Contact

Duke Health News

919-660-1306

When high-ranking monkeys are shown images of other monkeys
glancing one way or the other, they more readily follow the
gaze of other high-ranking monkeys, Duke University Medical
Center neurobiologists have discovered. By contrast, they tend
to ignore glance cues from low-status monkeys; while low-status
monkeys assiduously follow the gaze of all other monkeys.

The discovery represents more than a confirmation of what
most people believe about their bosses, said the researchers.
The findings reveal that gaze-following is more than a reflex
action; that it also involves lightning-fast social
perception.

Such a discovery in monkeys gives the researchers an
invaluable animal model that enables them to tease apart the
reflexive-versus-social mechanisms that govern behavior, they
said.

In particular, they can begin to understand the physiology
and neural machinery of status, they said. Further animal
studies will enable them to use drugs and genetic analysis to
figure out what hormonal and/or genetic influences determine
who becomes the monkey or human equivalent of Donald Trump, and
who becomes a Woody Allen.

The researchers -- graduate student Stephen Shepherd,
postdoctoral fellow Robert Deaner and Assistant Professor of
Neurobiology Michael Platt -- published their findings in the
Feb. 21, 2006, issue of Current Biology. The research was
supported by the Cure Autism Now Foundation and the National
Institute of Mental Health.

"By and large, most studies of gaze-following in humans
supported the idea that it was a reflexive attention
mechanism," said Platt. "People in those studies would tend to
shift their attention where they saw another person looking,
even if it wasn't predictive of some event happening around
them. And people didn't seem able to inhibit or control their
reaction." However, he said, there were hints that
gaze-following didn't have all the features of a purely
reflexive action, but these were only hints.

Such hints -- as well as previous studies in the Platt
laboratory -- led Shepherd and Platt to explore whether social
stimuli might also play a role in such decisions. Those
previous studies showed both that monkeys will follow the gaze
of other monkeys and that they will forego a juice reward to
look at high-status monkeys.

Said Shepherd, "It seemed reasonable to me that in the
natural environment monkeys would preferentially follow some
individuals' gaze and not others. High-status monkeys, for
example, do more to determine where the group is going to go.
So there's more information to be gleaned by finding out where
high-status individuals are looking. Also, it's fairly
important, if you're a low-ranking macaque, not to compete with
a high-ranking individual, so you want to know where they're
paying attention."

In the experiments, Shepherd showed macaque monkeys images
of monkeys known to be of higher or lower status than
themselves. The images depicted the monkeys looking left or
right. Immediately after each image, a target was flashed onto
the screen, randomly in the direction the monkey image was
looking or in the opposite direction. The monkeys were given
juice rewards for their participation in each trial.

After a large number of trials, the researchers
statistically analyzed whether status played a role in the
monkeys' tendency to follow the gaze on the screen. They found
that the high-status monkeys were significantly more likely to
follow the gaze of other high-status monkeys than low-status
monkeys; while the low-status monkeys tended to follow the gaze
of all the other monkeys.

However, noted, Shepherd and Platt, it was entirely possible
that low-ranking monkeys might be too anxious at seeing images
of high-ranking images, and would avoid eye contract
altogether.

"But our results were pretty striking," said Shepherd.
"Low-ranking macaques are extremely fast to follow gaze, while
the high-ranking monkeys were pretty blasé about it, being
slower to respond."

Said Platt, "So, now we have an excellent model of how
temperament or status can modulate the strength of these two
seemingly independent attention systems -- cognitive and
reflexive -- in the brain. We can begin to trace the neural
pathways by which social information feeds into the structures
that control the eyes. And, we can explore whether such
influences as hormonal levels, particularly testosterone,
influence ranking. For example, we can manipulate testosterone
levels, or give anxiety-reducing drugs, to determine an effect
on social status, using gaze-following as a measure."

The neurobiologists' basic studies could also have
application to understanding the origins of autism, said Platt.
One theory, for example, holds that high levels of testosterone
in utero cause "hypermasculinization" of the brain, which
suppresses the reflexive ability to orient socially -- a
characteristic of autism, he noted. Also, he said, such studies
could aid understanding a wide range of disorders such as
social anxiety.

More broadly, said Shepherd, such studies in monkeys will
enable greater insight into the basic machinery of social
interaction.

"Thanks to a combination of molecular and behavioral
studies, we're starting to be able to investigate the neural
machinery that allows humans to empathize, to form strong
social bonds, to do things like share food and to cooperate,"
he said. "Besides suggesting ways of diagnosing or assisting
people with autism and other disorders, such studies are also a
means of understanding what enables us to be social."