acoustic "colors": SUMMARY

Michael A Kisley kisley at blue.seas.upenn.edu
Mon Mar 27 18:49:19 EST 1995


This post is a general summary of the e-mail responses I received
regarding the topic of acoustic "colors".  I did not include anyone's
actual response (I just pooled them into a cohesive whole).  I am
only summarizing e-mails because all of the responses posted to the
newsgroups were about synesthesia.  Although the phenomenon of
synesthesia could be related to topic of my original post, I feel that
it is extremely tangential.  I apologize for not making my original
post more focussed.  The rest of this post consists of a repeat of the
original post, followed by the summary of responses.

*** original post *********************************************************

I am posting this message in hopes of receiving suggestions/references
regarding the general topic of auditory "color".  However, I would also
hope that this post might stir up some discussions.  This message is
cross-posted to the following newsgroups:

		bionet.neuroscience
		alt.sci.physics.acoustics
		bionet.audiology

While doing an introductory review of the topic of echolocation in bats,
I came across a very interesting idea in a 1990 review paper by G.
Neuweiler:  the idea of echo "colors" in bat echolocation.  The general
idea behind the use of the word "color" is that the bats might be
analyzing the spectral patterns of the targets which are reflecting
their broadband emissions.

In the general discussion of this review, Neuweiler goes on to hint that
perhaps this "color"-analysis is a general feature of auditory systems.
His reference to humans is that they might perceive acoustical "Gestalt"
during speech recognition and when listening to music.

A somewhat tangential but nevertheless related topic is the use of
sonar aids for blind people.  Apparently there were attempts to use
artificial sonar to help steer blind people around obstacles.  However,
someone I spoke with was under the impression that the sonar could also
be used to characterize textures of objects, not unlike the bats'
ability to characterize targets by their unique spectral pattern.

These fascinating topics have not been easy to trace, and I would appreciate
any help people can give me tracking down the following lines:

1)  Has any work been done since 1990 on the idea of echo "colors" in
    echolocating bats (or dolphins)?

2)  Does anyone know of any good, and recent, papers on the topic of
    general auditory "colors" or "Gestalt" (including the areas of
    speech recognition and music-listening)?

3)  Are the sonar aids for the blind still being pursued as a viable
    option, and are there any good references that discuss why/why not?

*** summary of responses ***************************************************

In general, there really has not been much research done into the
phenomena I mentioned.  Of the several responses I received, none
really addressed any of the three questions I laid out.  This is not
due to negligence on the part of the people reading the responses, but
rather due to the overall lack of knowledge and research about these
issues.  The most substantial topic of the responses was regarding the
use of the word "color" in this context.  Two points made (by two
different people) have convinced me that the word "color" is
inappropriate:  1) the tremendous differences between the processing
that occurs in the visual and auditory systems, 2) use of the term "color"
when talking about auditory signals is confusing.  The tremendous
difference between processing in the auditory and visual systems stems
ultimately from the fact that visual signals are mapped onto receptors
with 3 different frequency-responses, and auditory signals are mapped onto
receptors with several thousand different frequency-responses.  Thus,
the obvious difference in the "dimension" of the two types of signals
suggests difference types of processing in the CNS.  Therefore, it is
unlikely (but not impossible) that the auditory system represents a 
quality akin to

I received no responses to questions 2 and 3.  I received one response
addressing question 1 in a somewhat tangential manner.  However, since my
original post I have dug up a few references about the way
in which bats identify targets based on their fine structure.  The
following reference is a signal processing model of the phenomemnon:

Saillant, P.A. et al (1993) A computational model of echo processing and
acoustic imaging in frequency-modulated echolocating bats:  The
spectrogram correlational transformation receiver.  Journal of the
Acoustical Society of America, vol. 94: 2691-2712

Overall, I'd say that this area of research is about as poorly
defined as my original post is vague.  I suspect that several lines
of study will begin to converge, especially as the bat echolocation
research really narrows in on the way in which bat's identify targets
from their complex echoes.

Michael Kisley
kisley at eniac.seas.upenn.edu





More information about the Audiolog mailing list