and this phenomenon is a central concept in our ability to produce coherent speech. Although phoneme learning has been studied using event-related potentials (ERPs) in the past, most of the research has focused on listening paradigms. Little research has been done on electroencephalogram (EEG) correlates that take place during the active pronunciation of a foreign phoneme. Our study addressed this gap in literature by focusing on an ERP amplitude difference called Speaking Induced Suppression (SIS), during the pronunciation of an unfamiliar phoneme. The SIS event refers to the brain’s tendency to show suppressed auditory responses to self-produced speech in comparison to the same sounds that are passively heard (Niziolek et al., 2013). SIS is thought to reflect a process in the speech production system that compares how well produced speech matches the intended speech (Guenther & Vladusich, 2011), and there seems to be more suppression in the auditory cortex when the produced and attempted sound match closely (Ventura et al., 2009). Our study investigated how SIS behaves in relation to phoneme learning. We analyzed ERPs in response to Finnish participants’ pronunciations on two phonemes (Speak condition): the Estonian phoneme /õ/ (unfamiliar) and the Finnish phoneme /ö/ (familiar). After pronunciation the participants heard an immediate playback of their own vocalizations (Listen condition). We hypothesized that SIS would increase towards the end of the experiment in the Estonian phoneme condition, because the attempted sound and produced sound would match more closely as a result of learning the phoneme. We ran analyses in three time-windows (N1, P2, and Slow-Wave). We assessed learning by having a native Estonian researcher rate the participants’ attempts on the Estonian phoneme from 1 (not resembling /õ/ at all) to 4 (excellent pronunciation of /õ/). Based on our behavioral data analysis, our experiment did produce improvements on the Estonian phoneme pronunciations as the trials went on. However, we did not observe any significant changes in ERPs in the N1 time-window or the P2 time-window. These results indicate that the SIS event did not change as the trials moved forward, nor differed between the Finnish and Estonian phoneme conditions. Therefore, phoneme learning did not seem to affect the magnitude of SIS. We found that the ERPs changed as a function of trials in the Slow-Wave time-window for the Estonian phoneme in the Speak condition, turning more positive as trials went on. These results indicate that the brain responds differently to the Estonian phoneme pronunciation compared to the Finnish pronunciation in the Slow-Wave time-window (300-500ms). This effect took place parallel to improvements on the pronunciation, possibly reflecting high-level cognitive processes related to phoneme learning and the production of a new sound.