Dear Eckard Blumschein -
I am glad you enjoyed my website, and quite thankful
that you directed me to the Magdeburg message board. There were many interesting
discussions that I look forward to reading in depth over the coming weeks.
Before I address your comments I think it would
be appropriate to qualify my remarks by saying that my lab is a direct
perception/Gibsonian adherent. While personally I have my doubts about
the sufficiency of this approach, as a graduate student I worked within
the assumptions of this theory for the bulk of my research. You joked about
your ideas being
heretical, in many ways that description sums the entirety of the direct
approach.
So, from that perspective I approach echolocation
and all auditory perception as being a means to detect information for
successful behavior. This suggests that auditory perception is best
studied in relation to a particular event (e.g., walking to a surface,
determining the content of speech, etc.).
Returning to the mechanisms of echolocation my theoretical
perspective drives me to look for the environmental parameters that would
support echolocation directly - the vergence angle formed between the sound
emitter, the reflective surface, and the perceiver's ears (Lee et al.,
1991; 1992); the spectral changes caused by the sound reflecting
back on itself to the emitter (Schenkman, 1985). To the extent that the
physiological mechanisms detect these sorts of invariant properties that
have been found to be part of echolocation I agree they exist. Psychophysically
there seems to be very clear evidence interaural transients as evidenced
by the localization research so I certainly agree with you J. Manger.
You mentioned as one of the differences between
humans and bats that only bats rely on echolocation for survival. For normal,
sighted humans that is of course true. However our lab has worked with
blind individuals who use echolocation as their mainstay of navigation.
Furthermore I think that auditory perception may be an important enhancement
to visual perception (in a variety of scene analysis behaviors including
echolocation) throughout our daily interaction with the world.
I feel I have rambled on, but to summarize I agree
with you that there are differences in the salience of echolocation between
humans and bats that do separate how we should analyze their perceptual
mechanisms. However, I do
not think that on the whole echolocation differs across species - I
think the greatest difference is one of experience. The physical constraints
of echolocation force there to be a limited number of sources from which
the animal can derive distance judgments using reflected sound.
Would you expect large scale physiological differences
between bats and highly trained human echolocators? Would you expect the
auditory cortex of a bat to have response patterns to echolocation entirely
different than that of humans?
Mike
__________________________________
Eckard Blumschein replied:
What about the Gibsonian point of view, my knowledge is limited to what Richard Warren described in his 1999 book. Nonetheless, I feel it is a valuable approach that is not biased by the intention to belittle obvious discrepancy between questionable mathematical bases and observations of rather limited scope. Of course, the ultimate understanding requires experts of acoustics as well as physiology. With respect to the latter the I refer to the book "Hearing by bats". However, I am not sure whether or not your ray geometry is already sufficient. Acoustics also deserves attention. The two ears of a bat may act like a beam former. Frequencies around 100 kHz correspond to wave length as short as about 3 mm. You have to check the possibility of whispering gallery modes around larger obstacles. Furthermore, short acoustic waves are subject to strong attenuation. I am aware of extensive scientific investigation into echolocation.
So my own guesswork is rather primitive and possibly wrong. As written
by Koessl and Vater, Eptesicus is reported to have a two-glint resolution
as smaller than 10 microseconds or 1.7 mm on target. In the two CF-FM bat
species Pteronotus parnellii and Rhinolophus rouxi the maximum frequency
expansion was found to amount 40mmBM/octave at roughly 60 and 80 kHz, respectively.
Remember, normally, about 2.5 mm of BM length/octave are used.
I imagine the distance on a two-glint target represented in the positions
of two corresponding peaks of an epiphenomenal traveling wave on partition
of cochlea. Given the delay amounts just 1 microsecond. What distance on
BM corresponds to that delay? 1/80kHz equals 12.5 microseconds. 1/87kHz
roughly equals 11.5 microseconds. 87/80 = 1.088. 40mm times 0.088 = 3.5
mm. Total length of BM is roughly 15mm. This does not contradict Simmons
who reported measured binaural acuity within the nanosecond range. I guess,
the bottleneck is the neural processing of these transient profiles of
firing along partition.
There are many quantitative physiological differences between bats and other mammals. Maximal length of OHCs and their cilia is reported smaller for bats than minimal length measured in guinea pigs. Width of BM is nearly constant in Pteronotus while it changes by a factor near to 4 in cat. The maximal ratio of afferent fibers to IHCs amounts to 34-37 in Pteronotus and even 70 in the FM bat Myotis. Cochlea of several bats exhibits a fovea. Spatial orientation by the blind is quite different from echolocation by bats in that, the call of the latter has an exceptional high amplitude. High frequency increases directivity. For further details you might look into literature or consult experts.
What about cortical processes, I guess, human-like perception would be too sluggish. Echolocating bats have to react within a few milliseconds. They must not have a long reaction period.
Eckard