Fifth-year graduate student Ian Calloway was born in Norfolk, Virginia, and grew up in coastal North Carolina--a diverse region of the country that inspired his passion for speech and language.

With the area’s strong military presence, and his own father and aunt in the Navy, Ian recalls that many of the adults he encountered as a child came from other states and often relocated within a year. 

“The entire region was composed of a mix of multigenerational locals and people who were stationed there for a few months at a time,” says Ian. “As a result, the distinctive ways people sounded was always an aspect of speech that I would notice. Part of what led me to take linguistics classes in undergrad at the University of Chicago was a desire to learn about how accents are described and where they come from.”

As his undergraduate studies progressed, Ian’s interest evolved into a focus on speech perception, speech production, and sound change. A particular journal article--required reading in his phonetics lab--first piqued Ian’s interest in attending the University of Michigan. In the article, Linguistics professor Pam Beddor explored how variability in production and perception of coarticulated segments may help to explain a certain class of sound changes. 

“By this point in my program, I knew I was interested in phonetics and sound change,” explains Ian, “but this paper offered an outline of how to conduct research that bridges the synchronic and diachronic.” 

Phonetics and Sound Change: Was that ‘ki’ or ‘ti’?

Fast forward several years, and Professor Beddor now advises Ian’s own research on sound change. Ian’s PhD research investigates the often systematic way that people identify--or misidentify--spoken words in certain situations.

“It’s not surprising to find that people will sometimes misidentify a speech sound (perhaps in a noisy cafeteria), or that the sound the listener identifies tends to show phonetic similarity to what the speaker produced,” says Ian. “I didn’t expect to learn, however, that listeners show bias in their perceptual confusions.

For instance, a listener that hears sound “ki” will be much more likely to report hearing “ti” than to report “ki” when hearing “ti.” These instances of “phonetic perceptual asymmetry,” are not uncommon in the literature, but there hasn’t yet been an explanation proposed as to why listeners would tend to prefer one segment over another:

My dissertation looks at this phenomenon from several perspectives--by using MRI to see how the articulation of the two segments leads to confusions and by using acoustic analysis to learn what aspects of these sounds drive misperception and why the misperception would show asymmetry.”

Evolution of a Language Community

Understanding this sound-change phenomenon can provide clues to the long-term evolution of a language community. “Some long-term sound changes in a speaker community look similar to the types of segment misidentifications that listeners tend to make in the laboratory,” explains Ian. 

“These two phenomena have been suggested to be related to one another somehow, but no mechanism has been proposed to link the two. My dissertation also uses some computational modelling to test the theoretical conditions under which the tendency to misperceive segments in an asymmetrical way can influence the long-term course of a language community’s sound system.”

Ian successfully defended his research prospectus in November 2018, and is currently completing his dissertation. After graduation, Ian hopes to continue using his research and data analysis skills in academia or data science work.