Anne Olmstead and Navin Viswanathan
Interviewed by Ingrid Chuang and Pedro Millán
Our featured partners for this issue are Center for Language Science faculty members Navin Viswanthan and Anne Olmstead, both of whom are professors in the Communication Sciences and Disorders department within Penn State’s College of Health and Human Development. Drawing from their extensive experience in research on speech and communication, they offer us a unique glimpse into the dynamic field of conversation adaptation. Their perspectives highlight one way in which research can be applied to genuine and important practical use, while simultaneously providing new and interesting insight into the nature of speech and communication. We hope you enjoy taking a peek into their research and its broad impacts through this featured partner interview.
Anne Olmstead
Navin Viswanathan
1. Your research looks at how people adapt their speech in conversation. Can you describe what that means, and why you think it’s interesting and important?
Anne: I am interested in looking at what people do when they are in conversations with one another—how they adapt and shift speech to facilitate their interaction. Our descriptions about speech and spoken language are powerful. However, we don’t know much about ordinary, day-to-day speech use.
Navin: Our research looks at how people change the way they talk and listen when they communicate with others. We think that some of these changes are “adaptive” if it helps the task they are doing. This can happen during a normal conversation, but we also use special tasks to study these changes in more detail. We look at different parts of language, like sounds, words, and sentence structures. We find this interesting because the ideas we develop by studying language carefully in controlled settings should help us understand linguistic behavior in everyday life.
2. Can you describe some of the methods you use to understand how people adapt their speech in conversation?
Anne: General methods include the use of linguistic tasks, like tasks assessing differences in words such as “bat” and “cat”. We also implement collaborative approaches in which participants work with other language users to solve problems. What happens in situations in which you are told to be as clear as possible? What is more difficult is analyzing data in a way that tells us what people are doing and how they can adapt during a conversation. Sometimes it is hard to understand how a speaker reacts to a listener, but we are still working on it!
Navin: We have used tasks like word matching, picture comparisons, etc., to study how people talk. Collaborating with others, we plan to include multiple methodologies (e.g., eye tracking) to get a picture of how and when these speech interactions happen.
3. What is the most surprising thing you have learned in your research about speech adaptation in conversation?
Anne: In an exploratory finding for our study on people with ALS (amyotrophic lateral sclerosis, also known as Lou Gehrig’s disease), we observed that participants were rated as more understandable when they engaged in an interactive task as opposed to when they were simply told to “be as clear as possible” in a non-interactive task. This was very surprising to us and motivated us to continue researching new clinical interventions for this population.
Navin: I was drawn to broadening my work to study language in interaction because I thought that the social factors shaping language use were closely related to cognitive factors that I had been focusing on. In what I have been reading and what we have been seeing, it appears that language use is almost never separate from social context. Even in experimental settings in the laboratory, specific demands related to the social context likely shape the phenomena we study in surprising ways.
4. What are some of the findings from research on speech adaptation that you think the general public should know about?
Anne: Ordinary language users may think of themselves as “stable” or “unchanging”.However, one of the most beautiful things about language use is that it is so incredibly flexible. It can change in significant ways in how we talk to and accommodate people in circumstances in which we don’t understand them well at first. I find this nuance very compelling. When we are in a conversation, we are not separate islands; instead, we act as cooperative units.
Navin: I think the studies of linguistic alignment (i.e. when we align ourselves to our conversational partners by using similar words or speech patterns) are well known. But even though this happens a lot, it doesn’t last long, and many studies have mixed results or even find the opposite. So, there is a lot more to learn about interaction beyond just mimicking each other.
5. Is there anything else you’d like to share about speech adaptation, or about interactive communication more generally?
Anne: Generally, it is good to remember that we still know very little about the task of conversing with someone. We have powerful models and theories, but they are only directed to a very small number of conversations between people who basically understand each other. These models are not very good at telling us about conversations that include things such as unfamiliar accents, varied language proficiency, or difficulty with speech. I want to encourage us all to think about it, and we won’t know until we look at these more closely. So, let’s keep looking!
Navin: I believe we need to do more work to better understand spoken language processing during interactions. Studies involving interaction can be more challenging to manage because we need several participants and might have to give up some control over the experiment. However, this could open up new possibilities and insights in different areas of linguistics.