EMBARGOED FOR RELEASE:
2:00 PM (EDT), WEDNESDAY, MAY 7, 1997
TO COINCIDE WITH PUBLICATION IN NATURE
UCSF RESEARCHERS OFFER MORE PRECISE EXPLANATION
FOR A LANGUAGE DISORDER AFFECTING MILLIONS OF CHILDREN
A startling new finding made by researchers from the University of
California San Francisco strengthens evidence that a disorder which
prevents millions of U.S. children from keeping up in the classroom stems
from an inability to process sound normally, and not from problems that
originate primarily in higher brain regions as many other scientists had
earlier proposed.
The discovery, reported in the May 8 issue of the scientific journal
Nature, holds the key to easier diagnosis of the disorder, called specific
language impairment, the researchers say. The finding already is being
incorporated into the design of new training programs to improve
language-learning skills in language-impaired children.
Children with specific language impairment fall behind in language skills
despite normal intelligence and apparently normal hearing. The source of
the problem has been poorly understood, and standard speech therapy
techniques used to treat the disorder have proved unsatisfactory.
The researchers found that children with specific language impairment are
affected much more profoundly than unimpaired children by a phenomenon
known as "masking." Masking refers to a natural limitation in the human
ability to detect any particular sound that is presented simultaneously or
within a small fraction of a second of other "masking" sounds.
The new study refines scientific understanding of why the language-impaired
children find it difficult to distinguish individual sounds presented
during normal speech. The study specifies more clearly the cause of the
impairment, which in turn leads to slow progress in school.
Speech fits into the category of masking because it consists of a stream of
auditory stimuli occurring sequentially in time. In normal individuals the
masking of individual speech sounds by preceding or following sounds is not
sufficient to impair speech processing.
However, Beverly Wright, PhD, found that, compared to normal children,
those with the disorder in some masking situations required that tones be
about 45 decibels more intense before they could be heard over masking
noise. This is comparable to the difference between the sound level in a
quiet room and at the side of a superhighway, Wright says.
Wright conceived and conducted the study as a UCSF research scientist in a
group led by Michael Merzenich, PhD, a professor of otolaryngology and a
research scientist with the W.M. Keck Foundation Center for Integrative
Neuroscience at UCSF. Wright, who is trained in acoustics and psychology,
collaborated with researchers at the University of Florida to design and
conduct testing. Wright recently completed her UCSF research and has taken
a position as an assistant professor of audiology and hearing sciences at
Northwestern University, in Evanston, Ill.
"For a tone to be detected in the presence of masking noise, it must be
more intense than usual or else must be separated in time or frequency from
the masking noise," Wright says. "We observed that the language-impaired
children were consistently poorer than the controls at detecting a brief
tone presented with a masking noise, particularly when the brief tone was
turned on immediately prior to the masking noise. This phenomenon is
called backward masking."
Although Wright studied only eight language-impaired children and eight
unimpaired children for comparison, the differences in their responses was
significant, she says. Despite their distinctively different responses to
the tests Wright administered, the differences between normal and
language-impaired children would not be detected on standard hearing tests
given in school, she says.
Merzenich estimates that specific language impairment affects about ten
percent of children. "This study provides a basis for early identification
of this disorder, and helps us to define the problem more precisely," he
says. "It also has implications for how we might better treat
language-impaired children. Additionally, the study may lead to new
insights regarding the neurological basis of this disorder and of language
processing in general."
Although masking is a phenomenon that has been well known for decades to
experts in psychoacoustics and is presented in standard textbooks in that
field, it apparently had not been investigated until now in children with
specific language impairment.
Wright and Merzenich say the new results offer a refinement and elaboration
of earlier findings by Paula Tallal, PhD, professor and co-director of The
Center for Molecular and Behavioral Neuroscience at Rutgers University.
Tallal was the first to propose and demonstrate that a difficulty in
understanding rapidly changing speech frequencies is the fundamental source
of language problems in children with specific language learning
impairment. She found that the children are deficient in their ability to
perceive certain consonant sounds in normal speech, such as "ba" and "da."
In his research, Merzenich studies changes in sensory systems as a way to
learn more about the brain's "plasticity," which is its potential to
establish new networks of connections between nerve cells in support of a
variety of brain functions. His interest in this area earlier led to the
development of the cochlear implant, a device that permits the deaf to hear
sound and even to understand speech, and led to a collaboration with Tallal.
In two articles published in the January 3, 1996 issue of Science, the
collaborators described studies of a new training technique they developed
to speed sound-processing and improve language skills in children with
specific language impairment. Their approach relies on slowing and
amplifying the difficult-to-process sounds. The modified speech and other
sounds used in the training were incorporated into colorful and engaging
computer games.
The rapid improvements they measured, typically two grade levels in four
weeks, were incompatible with the idea that the disorder could arise in
higher processing centers in the brain, the researchers concluded.
Additional authors of the current Nature study include Linda Lombardino,
professor of communication processes and disorders; Wayne King and Cynthia
Puranik, graduate students; and Christiana Leonard, professor of
neuroscience, all of the University of Florida.
Last year, UCSF granted an exclusive worldwide license for
commercialization of the training technique and its computer software
applications to Scientific Learning Corporation (SLC), a privately held San
Francisco company.
For more information about Scientific Learning Corporation or its products,
call (415) 296-1470, or contact its web site at http://www.scilearn.com.
The address for the UCSF web site on language-based learning disabilities
is http://www.ld-ucsf.edu.
###