Previous linguistic research has shown that speakers accommodate by imitating the same linguistic forms they observe an interlocutor using. It is not known whether speakers also accommodate by using linguistic forms they expect from an interlocutor based on social cues, even if the forms are absent from the input. The dissertation targets a Southern variant---monophthongal /ay/---where the vowel in 'ride' sounds more like 'rod.' The project examines whether speakers produce more monophthongal /ay/ when conversing with a Southern talker who never uses the /ay/ vowel, as well as the cognitive mechanisms responsible for this behavior. Understanding how sociolinguistic expectations are formed and accessed during expectation-driven accommodation contributes to a general understanding of how social stereotypes are formed, mentally represented, and utilized.
The project employs a novel word-naming game paradigm designed to elicit tokens of /ay/ from participants while exposing them to Southern speech that contains no /ay/ tokens. Acoustic properties of participants' /ay/ tokens are compared before and after exposure to a Southern or Midland (control) talker to determine whether accommodation occurs. Experiment 1 investigates whether accommodation based on expectations alone occurs. Experiment 2 tests whether expectation-driven accommodation is socially or structurally rooted by manipulating social cues (dialect labels) and structural cues (linguistic variants). Experiment 3 uses a joint perception-production task to determine whether participants exhibit /ay/-category boundary shifts, and whether such perceptual shifts predict accommodative behavior. This task attempts to pinpoint the perceptual changes that must take place for accommodation to occur, as well as understand the role of social information in both early perceptual changes and subsequent production behaviors. Self-reported surveys are paired with all three experiment to gauge the influence of social, cognitive, and attitudinal measures in accommodative behavior. Results contribute to scientific understanding of the relationship between perception and linguistic production, as well as the mental relationships between social and linguistic knowledge.
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.
