173
zebrafi sh mounted on a shaker system (Fig. 5 ). After determining the baseline startle response probability, a set of acoustic stimuli ranging from 90 to 1200 Hz at
sub-startle threshold levels were used as prepulse stimuli. Startle response probabilities were measured, and the change in probability of evoking the ASR was quantifi ed. Auditory threshold was determined as the prepulse sound level that effectively
reduced the ASR probability by 5 % from the baseline response probability. Using
this method, the authors were able to show that auditory thresholds in 5 dpf zebrafish were 10–15 dB (re. 1 m/s
2
) lower than was previously shown by just using an
ASR assay (Fig. 6 ).
The PPI assay has similar limitations to ASR assays. Habituation to PPI can be
reduced, but it still imposes an upper limit on the number of stimulus presentations
that can be effectively performed. The use of PPI also requires the presence of the
M-cell circuitry and the associated PHP cells, which are known to occur in zebrafi sh
and goldfi sh (Neumeister et al. 2008 ; Medan and Preuss 2011 ). Future work in this
fi eld should investigate whether the PPI of the startle response can be implemented
to study of auditory function in species that lack Mauthner cells.
PPI is potentially a powerful tool to study ontogenetic changes in auditory sensitivity, particularly in larval fi shes. In this context, electrophysiological methods can
be too invasive, conditioning methods can be too time-consuming, and ASR assays
are not sensitive enough to study the auditory capabilities of larval fi shes. PPI assays
are sensitive enough to determine changes in auditory sensitivity of the animal during development and the same animals can be tested again at different stages of
development. Furthermore, the acquired threshold estimates can be just as or more
sensitive than those derived from electrophysiological methods. Studies using the
Mongolian gerbil ( Meriones unguiculatus ) have shown that auditory thresholds
determined by PPI are equivalent to thresholds determined using an ABR approach
(Walter et al. 2012 ). Future work on PPI will inform us whether this fi nding can be
generalized in larval and adult teleosts.
3.2 Phonotaxis
Phonotaxis is a refl exive locomotor response toward (positive) or away from (negative) an external auditory stimulus. This innate response is associated with a biologically relevant function, and as a result can only be elicited with a specifi c
stimulus. Stimuli that “release” phonotaxis can be biogenic (in the case of conspecifi c advertisement calls) or produced by physical features of the environment (e.g.,
reef sounds that attract pelagic fi sh larvae; Pijanowski et al. 2011 ). Such acoustic
“sign” stimuli are known to have unique spectral and temporal features (Amorim
2006 ) that are strongly stereotyped, complex, and behaviorally relevant to the animal. Many fi sh species have inherent behaviors that are evoked by specifi c acoustic
stimuli including acoustic signals used for communication. Courtship and agonistic/
distress sounds have been recorded in mormyrid fi sh ( Gnathonemus petersii ; Rigley
and Marshall 1973 ), cod ( Gadus morhua ; Hawkins and Chapman 1966 ), toadfi shes
Revisiting Psychoacoustic Methods for the Assessment of Fish Hearing
zebrafi sh mounted on a shaker system (Fig. 5 ). After determining the baseline startle response probability, a set of acoustic stimuli ranging from 90 to 1200 Hz at
sub-startle threshold levels were used as prepulse stimuli. Startle response probabilities were measured, and the change in probability of evoking the ASR was quantifi ed. Auditory threshold was determined as the prepulse sound level that effectively
reduced the ASR probability by 5 % from the baseline response probability. Using
this method, the authors were able to show that auditory thresholds in 5 dpf zebrafish were 10–15 dB (re. 1 m/s
2
) lower than was previously shown by just using an
ASR assay (Fig. 6 ).
The PPI assay has similar limitations to ASR assays. Habituation to PPI can be
reduced, but it still imposes an upper limit on the number of stimulus presentations
that can be effectively performed. The use of PPI also requires the presence of the
M-cell circuitry and the associated PHP cells, which are known to occur in zebrafi sh
and goldfi sh (Neumeister et al. 2008 ; Medan and Preuss 2011 ). Future work in this
fi eld should investigate whether the PPI of the startle response can be implemented
to study of auditory function in species that lack Mauthner cells.
PPI is potentially a powerful tool to study ontogenetic changes in auditory sensitivity, particularly in larval fi shes. In this context, electrophysiological methods can
be too invasive, conditioning methods can be too time-consuming, and ASR assays
are not sensitive enough to study the auditory capabilities of larval fi shes. PPI assays
are sensitive enough to determine changes in auditory sensitivity of the animal during development and the same animals can be tested again at different stages of
development. Furthermore, the acquired threshold estimates can be just as or more
sensitive than those derived from electrophysiological methods. Studies using the
Mongolian gerbil ( Meriones unguiculatus ) have shown that auditory thresholds
determined by PPI are equivalent to thresholds determined using an ABR approach
(Walter et al. 2012 ). Future work on PPI will inform us whether this fi nding can be
generalized in larval and adult teleosts.
3.2 Phonotaxis
Phonotaxis is a refl exive locomotor response toward (positive) or away from (negative) an external auditory stimulus. This innate response is associated with a biologically relevant function, and as a result can only be elicited with a specifi c
stimulus. Stimuli that “release” phonotaxis can be biogenic (in the case of conspecifi c advertisement calls) or produced by physical features of the environment (e.g.,
reef sounds that attract pelagic fi sh larvae; Pijanowski et al. 2011 ). Such acoustic
“sign” stimuli are known to have unique spectral and temporal features (Amorim
2006 ) that are strongly stereotyped, complex, and behaviorally relevant to the animal. Many fi sh species have inherent behaviors that are evoked by specifi c acoustic
stimuli including acoustic signals used for communication. Courtship and agonistic/
distress sounds have been recorded in mormyrid fi sh ( Gnathonemus petersii ; Rigley
and Marshall 1973 ), cod ( Gadus morhua ; Hawkins and Chapman 1966 ), toadfi shes
Revisiting Psychoacoustic Methods for the Assessment of Fish Hearing