showed insight into complex relationships
among the pattern pairs, the researchers
report in an upcoming Proceedings of the
National Academy of Sciences.
“The sleeping brain searches for distant
relationships in remembered
information as a way to
develop general rules for
dealing with new situations,”
says Harvard’s Matthew P.
Walker, a coauthor of the new
study.
The scientists trained
56 healthy young adults to
choose among six colored patterns as they appeared in a
series of five pairings. In the
training sessions, experimenters presented in random order the
pairs that defined the range of most-pre-ferred to least-preferred patterns. Participants were told that they had made the
correct choice when they selected pattern
A over pattern B, B over C, C over D, D over
E, and E over F.
After varying time delays, volunteers were
tested for their knowledge of the hierarchy
of patterns. For instance, discerning short-range relationships included choosing pattern B over pattern D, although participants
had not previously seen such pairings. Solving long-range relationships included selecting pattern B over pattern E or F.
After a 20-minute, a 12-hour, or a 24-hour
delay, volunteers displayed comparably good
memories for what they had initially learned,
such as picking pattern A over pattern B.
However, the 12 individuals tested after
20 minutes correctly discerned
the more-preferred item in new
pairings, such as B over D, only
about half the time, indicating
that they were guessing.
In contrast, 31 people were
given a 12-hour delay. Of
these, 14 were trained in the
evening and tested after a
night’s sleep. The rest were
trained in the morning and
tested later the same day.
Thirteen others were tested
after 24 hours that included a night’s sleep.
These three groups all correctly detected 75
percent of short-range relationships.
The 27 participants who had slept did
even better at identifying long-range relationships, correctly spotting them 93 percent of the time. After a 12-hour delay without sleep, the corresponding detection rate
reached only 69 percent.
Participants who had slept weren’t aware
of their newfound knowledge. This finding
clashes with the popular view among scientists that relational memory requires conscious thought.
The best evidence that sleep aids mem-
QUOTE
“The sleeping brain
searches for distant relationships
in remembered
information …”
MATTHEW WALKER,
Harvard Medical
School
ory comes from studies of automatically
executed tasks, remarks neuroscientist
Howard Eichenbaum of Boston University.
These tasks include detecting the orientation of diagonal bars (SN: 12/2/00, p. 358).
Ellenbogen’s team may have uncovered
an unconsciously controlled form of sleep-enhanced relational memory, Eichenbaum says. —B. BOWER
Diabetes from
Depression
Older adults face dual risk
Adults 65 and older who report depressive
symptoms are 50 to 60 percent more likely
to develop diabetes than are their peers,
according to a new study.
The study is the first to show that depression alone, apart from lifestyle factors such
as poor diet and lack of exercise, can trigger type 2 diabetes in older adults, reports
Mercedes Carnethon of the Northwestern
University School of Medicine in Chicago.
“This means doctors need to take depressive symptoms in older adults very seriously,” she says. Earlier research had shown
the connection in younger adults.
Among all age groups, adults 65 and older
Pregnancy and Pollution
Women living in areas with poor air quality have babies with lower birthweights
Pregnant women exposed
to moderate amounts of
several common air pollutants have babies with lower
birthweights than do women in
areas with cleaner air, according to a new study.
Newborns with low birthweights face an increased risk
of lifelong health problems.
Previous studies searching for
a link between air pollution and
birthweight had yielded mixed
results.
Now, in one of the largest
studies of this kind, scientists
at Yale University looked at
records of 358,504 births in
Massachusetts and Connecticut. The team found that four
types of air pollution correlate
with low birthweight. The culprits are carbon monoxide,
nitrogen dioxide, and two
classes of airborne particles:
those smaller than 10 and
smaller than 2.5 micrometers
(designated PM2.5).
“Maternal exposure to air
pollution may adversely affect
risk of low birthweight, even in
areas without high pollution
levels,” says Michelle L. Bell,
lead scientist on the newly
reported work. Air-pollution
amounts were based on Environmental Protection Agency
records for the 15 counties in
which the women lived while
pregnant. Only two counties—
New Haven and Fairfield,
Conn.—didn’t meet EPA’s air-quality standards, exceeding
the standard for PM2.5.
Carbon monoxide showed
the largest effect. In one comparison, the scientists considered the average birthweights
in counties at the 75 percent
point in rank for a given pollutant and in counties at the
25 percent mark. For carbon
monoxide, infants in those
groups differed in birthweight
by an average of 16.2 grams.
The next-worst offender was
PM2.5, which showed a difference of 14.7 g, the scientists
report online and in an upcoming Environmental Health Perspectives.
These differences in birthweight can increase the newborn’s risk of complications such
as gastrointestinal infections and
respiratory problems in the first
weeks of life, comments
Srimathi Kannan of the University of Michigan in Ann Arbor.
However, only 4 percent of
the babies in the study met the
clinical standard for low birthweight—less than 2,500 g
(about 5.5 pounds)—which is
associated with life-threatening
complications in infancy and
heart disease in adulthood. A
woman’s risk of having a low-
birthweight baby increased by
no more than 5.4 percent when
she lived in a county at the
75 percent mark for air pollutants rather than in a county
at the 25 percent mark.
In arriving at these results,
the researchers adjusted for
many factors that can influence birthweight, such as prenatal care, gestational length,
type of delivery, and the child’s
sex and birth order. They also
considered the mother’s race,
education, marital status, age,
and tobacco use, all of which
have been shown to influence
the weights of newborns.
The new study is “much
more comprehensive in its
investigation” than previous
research, Kannan says, noting
that the biological mechanisms
linking these pollutants to
reduced fetal growth are still
poorly understood. —P. BARRY
