One Reason for Difficulty Understanding at Parties
Editor: We've all been there - trying to follow a conversation at a
noisy party and just not being able to do it! One crucial aspect of the
ability to follow such a conversation is the brain's control of auditory
attention. Here's what the researchers at Boston University want you to
know about this.
~~~~~~~~~~~~~~~~~
August 2008
Call it the cocktail party effect: how an individual can participate in
a one-on-one conversation within a cluster of people, switch to another,
pick up important comments while tuning out others, change topics and
return to the first conversation.
This selective switching of attention which relies on disengaging and
re-engaging attention to different voices on a time scale of a tenth of a
second, can pose challenges for anyone with normal hearing.
However, the same crowded scene presents far more problems for the
hearing impaired who have trouble listening to one sound and ignoring
others in everyday settings like a restaurant or in a business meeting.
They struggle to listen -- even with a hearing aid -- and are often
exhausted and frustrated by their efforts, unable keep pace with other
people who can tune out voices and more precisely pick out and stay with
one conversation.
A first step toward helping hearing impaired listeners requires a
better understanding of how people with normal hearing perceive and
process a mix of sounds over time. This auditory ability to switch
attention and, in the next instant, reset focus on whatever the new
speaker says is something about which little is known. The dynamics of
this process-- the time it takes to more precisely hear what is being said
in environments with competing sounds - has received little study in the
field of neuroscience. It is also the subject of a recently completed
research study entitled "Object continuity enhances selective auditory
attention," published online in the Proceedings of the National Academy of
Sciences (PNAS) on August 21, 2008.
The authors are Barbara Shinn-Cunningham, a Boston University professor
and Director of Graduate Studies in the Department of Cognitive and Neural
Systems, and graduate students Virginia Best, Erol J. Ozmeral and Norbert
Kopco.
The research team measured this complex acoustic scene by studying how
switching attention spatially influenced a listener's ability to recall a
sequence of spoken digits. Five loudspeakers were distributed horizontally
in front of the listener. The listener identified sequences of four digits
presented either from the same loudspeaker or from different ones chosen
randomly on each digit. Visual cues - lights - indicated the target
loudspeaker at each temporal position in the sequence. The remaining four
loudspeaker presented simultaneous distractor digits, the study noted.
The study also examined normal listeners' reactions when the target
voice changed from digit to digit as well as conditions in which the
target voice was the same.
The results showed that the recall of the spoken digits was best when
they all came from the same loudspeaker compared to hearing each number
from different speakers. Recall of the sequence degraded when listeners
had to instantaneously switch attention to a new location for each digit.
Thus a listener got better at filtering out sounds from others when they
focused attention to a voice at a fixed location.
Sustaining attention to one continuous auditory stream led to
refinements in selective attention over time. This improvement over time
depended on the perceived continuity of the stream of target digits-the
improvement was greatest when the digits sounded like they came from one
person talking from a fixed location. The progress was reduced when
different voices spoke each target digit and when delays between the
digits were abnormally long (so that each digit was perceived as an
isolated number). Researchers also measured how fast a listener would
switch or redirect their hearing the stream of digits - the finite time
required to disengage and then re-engage attention.
"These findings shed light on why, in listening environments such as
noisy parties or restaurants, it is more difficult to follow a
conversation involving many people (where the relevant talker often and
unexpectedly changes locations) than to focus on one talker (at one
location) exclusively," the study concludes. "In addition, these results
may have implications for visual attention in tasks where object formation
and target segmentation is challenging, or where the identity of a visual
object depends upon continuity of visual features over time."
Prof. Shinn-Cunningham will be continuing studies of how the brain
controls auditory attention in complex settings through a National
Security Science and Engineering Fellowship. The prestigious fellowship,
recently awarded to six individuals from an initial pool of over 500
applicants, will enable Shinn-Cunningham to undertake a five-year program
that uses both behavioral experiments and direct, non-invasive measures of
electrical brain activity to extend her studies of how attention enables
us to communicate in settings with multiple, competing sounds.