138
canal neuromasts that enabled the cryoablation of the lateral line system of midshipman females for the subsequent sound source localization experiments.
Coffin et al. (2014) observed no difference in the proportion of females exhibiting positive phonotaxis with cyroablated- (37 %) and sham-ablated (47 %) lateral
line systems, though complete ablation was not achieved because some of the mechanosensory neuromasts were incompletely damaged when the liquid nitrogenchilled probe was not properly applied. In addition, the authors reported that the
mean approach angle to the sound source was significantly different for females
with ablated lateral line systems (mean approach angle = 39°) compared to females
with sham-ablated lateral line systems (mean approach angle = 15°). Based on these
results, the authors suggested that the lateral line system is likely not required for
sound source localization, but it may be important for fine-tuning the approach to
the sound source.
3 Models of Directional Hearing and Sound Localization
Over the past five decades, a number of models have been proposed to account for
directional hearing in fish. In the mid-1960s, van Bergeijk (1964, 1967) proposed
one of the first influential such models, later known as the “acoustic-lateralis”
model. Van Bergeijk hypothesized that pressure sensitivity in the inner ear (via the
swim bladder or another gas bubble acting as a displacement-to-pressure transformer) allowed for sound source detection, but that mechanisms of the inner ear
alone were insufficient to account for source localization. He instead theorized that
the functionally similar lateral line system must supply directional information to
the auditory system via unspecified channels (Harris and van Bergeijk 1962; van
Bergeijk 1967). While some functional overlap of lateral line and auditory modalities has since been demonstrated and empirical measurements have not definitively
ruled out lateral line contributions to sound localization behavior, Van Bergeijk’s
hypothesis has largely fallen out of favor in recent years. Most especially, it has
become clear that the inner ear is intrinsically sensitive to source directionality, by
virtue of the accelerometer-like otoliths and underlying mechanically tuned hair
cells. In fact, saccular afferent particle displacement detection thresholds at low
frequencies can approach 0.1 nm (Fay 1984; Fay and Edds-Walton 1997).
During the 1970s, Schuijf and his colleagues established behavioral (conditioning) evidence that was consistent with sound localization by several species of fish
(Schuijf et al. 1977; Buwalda et al. 1983; Schuijf and Hawkins 1983). Fish could be
conditioned to discriminate between sources of sound that were spatially separated,
and from opposing directions. The investigators developed a complex mathematical/acoustical model that became known as the “phase model” to account for their
experimental data (e.g., Schuijf 1975; Schuijf and Buwalda 1975; Chapman and
Hawkins 1973). There are two essential components of the phase model. The first is
a determination of the axis of acoustic particle motion impinging on the ears, which
is thought to be accomplished through a process of resolving the axis of particle
J.A. Sisneros and P.H. Rogers
canal neuromasts that enabled the cryoablation of the lateral line system of midshipman females for the subsequent sound source localization experiments.
Coffin et al. (2014) observed no difference in the proportion of females exhibiting positive phonotaxis with cyroablated- (37 %) and sham-ablated (47 %) lateral
line systems, though complete ablation was not achieved because some of the mechanosensory neuromasts were incompletely damaged when the liquid nitrogenchilled probe was not properly applied. In addition, the authors reported that the
mean approach angle to the sound source was significantly different for females
with ablated lateral line systems (mean approach angle = 39°) compared to females
with sham-ablated lateral line systems (mean approach angle = 15°). Based on these
results, the authors suggested that the lateral line system is likely not required for
sound source localization, but it may be important for fine-tuning the approach to
the sound source.
3 Models of Directional Hearing and Sound Localization
Over the past five decades, a number of models have been proposed to account for
directional hearing in fish. In the mid-1960s, van Bergeijk (1964, 1967) proposed
one of the first influential such models, later known as the “acoustic-lateralis”
model. Van Bergeijk hypothesized that pressure sensitivity in the inner ear (via the
swim bladder or another gas bubble acting as a displacement-to-pressure transformer) allowed for sound source detection, but that mechanisms of the inner ear
alone were insufficient to account for source localization. He instead theorized that
the functionally similar lateral line system must supply directional information to
the auditory system via unspecified channels (Harris and van Bergeijk 1962; van
Bergeijk 1967). While some functional overlap of lateral line and auditory modalities has since been demonstrated and empirical measurements have not definitively
ruled out lateral line contributions to sound localization behavior, Van Bergeijk’s
hypothesis has largely fallen out of favor in recent years. Most especially, it has
become clear that the inner ear is intrinsically sensitive to source directionality, by
virtue of the accelerometer-like otoliths and underlying mechanically tuned hair
cells. In fact, saccular afferent particle displacement detection thresholds at low
frequencies can approach 0.1 nm (Fay 1984; Fay and Edds-Walton 1997).
During the 1970s, Schuijf and his colleagues established behavioral (conditioning) evidence that was consistent with sound localization by several species of fish
(Schuijf et al. 1977; Buwalda et al. 1983; Schuijf and Hawkins 1983). Fish could be
conditioned to discriminate between sources of sound that were spatially separated,
and from opposing directions. The investigators developed a complex mathematical/acoustical model that became known as the “phase model” to account for their
experimental data (e.g., Schuijf 1975; Schuijf and Buwalda 1975; Chapman and
Hawkins 1973). There are two essential components of the phase model. The first is
a determination of the axis of acoustic particle motion impinging on the ears, which
is thought to be accomplished through a process of resolving the axis of particle
J.A. Sisneros and P.H. Rogers