Rory Turnbull

Constraints of Tones, Vowels and Consonants on Lexical Selection in Mandarin Chinese

Previous studies have shown that when speakers of European languages are asked to turn nonwords into words by altering either a vowel or consonant, they tend to treat vowels as more mutable than consonants. These results inspired the universal vowel mutability hypothesis: listeners learn to cope with vowel variability because vowel information constrains lexical selection less tightly and allows for more potential candidates than does consonant information. The present study extends the word reconstruction paradigm to Mandarin Chinese–a Sino-Tibetan language, which makes use of lexically contrastive tone. Native speakers listened to word-like nonwords (e.g., su3) and were asked to change them into words by manipulating a single consonant (e.g., tu3), vowel (e.g., si3), or tone (e.g., su4). Additionally, items were presented in a fourth condition in which participants could change any part. The participants’ reaction times and responses were recorded. Results revealed that participants responded faster and more accurately in both the free response and the tonal change conditions. Unlike previous reconstruction studies on European languages, where vowels were changed faster and more often than consonants, these results demonstrate that, in Mandarin, changes to vowels and consonants were both overshadowed by changes to tone, which was the preferred modification to the stimulus nonwords, while changes to vowels were the slowest and least accurate. Our findings show that the universal vowel mutability hypothesis is not consistent with a tonal language, that Mandarin tonal information is lower-priority than consonants and vowels and that vowel information most tightly constrains Mandarin lexical access.

The Role of Predictability in Intonational Variability

Predictability is known to affect many properties of speech production. In particular, it has been observed that highly predictable elements (words, syllables) are produced with less phonetic prominence (shorter duration, less peripheral vowels) than less predictable elements. This tendency has been proposed to be a general property of language. This paper examines whether predictability is correlated with fundamental frequency (F0) production, through analysis of experimental corpora of American English. Predictability was variously defined as discourse mention, utterance probability, and semantic focus. The results revealed consistent effects of utterance probability and semantic focus on F0, in the expected direction: less predictable words were produced with a higher F0 than more predictable words. However, no effect of discourse mention was observed. These results provide further empirical support for the generalization that phonetic prominence is inversely related to linguistic predictability. In addition, the divergent results for different predictability measures suggests that the parameterization of predictability within a particular experimental design can have significant impact on the interpretation of results, and that it cannot be assumed that two measures necessarily reflect the same cognitive reality.

Prominence perception is dependent on phonology, semantics, and awareness of discourse

ABSTRACT The perception of prosodic prominence is thought to be influenced by multiple competing factors. Three experiments tested the effects of phonological salience, discourse context and listener’s knowledge about the discourse on prosodic prominence judgements, using short adjective–noun phrases extracted from a corpus of spontaneous speech. These phrases had either a prominent L + H* 0 or a less prominent H* !H* pitch accent contour. The phrases were presented in a discourse context which either supported or did not support a contrastive interpretation of the adjective. Effects of the contrastive context to increase the perception of prominence only emerged for the phrases with the phonologically prominent L + H* 0 pitch accent sequence. Additionally, the magnitude of the contrast effect was correlated with the listener’s awareness of the discourse context, suggesting an ample interplay between linguistic context, pragmatic context, and phonology in prominence perception.

Contextual predictability and the prosodic realisation of focus: a cross-linguistic comparison

This study explored the effect of contextual predictability on the prosodic realisation of focussed expressions in American English and Paraguayan Guaraní. Pairs of native speakers played an interactive game to elicit utterances that varied in the location of focus in the NP and whether this location was predictable from visual context. The English results confirmed that focussed expressions had more rising pitch accents, longer durations, and higher f0 than non-focussed expressions. Differences between focussed and non-focussed expressions were enhanced when the location of focus was not predictable from context. The Guaraní results confirmed that focussed expressions had distinctive pitch accent and duration patterns relative to non-focussed expressions. Overall prosodic prominence was enhanced when the location of focus was not predictable from context. These results, which are discussed within information-based theories of language production, suggest contextual predictability affects the prosodic realisation of focus, and that this predictability-dependence varies across languages.

Interactions among lexical and discourse characteristics in vowel production

Various factors are known to affect vowel production, including word frequency, neighborhood density, contextual predictability, mention in the discourse, and audience. This study explores interactions between all five of these factors on vowel duration and dispersion. Participants read paragraphs that contained target words which varied in predictability, frequency, and density. Each target word appeared twice in the paragraph. Participants read each paragraph twice: as if they were talking to a friend (“plain speech”) and as if they were talking to a hearing-impaired or non-native interlocutor (“clear speech”). Measures of vowel duration and dispersion were obtained. Results from the plain speech passages revealed that second mention and more predictable words were shorter than first mention and less predictable words, and that vowels in second mention and low density words were less peripheral than in first mention and high density words. Interactions between frequency and mention, and density and ment...

