Japanese Perceptual Epenthesis is Modulated by Transitional Probability

Abstract

Perceptual epenthesis is the perception of illusory vowels in consonantal sequences that violate native phonotactics. The consensus has been that each language has a single, predictable candidate for perceptual epenthesis, that vowel which is most minimal (i.e., shortest and/or quietest). However, recent studies have shown that alternate epenthetic vowels can be perceived when the perceptual epenthesis of the minimal vowel would violate native co-occurrence restrictions. We propose a potential explanation for these observed patterns: speech perception, and thus also vowel perceptual epenthesis, is modulated by transitional probability whereby epenthetic vowels must conform to the language specific expectations of the listener. To test this explanation, we present two experiments examining perceptual epenthesis of two Japanese vowels—/u/ and /i/—against their transitional probability in CV sequences. In Experiment 1, Japanese listeners assigned VCCV tokens to VCuCV and VCiCV categories. In Experiment 2, participants discriminated VCCV tokens from VCuCV and VCiCV tokens. The results show that sequences where /i/ is transitionally probable are more likely to elicit /i/ perceptual epenthesis.

Multiple Sources of Surprisal Affect Illusory Vowel Epenthesis

“…When another vowel has lower surprisal in a given context (e.g., [i] after palatal consonants), the vowel with the lower surprisal is epenthesized instead, suggesting that phonotactic information can override the use of a phonetically minimal segment. Kilpatrick et al ( 2020 ) also found similar results with [ɡ, ʧ, ʃ], where [u] was epenthesized more often after [ɡ] but [i] was epenthesized after the palatalized consonants [ʧ, ʃ].…”

“…The epenthetic vowel after [p, k, g], [t], and [d] are predicted to be [a, o, e], respectively. In the case of [p, k, g], previous studies have shown repeatedly that it is in fact [u] that is epenthesized after these consonants (Dupoux et al, 1999(Dupoux et al, , 2011Whang, 2019;Kilpatrick et al, 2020). The results also show that neither [u] nor [i] In fact, high vowels are prohibited in the native and Sino-Japanese lexical strata of Japanese, and it is most often [o] that is epenthesized after coronal stops in loanwords (e.g., /faIt/ → [Faito] "fight"; Ito and Mester, 1995).…”

“…When another vowel has lower surprisal in a given context (e.g., [i] after palatal consonants), the vowel with the lower surprisal is epenthesized instead, suggesting that phonotactic information can override the use of a phonetically minimal segment. Kilpatrick et al (2020) When framed rationally, the phonotactic predictability account appeals to an idealized optimal listener's knowledge of context-specific segmental frequency to select the most probable output for a given input. This can be summarized as in ( 2) and (3), where relative probabilities are assigned according to the listener's knowledge of native phonotactics K p rather than the physical signal A as was the case for phonetic minimality in (1).…”

“…Illusory epenthesis, or perceptual epenthesis, is a phenomenon where listeners perceive C 1 C 2 consonant clusters that are phonotactically illegal in their native language as C 1 VC 2 sequences (Dupoux et al, 1999 , 2011 ; Dehaene-Lambertz et al, 2000 ; Monahan et al, 2009 ; Durvasula and Kahng, 2015 ; Whang, 2019 ; Kilpatrick et al, 2020 ). The misperceived medial vowel is not present in the original speech signal but makes the resulting sequence phonotactically legal.…”

“…The main weakness of the phonotactic predictability account is not necessarily inherent to the approach itself but lies instead in the specific assumptions that previous studies have made. First, both Whang ( 2019 ) and Kilpatrick et al ( 2020 ) assume a priori that only high vowels are considered for perceptual epenthesis due to knowledge of high vowel devoicing, and fail to show empirically that non-high vowels do not participate in illusory epenthesis. Second, and more importantly, the two studies do not distinguish voiced and devoiced vowels before calculating predictability, assuming that they belong to the same underlying vowel.…”