Alumni Bulletin - Winter 2008 - (Page 9)

RESEARCH psychology THE SYLLABLE SOLUTION Try repeating the words “margin” and “marble” multiple times. It’s a bit of a tongue twister, isn’t it? Padraig O’Seaghdha, associate professor of psychology, believes that any difficulty you may have experienced in clearly repeating this pair of words arises in filling the individual sounds of the words. Because “margin” and “marble” share the same first syllable, it becomes tricky to repeat a few times over. You may say the wrong word (e.g. saying “margin” instead of “marble”) or produce an incorrect sound (e.g. saying “marbin” instead of “margin”). The precise nature of the difficulty may also vary in different languages. The outcome of the recitation task may be different in English than in other languages such as Spanish or Chinese, where syllables play a greater role. Little is known about the role syllables play in language processing, and there is no consensus on whether syllables are represented in word memory. But what is known is that the syllable’s role varies greatly among languages around the world. O’Seaghdha believes that comparing different languages can make all the difference in understanding syllable production, and that it has great practical applications. His current research examines three of the most commonly spoken languages in the world—English, Spanish, and Mandarin Chinese. “Addressing the status of the syllable in language production theory is critical because of the syllable’s central role in making speech possible,” O’Seaghdha says. “This project will provide a multipronged attack on this signiﬁcant weakness in a core aspect of our understanding of language production.” The “margin”-“marble” recitation task is one example of an experiment that serves as the basis for O’Seaghdha’s research. The subjects are prompted by words on a computer screen that share certain properties, and they are then asked to recite the words into a microphone. Research assistants track how many times the recitation was correctly completed in the allotted time. The word combinations that take longer to recite correctly, or trigger errors, provide clues to the underlying principles of how correct production is organized. These same experiments are replicated by O’Seaghdha’s coinvestigators at the University of Illinois, at the Universities of La La- O’Seaghdha believes his research will improve computational models guna and Granada in Spain, and at of speech projection, speech synthesis software, and speech therapy. the National Cheng Kung University in Taiwan—each in their native languages. Communication Disorders, O’Seaghdha be“In the research literature, most of the time lieves his research will ultimately provide guidwe compare informally across languages, but beance in computational models of speech cause experiments will vary in a lot of details production, in speech synthesis software, aphafrom site to site, cross-language comparisons sia rehabilitation, and speech therapy. A deeper will be difﬁcult and imperfect,” O’Seaghdha understanding of the language process will ulsays. “We’re doing closely matched compartimately translate to more effective applicaisons, which make it easier to draw much bettions and treatments. ter conclusions.” Aphasia is a communication disorder that Some things are already clear about the difimpairs a person’s ability to process language, ferent use of syllables across languages. For inaffecting their speech as well as their ability to stance, syllables are well deﬁned in the Chinese understand others. For instance, a stroke viclanguage, because speakers go to the syllable tim will need to relearn how to process lanfirst when forming words. English speakers, guage. But the theories that currently guide therapy for such disorders, O’Seaghdha says, are often too general and English-centered, failing to take into account the differences in word production from language to language. “Theories, computational models, and therapeutic interventions that misconstrue the syllable are, in a very real sense, ‘unable to speak.’ The project will clarify the prospect for a language-general model of word production that encompasses variation in the roles syllaKurt Schmenger ’08 recites similar words into a bles play, and will also specify languagemicrophone. Later, O’Seaghdha and his assistants will speciﬁc process models for three of the world’s look for errors and pauses that would give clues to how major languages,” he says. language is produced. O’Seaghdha and his co-investigators are however, rely more on whole words in language finding that the research has implications in production and only put syllables together other areas of language study and are curright before articulating the words. That is why rently examining how second language learnthe outcome of recitation tasks could poteners navigate the differences between two tially differ between languages. languages, such as English and Chinese, that Funded through a three-year grant from build the sound patterns of words in very difthe National Institute on Deafness and Other ferent ways.—Tricia Long PHOTO BY THEO ANDERSON winter 2008 9

Table of Contents Feed for the Digital Edition of Alumni Bulletin - Winter 2008

Alumni Bulletin - Winter 2008 Contents From the President's Desk Mailbox On Campus Research Arts & Culture Sports Ugly is in the Eye of the Beholder Fitch's Law of Motion A Blessing to the Community The Amidonian Acclamation China Trip Bridges Cultures, Generations From the Publisher's Desk Alumni News The Last Word

Alumni Bulletin - Winter 2008

For optimal viewing of this digital publication, please enable JavaScript and then refresh the page. If you would like to try to load the digital publication without using Flash Player detection, please click here.