First published online March 8, 2005
Journal of Experimental Biology 208, 915-927 (2005)
Published by The Company of Biologists 2005
doi: 10.1242/jeb.01452
Mechanisms underlying phonotactic steering in the cricket Gryllus bimaculatus revealed with a fast trackball system
B. Hedwig* and
J. F. A. Poulet
University of Cambridge, Department of Zoology, Downing Street,
Cambridge CB2 3EJ, UK
View larger version (21K):
[in a new window]
|
Fig. 1. (A) Experimental arrangement for measuring cricket phonotaxis. The animal
was fixed on top of a trackball and during walking rotated the trackball with
its legs. An optical sensor aligned below the trackball (not shown) monitored
the forward–backward and left–right trackball movements. Speakers
at a distance of 87 cm were arranged at 45 deg off the animal's length axis.
(B) Simultaneous recording of the forward–backward and left–right
movement component of a walking cricket, as picked up from the rotating
trackball. From these two components, the translational velocity of the animal
was calculated as well as the lateral deviation of the path and the total path
length. (C) The step cycle of a foreleg recorded with an optoelectronic camera
(Hedwig, 2000 ) and the
translation velocity of a walking cricket. Due to the insect tripod gate the
movement of the foreleg was in phase with every other peak in the velocity
signal.
|
|
View larger version (36K):
[in a new window]
|
Fig. 2. Intensity function of phonotactic behaviour. (A) Phonotactic orientation of
a cricket towards an optimal sound pattern presented from the left and right
for 30 s at increasing intensities from 45 to 85 dB SPL. For each sound
intensity the lateral deviation (top) is plotted over time. At the start of
each trial the lateral deviation is reset to zero. The lateral steering
velocity (middle) oscillated around zero at low sound intensities. At high
sound intensities oscillations in the velocity signal were still evident, but
the signal was shifted towards the side of acoustic stimulation. The
translation velocity (lower trace) exhibited a broad variation in amplitude
due to the temporal resolution of individual velocity peaks underlying
walking. (B) Intensity dependence of the behaviour. With sound intensity
increasing from 45 to 75 dB SPL the overall lateral deviation from a straight
line increased 6.1-fold. The overall path length walked increased 1.7-fold
with increasing sound intensity as calculated for the 60 s trial.
|
|
View larger version (24K):
[in a new window]
|
Fig. 3. High-resolution recording of an animal steering towards the left speaker,
which presented a sound pattern at 75 dB. (A) Each chirp released a steering
velocity transient to the left whereas the forward velocity was only weakly
modulated by the stimulus pattern. Each steering transient caused a shift of
the animal's path to the left. (B) Averages of the velocity components
demonstrate the strong modulation of the steering velocity in the chirp rhythm
and the weak modulation of the forward velocity. As a consequence, the
animal's deviation to the left was modulated by the sound pattern.
|
|
View larger version (36K):
[in a new window]
|
Fig. 4. Intensity differences and steering responses. (A) A reference sound pattern
of 60 dB SPL was presented simultaneously with a test pattern of equal or
higher amplitude. Intensity differences are indicated above the top trace.
Lateral deviation towards the louder sound pattern was minute at 1 dB and
increased with increasing intensity difference. Oscillations in the lateral
steering velocity occurred during each test together with oscillations in the
translational velocity. (B) Lateral deviation as a function of the intensity
difference is presented. The lateral deviation is expressed as a percentage of
the steering towards a 75 dB reference pattern presented from one side only.
(C) Average steering response of a cricket when exposed to intensity
differences of 1, 2 and 10 dB. (D) Overall path length walked during the
different tests.
|
|
View larger version (30K):
[in a new window]
|
Fig. 5. Latency differences and steering responses. (A) Two identical sound
patterns of 75 dB SPL were presented with increasing delay. Latency
differences are indicated above the top trace. Lateral deviation towards the
leading sound pattern was negative at 0 and 2 ms. It was significant at 4 ms
and weak thereafter. Oscillations in the lateral steering velocity during each
trial and oscillations in the translation velocity are given below. (B)
Lateral deviation in dependence of the latency difference presented. The
maximum response occurred at 4 ms difference. The lateral deviation is
expressed as percentage of the steering towards a 75 dB reference pattern
presented from one side only. (C) Average lateral steering response of a
cricket when exposed to latency differences of 4 and 10 ms did not indicate an
obvious response towards the leading sound pattern.
|
|
View larger version (23K):
[in a new window]
|
Fig. 6. Bilateral pulse numbers and steering. (A) Patterns with different numbers
of sound pulses were presented from the left and right side. The numbers of
pulses presented from the first and second speaker varied from 6:0 to 3:3 to
0:6. Lateral deviation depended on the number of sound pulses presented from
each side. When the same number of pulses was presented from both sides, the
animals almost walked straight ahead. (B) Lateral deviation in dependence of
ratio of sound pulses presented from the left and right speaker. The lateral
deviation depended on the pulse numbers perceived at each side. (C) Overall
path length walked during the trials. Translation velocity increased with the
amount of steering.
|
|
View larger version (25K):
[in a new window]
|
Fig. 7. Averages of lateral steering velocity in response to different numbers of
sound pulses presented by the leading and following speaker. (A) With six
pulses presented from the left side, steering occurred towards the side of
sound stimulation. (B) When the last pulse was presented from the right side,
the animal started to turn towards the right side in response to the single
sound pulse. (C–E) With increasing numbers of sound pulses presented
from the right side, the animals steered stronger to the right side. (E) The
cricket started to steer towards the single pulse presented from the left side
and then turned towards the right side presenting five pulses. For all tests
the steering response to six pulses from the left is given as a reference (in
blue). Number of trails 60.
|
|
View larger version (23K):
[in a new window]
|
Fig. 8. Steering response to extreme split-song pattern. (A) The crickets were
exposed to a split-song paradigm with every other sound pulse presented from
opposite sides. The animals walked straight ahead or slightly deviated towards
the leading pattern. The steering velocity oscillated around zero. (B) A plot
at high temporal resolution demonstrates, that each sound pulse elicited a
rapid steering response directed towards the corresponding speaker. (C)
Superposition of consecutive steering responses during split-song chirps. (D)
Average of the steering velocity when the crickets were exposed to the
split-song paradigm. The steering responses towards single sound pulses
occurred with a latency of about 55–60 ms and were initiated just at the
end of the consecutive sound pulse.
|
|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati 
Twitter What's this?
© The Company of Biologists Ltd 2005
