Research aids people with communication disorders
Oct. 20, 2005
KALAMAZOO-- Being able to say what you want to say when you want to say it is taken for granted by most people.
But for people with severe speech impairments, communication is not so easy.
With that spirit in mind, Western Michigan University is leading a three-university research project with the help of a federal grant that will examine how people with severe communication impairments can communicate more effectively using voice output computers.
The grant, from the National Institute on Deafness and Other Communication Disorders of the National Institutes of Health, is for $340,000 for the first year, but subsequent funds for the next two years are expected to bring the total to nearly $1 million. Researchers at WMU, Kansas State University and the University of Delaware will use the funding to test a model predicting the impact of voice output computer messages on the attitudes of public service providers. The model ultimately could lead to better design of assistive technology for people who rely on those computer systems for communication.
"For the past five years, we have been experimenting with one type of voice output computer," says Dr. Jan Bedrosian, WMU professor of speech pathology and audiology and the project's lead researcher. "The computer allows for the storage and relatively quick retrieval of one or more sentences in an effort to increase the user's rate of communication, which is especially important when interacting in public places. The people using the computers have to anticipate their communication needs, compose and store the messages and then hope that these pre-stored messages 'fit' the communicative situation."
But even ordinary social interactions can become complex, and no matter how hard people try to anticipate every possibility, problems still arise. For example, "the computer message may not have enough information or may be only partly relevant to that situation," Bedrosian says. "When this happens, the person could choose to use the message anyway, even though it may be slightly 'off,' so that there won't be a long delay in the conversation. As a second choice, the person could take the time to edit the message letter-by-letter and create a message to better fit the situation. But then this could be a problem if the person is in a checkout line and holding everybody else up."
With either choice, communication is flawed.
In their previous NIH-funded project, the researchers developed a theoretical model that consists of a hierarchy of these types of message choices. They found that sales clerks responded more favorably to some message choices than to others, at least within the type of public setting examined. In the current project, the research team will test how well this model holds up across a greater variety of message choices, public settings and service providers.
"We're trying to help the people use these systems more effectively by making the best communication choices, so they will be viewed more positively in public," Bedrosian says. "We want people who have severe speech impairments to be welcome and included in public settings, and frequently they're not. If we can help them make the best choices when using these computers, then it might give them and members of the public a better communication outcome."
The research ultimately could help designers of voice output computers create better software that fits more situations and lets people alter messages more quickly and easily.
"Some people don't want to go out in public and use these systems," Bedrosian says. "So we really want to help people use the computers more effectively and also expose the public to these types of communicators so they can better deal with them.
"But there's no way that we can train the world. If we can build a model that works in a lot of different situations, then it will be very helpful to researchers developing new assistive technology, and their efforts will be more fruitful."
Media contact: Mark Schwerin, (269) 387-8400, email@example.com