Variability in vowel production: Exploring interactions among frequency, neighborhood density, predictability, and mention

Multiple factors are known to affect vowel production, including word frequency, neighborhood density, close predictability, and mention. In this study, we explored interactions between the effects of all four of these factors on vowel duration and dispersion in a fully crossed within-subjects design. Participants read a series of short stories that contained target words varying in frequency, neighborhood density, predictability, and mention. Vowel duration and dispersion from the center of vowel space were measured. Results from linear mixed effect modeling revealed the expected effect of neighborhood density on duration: vowels in words with higher neighborhood density were longer. Both frequency and mention had significant expected effects on vowel duration and dispersion: vowels in more frequent words and second mention words were shorter and less peripheral. An interaction between frequency and mention on dispersion was also observed, such that low frequency words underwent second mention reduction ...

2. Exploring variation in phonetic reduction: Linguistic, social, and cognitive factors

Substantial empirical research has revealed that temporal and spectral phonetic vowel reduction occurs in “easy” processing contexts relative to “hard” processing contexts, including effects of lexical frequency, lexical neighborhood density, semantic predictability, discourse mention, and speaking style. Theoretical accounts of this phonetic reduction process include listener-oriented approaches, in which the reduction reflects the talker’s balancing the comprehension needs of the listener with production effort constraints, talker-oriented approaches, in which reduction is argued to result entirely from constraints on speech production processes, and evolutionary approaches, in which reduction results directly from long-term interactive communication within a community. Recent research in our laboratory has revealed complex interactions among the linguistic, social, and cognitive factors involved in phonetic vowel reduction processes. These interactions reveal variation in the robustness of phonetic reduction effects across linguistic factors, as well as different patterns of interactions among linguistic, social, and cognitive factors across acoustic domains. These interactions challenge aspects of each of the three existing models of phonetic reduction. We therefore propose that a more complex view of the relationship between processing demands and phonetic vowel reduction processes is necessary to account for these observed patterns of variation.

Assessing predictability effects in connected read speech

Abstract A wide range of reduction phenomena have been described in the literature as predictability effects, in which more predictable units (i.e. words, syllables, vowels) are reduced in duration or other acoustic dimensions relative to less predictable units. The goal of the current study was to critically evaluate these predictability effects on vowel duration in read speech to explore the extent to which they reflect a single underlying phenomenon. The results revealed shorter vowel duration for words with high phonotactic probability, for high-frequency words (in clear speech only), and for words in plain lab speech relative to clear speech. However, the results also revealed qualitatively different effects of three measures of contextual probability (cloze probability, written trigram probability, and spoken trigram probability). Greater spoken trigram probability predicted longer vowel duration, contrary to expectations, and this effect was limited to high-frequency words in first mentions and in plain speech. Cloze probability and written trigram probability exhibited even more complex interactions with other predictability measures. These results provide evidence for fundamental differences in these measures of predictability, suggesting that a more nuanced perspective on predictability effects and the mechanisms underlying them is necessary to account for the complexity of the empirical data.

The phonetics and phonology of lexical prosody in San Jerónimo Acazulco Otomi

San Jeronimo Acazulco Otomi (SJAO) is an underdescribed and endangered Oto-Manguean language spoken in central Mexico. This paper provides an analysis of the phonology of tonal contrasts in SJAO and the phonetics of their realization based on pitch pattern data derived from audio recordings of citation forms of SJAO words. Each SJAO lexical word has one and only one tonal sequence – either /H/ or /HL/. This sequence is underlyingly associated with one syllable in the word. Other syllables are not specified for tone, and their phonetic realization is predictable depending on their position relative to the tonal syllable. A phonetic analysis revealed that underlyingly-tonal syllables are phonetically distinct from non-tonal syllables: those with /H/ are produced with greater vocal effort (measured by spectral tilt), and those with /HL/ are longer, louder, and bear a higher f0 (fundamental frequency), compared with non-tonal syllables. This analysis differs from previous accounts of lexical prosody in other Otomi varieties, which have either described a three-way system of high, low, and rising tones contrasting on every stem syllable, or a system where one syllable per word is assigned a stress-like ‘accent’. This difference from previous analyses suggests that there is a third possible characterization of lexical prosody for Otomi, which is appropriate for SJAO and potentially other understudied varieties.

What governs a language’s lexicon? Determining the organizing principles of phonological neighbourhood networks

The lexicons of natural language can be characterized as a network of words, where each word is linked to phonologically similar words. These networks are called phonological neighbourhood networks (PNNs). In this paper, we investigate the extent to which observed properties of these networks are mathematical consequences of the definition of PNNs, consequences of linguistic restrictions on what possible words can sound like (phonotactics), or consequences of deeper cognitive constraints that govern lexical development. To test this question, we generate random lexicons, with a variety of methods, and derive PNNs from these lexicons. These PNNs are then compared to a real network. We conclude that most observed characteristics of PNNs are either intrinsic to the definition of PNNs, or are phonotactic effects. However, there are some properties—such as extreme assortativity by degree—which may reflect true cognitive organizing principles.