Semantic focus mediates pitch auditory feedback control in phrasal prosody

ABSTRACT

This study investigated the effect of semantic focus on pitch auditory feedback control in the production of phrasal prosody through an experiment using pitch-shifted auditory feedback. We hypothesized that pitch-shift responses would be mediated by semantic focus because highly informative focus types, such as corrective focus, impose more specific constraints on the prosodic form of a phrase and require greater consistency in the production of pitch excursions compared to sentences with no such focus elements. Twenty-eight participants produced sentences with and without corrective focus while their auditory feedback was briefly and unexpectedly perturbed in pitch by +/−200 cents at the start of the sentence. The magnitude and latency of the reflexive pitch-shift responses were measured as a reflection of auditory feedback control. Our results matched our prediction that corrective focus would elicit larger pitch-shift responses, supporting our hypothesis that auditory feedback control is mediated by semantic focus.

KEYWORDS:

• Prosody
• auditory feedback
• pitch shifts
• feedback perturbations
• intonation
• semantic focus

Acknowledgements

This research was funded by the NIH NIDCD F31 DC017877-01A1 and the Council of Academic Programs in Communication Sciences and Disorders. We would like to thank the participants for their time and effort to participate in this study.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Declaration of Interest Statement

The authors have no financial or non-financial relationships to disclose.

Notes

1 In the prevailing framework of Autosegmental-Metrical theory (Ladd, 2008), pitch accents in American English are in terms of L(ow), H(igh) and downstepped high (!H) tones: L*, H*, L+H*, L+H*, H+!H*, where the * indicates the tone that is anchored to the stressed syllable, and consequently, the pitch target for that syllable.

2 Within the literature in intonational phonology, there have been claims that corrective and contrastive focus are produced with an L+H* pitch accent, while words introducing new information are produced with an H* pitch accent. However, empirical support for that claim is disputed (Chodroff & Cole, 2019; Ladd & Morton, 1997). We do not address the tonal specification of focus-marking pitch accents in this paper.

3 In the framework of alternative semantics (Rooth, 1992), focus is understood as a semantic notion which references a set of semantic alternatives to the focused expression and triggers a set of propositions which contrast in the focused element. Our hypothesis of precision scaling is grounded in this alternatives-based account of focus, and accordingly we predict precision scaling in the “wide” domain of the proposition (i.e., the sentence, in our materials). To the extent that we find evidence for precision scaling in a domain wider than the focused word, and within the domain bounded by the proposition, it can be taken as support for Rooth’s alternatives-based analysis.

4 Our work with data from the present experiment is ongoing, and the analysis of downstream effects of perturbed auditory feedback on the production of the focus-marked word is the subject of a manuscript in preparation.

5 The experimental software was set up to elicit the pitch perturbation 50 msec after voice onset. If the participant cleared their throat, clicked their tongue, or breathed too loudly, the software could mistake these noises as the onset of voice production. Therefore, the pitch perturbation was sometimes elicited before the phrase was produced or at the very onset of production. Every trial was manually checked and removed if the pitch perturbation did not occur 50 msec after voice onset. We anticipated the possibility of mistiming of the perturbation and, therefore, included many trials to have an adequate number after exclusion.

Additional information

Funding

This work was supported by Council of Academic Programs in Communication Sciences and Disorders: [Grant Number ]; the National Institutes of Health National Institute of Deafness and Other Communication Disorders: [Grant Number DC017877-01A1].