This study tests whether native speakers of American English exhibit a glide-vowel distinction ([j]- [i]) in a speech elicitation experiment. When reading sentences out loud, participants' pronunciations of 4 near-minimal pairs of pre-existing lexical items (e.g., Eston [iə] vs. pneumon [jə]) exhibit significant differences when acoustically measured, confirming the presence of a [j]- [i] distinction. This distinction is also found to be productively extended to the production of 20 near-minimal pairs of nonce words (e.g., Súmia → [sumiə] vs. Fímya → [fimjə]), diversified and balanced along different phonologically relevant factors of the surrounding environment. Multiple acoustic measurements are compared to test what aspects most consistently convey the distinction: F2 (frontness), F1 (height), intensity, vocalic sequence duration, transition earliness, and transition speed. This serves the purpose of documenting the distinction's acoustic phonetic realization. It also serves in the comparison of phonological representations. Multiple types of previously proposed phonological representations are considered along with the competing predictions they generate regarding the acoustic measurements performed. Results suggest that the primary and most consistent characteristic of the distinction is earliness of transition into the following vowel, with results also suggesting that the [j] glide has a greater degree of constriction. The [j] glide is found to have a significantly less anterior articulation, challenging the application of a representation based on place or articulator differences that would predict [j] to be more anterior.

No abstract

Loanword adaptation exhibits a bias favoring sound cue preservation, possibly due to a conservative caution against deleting cues of unsure expendability in a foreign language. This study tests whether listeners are biased to preserve an acoustically ambiguous sound cue in a nonce word framed as originating from a foreign language. Results show the opposite: Listeners are less likely to transcribe an ambiguous sound cue as a phonological segment when the word containing it is framed as a loanword. However, listeners who identify as more open and accommodating to foreign people and languages show relatively more preservation in the loanword condition.

Previous research demonstrates that simultaneous training of novel sound contrasts in both perception and production can disrupt rather than enhance perceptual learning, indicating that although perception and production are assumed to be closely connected, these modalities may have a competitive relationship. In spite of this perceptual disruption, subjects trained in perception and production show gains in producing the distinction they were trained on, compared to perception-only training. The current study examines how subjects learn to produce a new sound contrast after training in perception or production. L1 Spanish speakers were trained on an unfamiliar Basque sibilant fricative-affricate contrast: /s̺a/–/ʃa/. Since learners’ productions of the contrast may not be identical to the way native speakers distinguish it, and rather than exploring a single phonetic dimension, we apply Linear Discriminant Analysis to acoustic measurements of subjects’ post-test productions to classify whether and how they distinguish the categories in a potentially multidimensional space. This classification model is then applied across conditions to compare production learning across training modes and examine how production learning relates to perceptual learning.

Existing lexical items suggest that American English exhibits a [j]-[i] distinction (e.g., pneumonia [numonjə], Estonia [ɛstoniə]). This study tests if such a distinction can be experimentally elicited in both existing and new items and what acoustic cues most consistently convey it. A sentence reading task elicits the distinction by native speakers of American English using orthographically paired nonce names: 'y' stimuli (e.g., Chobya) expecting [jV] productions, 'i' stimuli (e.g., Shabia) expecting [iV] productions. Stimuli are controlled and diversified along the factors of place and manner of the preceding consonant and word position (initial vs. medial). Multiple acoustic factors of [V] sequences are measured and tested against each other as predictors of stimulus orthography, thus as cues to any elicited distinction, in a generalized linear mixed-effects model. Productions of 'y' stimuli are predicted by significantly earlier transition to the following vowel (represented by timepoint of the F2 maximum), lower F1, and lower intensity. This confirms the presence of the distinction and supports a constriction/height-based classification (Padgett 2008). A significant difference in F2 is not observed; these results are therefore not consistent with a classification of [j] as a coronal sound and [i] as dorsal (Levi 2008).