Purdue SLHS researcher makes novel connections between sight and sound through developmental language disorder research

Written By: Rebecca Hoffa, rhoffa@purdue.edu

Natalya Kaganovich observes a child participating in research in her Auditory Cognitive Neuroscience Laboratory.(Purdue University Photo/John Underwood)

In a noisy school cafeteria, two children speak over the conversations and clanging utensils happening around them. While many observers may see the children talking and think their ears are doing all of the comprehension, Natalya Kaganovich, associate professor in the Purdue University Department of Speech, Language, and Hearing Sciences, is exploring how the eyes play a role in understanding speech and how that might influence language disorders.

“We’re so used to thinking of speech as being auditory only, but actually, in the majority of cases, we will be speaking face-to-face,” Kaganovich said. “In a very noisy environment, which there are plenty of, like coffee shops, restaurants, or school cafeterias and playgrounds, there will be a lot of noise. We know that when it is noisy, people tend to fixate on the mouth, not the eyes, during the conversation.”

Natalya Kaganovich(Purdue University Photo/John Underwood)

Kaganovich studies school-age children in her Auditory Cognitive Neuroscience Laboratory during the critical, formative years when language continues to develop and become more complex. She looks at how their auditory and visual senses work together when they’re absorbing speech.

“Some researchers now argue that how well children are able to attend to the mouth, especially during certain times within development, can be predictive of their lexicon and later language skills,” Kaganovich said.

In studying this audiovisual nature of speech in the College of Health and Human Sciences, Kaganovich has become interested in looking at children with developmental language disorder (DLD), a language disorder in which the child is neurologically typical without any hearing difficulties but has atypical language acquisition. In the disorder, children often take longer to understand the rules for well-formed sentences, have smaller vocabularies and have poorer sentence comprehension. This can cause a wide range of challenges for children in school and puts them at a higher risk for developing dyslexia and literacy difficulties.

“It’s a complex disorder,” Kaganovich said. “On average about 7% of kids at about 5 years of age have this disorder. It’s much less known than autism or ADHD, so people are trying to raise awareness. If you’re not able to fully follow your teacher in class, if you struggle to do anything that’s language-based, or if you’re hesitant to speak because it takes you a long time to formulate a sentence and you’re afraid to make mistakes, all of this will affect your functioning.”

In combining her audiovisual work with her interest in DLD, Kaganovich has contributed valuable insights into some of the symptoms and risk factors to consider in children with DLD and how they may have a multisensory source.

“The work in my lab and in other labs points to the fact that this is not just auditory processing of language that’s impaired in children with DLD,” Kaganovich said. “The ability to use language-related cues from vision is also not typical in these children. It kind of expands the symptoms we should look at and what we consider to be a language disorder. It’s not common to think of visual processing as being a weakness in these children, so that was a novelty of my work, and also now we’re showing that actually even in some siblings there is a very similar pattern that is atypical for processing audiovisual information.”

Recently, Kaganovich published a study in Brain and Language focusing on unaffected siblings of children with DLD — a population she unintentionally stumbled upon in her work. While these children entered an unrelated study as typical children without language difficulties, she found that this group actually performed in the audiovisual assessments more like their siblings with a language disorder than like typical children, which suggests that difficulty with matching auditory and visual information during speech perception may be a risk factor for DLD.

“As we continued to test, we were noticing that even though these children do not come with a diagnosis of language impairment, many of them were not really performing in our tasks like typical kids would,” Kaganovich said. “We started to be concerned about it, and actually, we ended up excluding siblings of children with DLD from our data in the original studies and published their data separately.”

Audiovisual tasks in Kaganovich’s studies included matching a speech sound with the mouth shape of a speaker and listening to a word, viewing a visual articulation and then saying whether the words matched or not. Children with DLD and their siblings had lower accuracy in these tasks and a weaker brain response to audiovisual mismatches when measured through electroencephalography (EEG). While the sample size was relatively small in this initial paper, Kaganovich said this opens a new realm of additional research related to familial characteristics associated with developmental language disorder.

“If we see some significant effects even in a small group of siblings, to me, it also says that it may be worth pursuing this work and looking in more detail at a larger sample of siblings or at parents in order to understand how visual and audiovisual skills may be related to language ability,” Kaganovich said.

Kaganovich recently received a grant from Purdue’s Office of Research through the Laboratory and Core Facility Research Equipment Program, which has enabled her to purchase an eye tracker to further develop her research related to the visual component of speech perception.

“Going forward, I would like to also understand the reason for atypical audiovisual skills in DLD. Why is it that these children cannot reliably detect a difference between visual articulations of very different words? Is it that they’re not looking at the mouth?” Kaganovich said. “Are they looking at it less? Or is it that they are not able to process visual information — such as the mouth shape and movement — the same way typically developing children do? Especially if you add noise and make it more of a realistic environment, how are they fixating on the face and does that actually correlate with their language ability?

“This equipment can add a whole other dimension to my work. I’m excited about it.”

For Kaganovich, the clinical aspect of her work is important in helping provide new insights into how language continues to develop in school-age children both with and without DLD.

“We’re finding that the degree to which children with DLD are identified in schools and even by their parents varies,” Kaganovich said. “We’ve got quite a few children coming into the lab as typical, but when we test their language skills, they’re anything but typical. Often, if they’re not presenting with other disorders such as attention deficit-hyperactivity disorder, autism spectrum disorder or articulation difficulty, they tend to fall through the cracks of the school system and get little or no support. Yet, children need strong language skills to keep up with curriculum demands. When you see how much these children struggle and that we’re overlooking this disorder, it really pulls you in.”