Pausing, inhalation, and language structure

University Of Oregon Eugene, Eugene OR

Investigators

Abstract

It is difficult to imagine how what individuals plan to say in conversation interacts with breathing because language is so clearly a high-level cognitive function, and breathing is so clearly a basic physiological function. Yet, planning for language and motor planning for breathing do interact. During speaking, inhalations are timed to coincide with important grammatical junctures, like sentence boundaries. Also, the length of sentences that speakers produce can be predicted from the depth of the breaths taken before speaking. This project focuses on how language and breathing interact during speaking. The goal is to advance psycholinguistic theory and, with it, the conceptual models that frame understanding and treatment of communication disorders. Importantly, no current theories of language production address speech breathing despite its fundamental role in speaking: all of sound production, from whispers to shouts, depends on air flowing out from the lungs. Thus, this project addresses a substantial gap in understanding how spoken language works. The research team's hypothesis is that language and speech motor processes, such as speech-related breathing, influence one another because they develop together and are used together in communication. This position motivates the developmental approach taken in this project. School-aged children and adults engage in different spoken language tasks while their speech is audio recorded and their breathing (chest and abdominal) movements are measured. The spontaneous speech elicited from the tasks the tasks is segmented into pause-delimited utterances, with pauses coded for the presence or absence of inhalation and, when inhalations are present, for inhalation depth. At the same time, the utterances are transcribed and then tagged for part-of-speech and other linguistic information. Using the tools of corpus linguistics and model comparison, the researchers identify language factors that best explain the inhalation data. The researchers predict that discourse structure and language habits provide a better account of inhalation patterns than syntactic boundary strength or syntactic complexity—the language factors most often discussed in relation to inhalation. Additionally, physiology (e.g., measures of lung function and body size) likely provides a better account of children’s speech breathing patterns than adult speech breathing patterns because children’s discourse structure is flatter than adults’ and their language habits are less entrenched than adult language habits. The project establishes a developmental corpus of spontaneous speech marked out for inhalation patterns that other researchers can use to investigate speech breathing from different theoretical perspectives. The researchers also make publicly available the in-house software developed for the project. Finally, the research team engages in science education through community outreach and by providing undergraduate students with research experience in a setting that fosters cross-disciplinary interactions between students who pursue study in linguistics and psychology and those who pursue study in communication disorders and sciences or in disability studies. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.