Is There a Universal Strategy for Discourse Processing?
(Ip, M. H. K., & Cutler, A. (2020). Universals of listening: Equivalent prosodic entrainment in tone and non-tone languages. Cognition)

In these experiments, we examined whether listeners across languages could engage in prosodic entrainment to predict in real time the information structural status of an upcoming word. Prosodic entrainment is a strategy in which listeners attend to the intonational contour of the immediate speech stream, and entrain to it, to predict the location of an upcoming focused word; their attention to the contour allows them to be transported along with it to anticipate the prosodic form of an upcoming word. This strategy is important for discourse processing in many languages where intonation cues the focus status of a word. Using a phoneme-monitoring task, we asked whether this strategy is universally available, even in languages with very different phonological systems, for instance, in tone languages where the use of various suprasegmental cues to lexical identity take precedence over their use in discourse salience. We found that listeners across both tone and non-tone languages (e.g., English and Mandarin Chinese) could use intonation cues to predict and detect upcoming focused words, consistent with the view that prosodic entrainment in discourse processing may be a universal strategy. Interestingly, the extent to which listeners engaged in the strategy was exactly the same across both groups of languages.

*******

Speaker Variability and Listener Flexibility in Prosodic Entrainment
(Ip, M. H. K., & Cutler, A. (2022). In search of salience: Focus detection in speech of different talkers. Language & Speech)

In prosodic entrainment, many different intonation cues are at play in listeners’ anticipation of an upcoming discourse focus. Languages differ in the types of pre-focus intonation cues that listeners can use to predict an upcoming focused word (e.g., a variety of intonation cues exists in English, but only F0 range is a consistent cue in Mandarin Chinese). Here, we tested whether, within a language variety, speakers would differ in their pre-focus prosodic production, and if so, whether listeners would be flexible enough to use any cues signalled by the speaker. We used unsynthesized sentences recorded by four female native speakers of Australian English who happened to have produced different pre-focus intonation cues: a combination of pre-focus overall duration cues, longer pre-target interval duration before the focused word onset (sometimes characterized as “prosodic breaks”), and F0 and intensity (mean, maximum, range) (Speaker 1), only mean F0 cues, mean and maximum intensity, and longer pre-target interval duration (Speaker 2), only pre-target interval duration (Speaker 3), and only pre-focus overall duration and maximum intensity (Speaker 4). All these speakers differed in their pre-focus production despite being of the same age groups (i.e., late 20s – early 30s), same gender (i.e., female), same education level (i.e., postgraduate degrees), and living in the same part of the country (i.e., Western Sydney, New South Wales). Despite these production differences, listeners could, for the most part, entrain to almost every speaker’s cues, and could use whatever cues were available, even when one of the cue sources was rendered uninformative.

*******

Making Sense of Structural Ambiguity
(Ip, M. H. K., & Cutler, A. (2022). Juncture prosody across languages: Similar production but dissimilar perception. Laboratory Phonology.)

This is another set of experiments looking at real time prosodic processing. Here, we ask the question: How do speakers of languages with different intonation systems produce and perceive prosodic boundaries to resolve syntactic ambiguity? Past research has identified similar production strategies across languages (e.g., pausing, pre-boundary lengthening, domain-initial consonantal strengthening, F0 resetting), but very few have adopted a cross-linguistic approach. Here, we compared speakers of two typologically distant and phonologically distinct languages, English and Mandarin Chinese, that nonetheless allow ambiguous sentences with identical structural ambiguity. Consider the following sentences:

(a)
爺爺 / 給 / 她 (#) 嬰兒奶粉 / 喝
ye2ye5 / gei3 / ta1 (#) ying1er2nai2feng3 / he1
Grandpa / gave / her (#) baby formula / to drink

(b)
爺爺 / 給 / 她 / 嬰兒 (#) 奶粉 / 喝
ye2ye5 / gei3 / ta1 / ying1er2 (#) nai2feng3 / he1
Grandpa / gave / her / baby (#) formula / to drink

The two sentences differ in the direct object, and as a consequence, differ in juncture location. In (a), the juncture (#) is realized earlier on in the utterance, giving a sentence with a feminine personal pronoun as the indirect object and a compound noun as the direct object. In (b), the same (segmentally identical) sentence is produced with a later boundary, after “baby”, so that in this case “her” is functioning as a possessive determiner. This ambiguity can occur in English because “her” can be either a possessive or an indirect object. It can also occur in Mandarin because speakers ignore the alienable versus inalienable distinction in everyday speech where the possessive particle -de can be omitted. In fact, according to a large database of informal written and spoken Mandarin, almost half (45%) of associative noun phrases in Mandarin are produced without the particle (Chappel & Thompson, 1992).

Curiously, such ambiguity can also be identical across the two languages in sentences with compound verbs, sometimes with funny consequences. See below!

Consistent with prior studies, both English and Mandarin speakers adopted the same production strategies to disambiguate these sentences, but listeners across languages differed in their perceptual strategies.

*******

Cross-language Production Experiment on 5 Types of Prosodic Focus
(Ip, M. H. K., Shaw, J., & Cutler, A. (submitted). Prosodic strategies of focus expression across languages.)

Here, we compared production of five different types of focus by native speakers of English and of Mandarin Chinese, and we compiled prosodic data from 34,944 measurements, coming from 48 speakers (24 in each language) across five dialogues. We were interested in the relative roles of language-universal and language-specific mechanisms in the production of prosodic focus. Structured dialogue scripts were constructed for each language, with the same words appearing in focused and unfocused position. Duration, F0 (mean, maximum, range), and RMS-intensity (mean, maximum, range) of all critical word and phrase tokens were measured. Overall, English and Mandarin speakers were alike in the ways in which they used prosody to effect focus, consistent with many studies in the literature. However, there were also some cross-language differences: Mandarin speakers produced greater increases in mean and maximum F0 and F0 range, while English speakers tended to produce focused words with higher increases in mean and maximum intensity and intensity range. Further, the pattern of language-specific differences also varied across different dialogues and focus types. Our findings provide evidence of language-specificity in prosodic processing and show that production of information structure can differ even when the same prosodic resources are employed in the same manner.

*******

Effect of Linguistic Experience on Non-Linguistic Representations
(Ip, M. H. K., Imuta, K., & Slaughter, V. (2018). Which button will I press? Preference for correctly ordered counting sequences in 18-month-olds. Developmental Psychology)

This is joint work with Virginia Slaughter and Kana Imuta at the University of Queensland. First, we asked whether counting is an innate ability or a result of a hard-won cultural construction. Second, we examined whether learning to count, a non-linguistic ability, is nonetheless influenced by language experience. We addressed these questions by examining monolingual and bilingual 18-month-olds’ recognition of correct (stable-ordered) and incorrect (mixed-ordered) counting. We found that all infants distinguished between correct and incorrect counting in their native language, and preferred the correct counting events. Interestingly, bilingual infants, but not monolinguals, could also recognize correct counting even when it was produced in an unfamiliar foreign language (i.e., Japanese). Our findings were consistent with the view that infants can represent some counting principles before learning a verbal count routine. Importantly, our findings show a bilingual advantage in infants’ counting abilities.