Language in Special Populations
Not all children and adults use language the same way. Variation in sensory, cognitive, and social abilities can have dramatic effects on language development and use. Thus, it is important to studying populations of language users that would not be considered "normal". This provides a unique window on normal language by revealing the ways in which it can develop differently. However, understanding the unique language skills of these children can also help us derive therapies, treatments and experiences that may help them.
In the MACLab we take this approach to heart with research on two populations: language impaired adolescents, and children with cochlear implants.
Projects with Special Populations
- Word recognition in SLI, with Vicki Samelson & J. Bruce Tomblin
- Speech perception and language in cochlear implant users, with Marcus Galle & J. Bruce Tomblin
Specific Language Impairment (SLI)
SLI is a language disorder that affects approximately 8% of the children in Iowa. To compare this
number to other well-known cognitive disorders, both autism and down's syndrom affect less than 1%.
SLI is marked by poor language use in the absence of any obvious cause: SLI children have normal
non-verbal intelligence, no developmental disorders, no neurological disorders, good articulation
(pronunciation) and good hearing. Nonentheless they struggle with many different aspects of language.
Together with Bruce Tomblin and Vicki Samelson
of the Child Language Research Center, we have been investigating how children with SLI understand and
process spoken words. We've been using similar eye-tracking tasks with these children as we
do with normal adults. However, since our subjects live all across the state of Iowa, we have
have installed our eye-tracker in vans which allow us to collect data anywhere we want.
One of the things we don't yet understand about SLI is whether or not it is a unique disorder (like down's syndrome), or whether it represents a population of language users who are simply at the lower end of the scale. The figure to the left represents the language and non-verbal IQ scores of a group of around 100 adolescents in Iowa. We can break this group into four diagnoses: unimpaired, SLI, SCI (who have good language skills, but poor cognitive skills) and NLI (poor language and cognition). The break-points, however, are arbitrary--you can see that there are no clear cut distinctions between the NLI and SLI kids and there are lots of kids in the middle. Thus, instead of seeing SLI as a unique disorder we are examining SLI as a way to understand the dimensions along which children vary in language skills.
In collaboration with Bruce Tomblin of the Child Language Research Center we have begun to examine word recognition in children who can be classified as SLI, SCI and NLI. The visual world paradigm is excellent for use with these children as it is easy for them (all they have to do is identify common words) and yet it can be highly revealing. One of the interesting twists it offers is that rather than studying the tasks that these children do poorly at, we can study a task they do well at (matching words and pictures), and yet see differences (in there eye-movements) in how they perform it. In addition, we are able to use models of word recognition like TRACE to understand their impairments at a deeper level. With these models we can selectively impair certain core processes. For example, we might slow the rate that activation flows from phonemes to words, or damage its ability to retain the input over time. We can than match the predictions of TRACE to our eye-tracking data and pinpoint what part of the system may not be working properly.
|
Students Vicki Samelson |
Collaborators J. Bruce Tomblin (Speech Pathology) |
Presentations
- Samelson, V., McMurray, B., Lee, S., and Tomblin, J.B. (2007, April) Eye Movement During Spoken Word Recognition in Normal and Language-Impaired Adolescents: Activation and Inhibition Effects. Poster to be presented at the Biennial Meeting of the Society for Research in Child Development, Boston, MA.
- McMurray, B., Samelson, V., Lee, S.H., and Tomblin, J.B. (2006, June) Eye-movements reveal the timecourse of spoken word recognition in normal and language-impaired adolescents. Poster presented at the Meeting of the Society for Research on Child Language Disorders, Madison, WI.
Children with Cochlear Implants (CIs)
In recent years a number of hard-of-hearing and deaf children have begun using
cochlear implants.
These small electrical devices are implanted directly in the cochlea in the inner ear and are
hooked up to a small microphone and computer which process sound. For many deaf children they
provide the first window into a new auditory world. While there are important
debates about
the effect of coclear implantation on the rich culture that the deaf and hard of hearing communities
have created, it is nonetheless true that many deaf children are being implanted with CI's and
are successfully learning language. The MACLAb member Marcus Galle is currently working
with Bruce Tomblin of the Child Language
Research Center and
Dr. Bruce Gantz and colleagues in the Dept. of Otolaryngology to understand how
these children learn to understand language. Are they using the same mechanisms as normal-
hearing children or do they get there by another route?
Children who are born deaf now have the opportunity to hear with cutting edge medical technology. Cochlear implants are essentially long thin electrodes which are surgically inserted into the cochlear and attached to a micro computer which translates auditory input into electrical signals. These devices are typically implanted at or around one year of age, and it takes about 4 months for the device to stabilize. This gives us the unique opportunity to both test the effectiveness of the technology and have access to a population which has had a full year of development with out any exposure to sound of spoken language.
Our lab is working with the Otolaryngology department of the University of Iowa Hospital and Clinics to conduct several experiments with children who have had cochlear implants, and a group of normal developing infants. These studies include accessing whether or not CI infants have a preference for infant or adult directed speech, and which particular acoustic properties may account for this preference. We are also looking at how CI infants categorize phonemes based on voice onset time (VOT), if they have a preference for their native language based on rhythm or timing (syllable timed, mora timed, etc), and how they discriminate native and non-native phonemes.
|
Students Marcus Galle Beth Walker |
Collaborators J. Bruce Tomblin (Speech Pathology) Linda Spencer (Otolaryngology) Bruce Gantz (Otolaryngology) |