Focuses on a study which examined the responses of people to
personality cues in computer-generated voices. How people categorize
computer-generated voices; Role of the similarity-attraction principle in
the responses.
By
Courtney Bennett, published on January 01, 2002
SUBLIMINAL MESSAGES
We befriend, hire and fall in love with people who remind us of
ourselves. And apparently we decide whether or not to remain on hold with
the automated bank teller for the same reason. New research reveals that
people respond to the personality cues in computer-generated speech just
as they would to flesh-and-blood individuals: Extroverts prefer ebullient
computer voices, while introverts gravitate toward a quieter
cadence.
Clifford Nass, Ph.D., a professor of communications at Stanford
University, and postdoctoral student Kwan Min Lee found that people
categorize computer-generated voices based on cues such as loudness and
speech rate. The findings will be published in an upcoming issue of the
Journal of Experimental Psychology: Applied.
We are drawn to our robotic coevals by the "similarity-attraction
principle," a consistent pattern in social psychology. "We like things
that are similar to us," Nass explains. "And those feelings create a
'halo' effect that makes the things we like seem smarter and more
trustworthy." Even if they are 100 percent mechanical.
When it comes to advertising, consumers want the medium to match
the message. Seventy-two subjects who listened to extroverted and
introverted synthetic voices and read identical book reviews were more
likely to buy the book if the synthesized voice's personality matched
their own.
Finally, in a mock online auction Nass manipulated both the
synthetic voice's personality and the description of an item from Tiffany
& Company. "This is a reproduction of one of the most famous pieces
of Tiffany stained glass. The colors are absolutely sensational!" cooed
an outgoing, over-the-top voice. The introverted version demurely
introduced "a reproduction of a Tiffany stained-glass piece. The colors
are quite rich." The items were identical, but subjects were more likely
to buy a product when the voice and product description were in
sync.
Nass expresses concern about advertisers manipulating this
technology: "Just as you respond to other tricks that advertisers have up
their sleeve, being aware seems to be the best defense."
Tags:
bank teller,
clifford nass,
computer generated voices,
computer voices,
consistent pattern,
flesh and blood,
glass piece,
gravitate,
journal of experimental psychology,
online auction,
postdoctoral student,
speech rate,
stanford university,
subliminal messages,
synthetic voice,
synthetic voices,
tiffany company,
tiffany stained glass
