Individuals With Congenital Amusia Show Degraded Speech Perception but Preserved Statistical Learning for Tone Languages

Abstract

Purpose: Individuals with congenital amusia exhibit degraded speech perception. This study examined whether adult Chinese Mandarin listeners with amusia were still able to extract the statistical regularities of Mandarin speech sounds, despite their degraded speech perception. Method: Using the gating paradigm with monosyllabic syllable–tone words, we tested 19 Mandarin-speaking amusics and 19 musically intact controls. Listeners heard increasingly longer fragments of the acoustic signal across eight duration-blocked gates. The stimuli varied in syllable token frequency and syllable–tone co-occurrence probability. The correct syllable–tone word, correct syllable-only, correct tone-only, and correct syllable–incorrect tone responses were compared respectively between the two groups using mixed-effects models. Results: Amusics were less accurate than controls in terms of the correct word, correct syllable-only, and correct tone-only responses. Amusics, however, showed consistent patterns of top-down processing, as indicated by more accurate responses to high-frequency syllables, high-probability tones, and tone errors all in manners similar to those of the control listeners. Conclusions: Amusics are able to learn syllable and tone statistical regularities from the language input. This extends previous work by showing that amusics can track phonological segment and pitch cues despite their degraded speech perception. The observed speech deficits in amusics are therefore not due to an abnormal statistical learning mechanism. These results support rehabilitation programs aimed at improving amusics' sensitivity to pitch.

Tone Deafness in Music Does Not Preclude Distributional Learning of Nonnative Tonal Languages in Individuals With Congenital Amusia

“…More importantly, the authors found that in early gates containing truncated speech (e.g., consonant onset and up to 120 ms of the vowel) amusics-like control listeners-identified frequent syllables more accurately than less frequent syllables and more probable tones more accurately than less probable tones (given the syllable). In other words, Zhu et al (2022) demonstrated that amusics tracked statistical regularities of segments and tones embedded in their native speech in a manner similar to control listeners (Wiener & Ito, 2016;Wiener & Lee, 2020). What remains unclear, and is the focus of the present study, is whether amusics can learn statistical regularities involving phonological tone in a non-native language within a brief training period involving distributional learning.…”

“…Pertaining to the learning outcome, amusics could show no improvement after training in either distribution condition given that they were afflicted with the pitch-processing disorder of amusia (Vuvan et al, 2015). Meanwhile, as mentioned earlier, because several studies did uncover that amusics were able to learn speech sounds in a statistical manner (Omigie & Stewart, 2011;Peretz et al, 2012;Zhu et al 2022), it was also expected that for non-native lexical tones, the perceptual gains would be observed in bimodally rather than unimodally trained amusics from pretest to posttest, which resembles the performance profile in typical listeners (Erickson & Thiessen, 2015;Ong et al, 2015aOng et al, , 2016Ong et al, , 2017aThiessen, 2017).…”

“…Nevertheless, because Omigie and Stewart (2011) removed all pitch information from their stimuli and Peretz et al (2012) did not cue F0 to alter lexical semantics, neither study fully evaluated amusics' statistical learning of speech in which pitch plays a phonological role. More recently, Zhu et al (2022) made use of the gating paradigm to present Mandarin-speaking control listeners and amusics increasingly longer fragments of syllable-tone words. The authors found that amusics' identification of both segments and tones was degraded relative to control listeners (e.g., Shao et al, 2019;Zhang et al, 2017a).…”

“…Several previous studies claimed that amusics' statistical learning of language was preserved since amusics could track regularities of transitional probabilities that distinguished within-word and between-word syllables similarly to control listeners; nevertheless, the training and test stimuli they exploited involved words whose pitch information was removed (Omigie & Stewart, 2011) or did not play a phonological role (Peretz et al, 2012). Besides, although Zhu et al (2022) revealed amusics' successful computation of speech when pitch was involved, their tokens that amusic participants heard were however from their native language. As a result, it was unclear whether amusics were truly able to compute the statistics embedded in the linguistic input with novel tone categories in the process of learning.…”

The Phonological Development of Mandarin Voiceless Affricates in Three- to Five-Year-Old Children

“…Therefore, /y/ was used to substitute for it, and there is a glide between alveopalatal affricates and the nucleus /a/. Despite the fact that Mandarin is a tone language ( Zhu et al, 2022 ), tones were not controlled when selecting the target words due to the issue of familiarity and picturability. The purpose was to the greatest extent to make sure that children in this age range could readily recognize these pictures without heavy cognitive load in case they felt frustrated and withdrew during recording.…”