133
man. The sound level at the release site was adjusted each night before behavioral
tests to 130 dB re 1 μPa, which is consistent with sound levels that nesting type I
males produce within or near their nests (Bass and Clark 2003). The acoustic
pressure gradients of the sound field were measured directly using an eight minihydrophone array that formed a cube, 5 cm × 5 cm on each side (Fig. 2). This
arrangement of the hydrophones permitted particle motion to be calculated in the x,
y, and z directions from the pressure gradient measurements between adjacent
hydrophones, which were then used to create a map of the sound fields produced by
the J9 sound projector in terms of sound pressure and particle motion (Fig. 3).
Behavioral tests occurred at night after sunset in the large outdoor testing tank.
Female midshipman that were used in these experiments were collected by hand
from the intertidal zone where midshipman nest and spawn during the summer reproductive season. All animals were collected during the morning low tides and then
were tested later at night on the same day of collection. The behavioral tests began
with an individual gravid female being placed in a 30 cm diameter plastic mesh cylinder that was positioned 109 cm away from the sound projector. Fish were then
released from the cylinder while the J9 sound projector was continuously playing the
90 Hz stimulus tone. The phonotaxic responses of the gravid females consisted primarily of straight to slightly curved paths to the monopole sound source. Once at the
sound source, females would then precede to either directly touch the face and/or
underside of the sound projector. The majority (73 %) of the tested gravid females
showed robust phonotaxis and localization of the monopole sound source (Fig. 4). In
contrast, none of the gravid females in the control (sound-off) group localized the
projector or made physical contact with it. The study by Zeddies et al. (2010) was
significant because it was the first to compare the paths fish take to a sound source
with a description of the available directional cues in the form of local particle motion
vectors. Zeddies et al. confirmed that gravid females exhibit highly directional pho-150
150
150
-100 -50
a
b
0
-150
-100
-50
0
-150
-100
-50
0
50 100
cm
-150 -100 -50
0
50 100
cm
Fig. 4 Response pathways of the test (a, sound playback of a simulated midshipman advertisement call) and control (b, no sound) to naïve female plainfin midshipman. The phonotaxic
responses of gravid females in the test (sound on) group displayed primarily straight to slightly
curved paths to the sound source. Adapted from Zeddies et al. (2010)
Directional Hearing and Sound Source Localization in Fishes
man. The sound level at the release site was adjusted each night before behavioral
tests to 130 dB re 1 μPa, which is consistent with sound levels that nesting type I
males produce within or near their nests (Bass and Clark 2003). The acoustic
pressure gradients of the sound field were measured directly using an eight minihydrophone array that formed a cube, 5 cm × 5 cm on each side (Fig. 2). This
arrangement of the hydrophones permitted particle motion to be calculated in the x,
y, and z directions from the pressure gradient measurements between adjacent
hydrophones, which were then used to create a map of the sound fields produced by
the J9 sound projector in terms of sound pressure and particle motion (Fig. 3).
Behavioral tests occurred at night after sunset in the large outdoor testing tank.
Female midshipman that were used in these experiments were collected by hand
from the intertidal zone where midshipman nest and spawn during the summer reproductive season. All animals were collected during the morning low tides and then
were tested later at night on the same day of collection. The behavioral tests began
with an individual gravid female being placed in a 30 cm diameter plastic mesh cylinder that was positioned 109 cm away from the sound projector. Fish were then
released from the cylinder while the J9 sound projector was continuously playing the
90 Hz stimulus tone. The phonotaxic responses of the gravid females consisted primarily of straight to slightly curved paths to the monopole sound source. Once at the
sound source, females would then precede to either directly touch the face and/or
underside of the sound projector. The majority (73 %) of the tested gravid females
showed robust phonotaxis and localization of the monopole sound source (Fig. 4). In
contrast, none of the gravid females in the control (sound-off) group localized the
projector or made physical contact with it. The study by Zeddies et al. (2010) was
significant because it was the first to compare the paths fish take to a sound source
with a description of the available directional cues in the form of local particle motion
vectors. Zeddies et al. confirmed that gravid females exhibit highly directional pho-150
150
150
-100 -50
a
b
0
-150
-100
-50
0
-150
-100
-50
0
50 100
cm
-150 -100 -50
0
50 100
cm
Fig. 4 Response pathways of the test (a, sound playback of a simulated midshipman advertisement call) and control (b, no sound) to naïve female plainfin midshipman. The phonotaxic
responses of gravid females in the test (sound on) group displayed primarily straight to slightly
curved paths to the sound source. Adapted from Zeddies et al. (2010)
Directional Hearing and Sound Source Localization in Fishes