Currently, there is controversy around whether rats can use interaural time differences (ITDs) to localize sound. Here, naturalistic pulse train stimuli were used to evaluate the rat's sensitivity to onset and ongoing ITDs using a two-alternative forced choice sound lateralization task. Pulse rates between 50 Hz and 4.8 kHz with rectangular or Hanning windows were delivered with ITDs between ±175 μs over a near-field acoustic setup. Similar to other mammals, rats performed with 75% accuracy at ∼50 μs ITD, demonstrating that rats are highly sensitive to envelope ITDs.

Binaural cues are important for sound localization. The two primary binaural cues are differences in sound level and arrival time between two ears, known as interaural level differences (ILDs), and interaural time differences (ITDs), respectively. The species most commonly used in binaural hearing research, including gerbils (Lingner et al., 2012; Tolnai et al., 2017), ferrets (Keating et al., 2013), cats (Brugge et al., 2001), and guinea pigs (Greene et al., 2018), all show sensitivity to both ILDs and ITDs.

Rats are relatively rarely used in studies of binaural hearing due to a reputation for relatively poor sound localization abilities (Kavanagh and Kelly, 1986; Heffner and Heffner, 1985) and controversy around the extent to which they can use ITDs to localize sound. While neurophysiological evidence for ITD sensitivity in rats has been documented in the literature (Kidd and Kelly, 1996), a recent behavioral study by Wesolek et al. (2010) used low frequency (0.5 to 2 kHz) pure tone free-field stimuli to conclude that rats were unable to use ITDs to localize sound. However, the restricted choice of stimuli did not allow the authors to fully explore the rats' sensitivity to envelope ITDs, which feature prominently in many natural sounds, and which can in principle provide both transient and ongoing ITD information over the entire frequency range (Bernstein, 2001).

Here we evaluated the rat's ability to use ITDs to localize pulse-resonance sounds, an important class of broadband sounds that also comprises communication calls used by humans and many other animal species (Patterson, 2015). These stimuli provided both transient onset and ongoing envelope ITD information across a wide frequency range. By comparing responses to pulse trains with either sharp onset rectangular windows or with very gradual Hanning windows, we were able to assess the rats' sensitivity to onset and ongoing envelope ITDs at different pulse rates.

Five female Wistar rats (220–260 g, two months old at the beginning of training) were used in this study. The rats' hearing thresholds were confirmed to be in the normal range by recording auditory brainstem responses (ABRs) under anesthesia [ketamine (80 mg/kg) and xylazine (12 mg/kg)], and their tympanic membranes and outer ear canals were inspected to confirm the absence of obstruction or outer or middle ear disease. All experimental procedures were approved by the Committee on the Use and Care of Animals at City University of Hong Kong, and under license by the Department of Health of Hong Kong [Ref. No. (16–86) in DH/HA&P/8/2/5 Pt.5].

Stimuli consisted of single sample pulse (delta function “click”) trains of 200 ms duration generated at a sample rate of 44 100 Hz and pulse rates of 50, 300, 900, 1800, or 4800 Hz. Click trains were presented at an average binaural acoustic level of 80–85 dB sound pressure level. To investigate the contributions of onset and ongoing ITD cues, click trains were enveloped with either rectangular or Hanning windows of 200 ms duration [Fig. 1(A)]. To produce ITDs of ±175 μs [∼130% of the rat's physiological range of ±130 μs (Koka et al., 2008)], identical stimulus pulse trains were presented to each ear, with the stimulus in one ear delayed relative to the other by an appropriate number of samples (22.7 μs steps). Negative ITD values are when the stimulus in the left ear leads relative to that in the right.

Fig. 1.

(Color online) Rat binaural psychoacoustics near-field setup. (C) Example waveforms of 300 Hz rectangular window (top) and Hanning window (bottom) pulse trains used. (A) Rat during a testing session, initiating a trial by making contact with the central “start” spout. Steel tube phones are positioned close to each ear. (B) 3D printed rat acoustic manikin with miniature microphones in each ear canal used for calibrating and validating the setup. (D) Waveforms of a single binaural pulse as recorded from the microphones inside each ear canal of the acoustic manikin (L: left ear, R: right ear) for ±100 μs ITD (top and bottom pair of traces, respectively). The pulse train stimuli shown in (C) are made from sequences of pulses like the one shown. Cross correlation functions of the recorded signals confirmed what is also apparent by inspection of the waveforms, namely, that the ITDs of the acoustically recorded signals corresponded precisely to the electric interaural delays. (E) Frequency spectra of the sound waveforms in (D) for ±100 μs ITD. (F) Acoustic ILDs (y axes) measured from microphones placed in the ear canals of the acoustic manikin for ±100 μs ITD. ILDs were computed as the difference in root mean square (RMS) power of the signals in (D). Waveforms were recorded for ten pulses at each ITD. Each dot therefore represents one trial. A small random x axis scatter was added for visualization. Dotted lines indicate average broadband ILD thresholds of ferrets (Keating et al., 2013) for comparison. Behavioral rat ILD thresholds have not been reported in the literature, but are estimated to be around 2.8–4 dB (Greene et al., 2018).

Fig. 1.

(Color online) Rat binaural psychoacoustics near-field setup. (C) Example waveforms of 300 Hz rectangular window (top) and Hanning window (bottom) pulse trains used. (A) Rat during a testing session, initiating a trial by making contact with the central “start” spout. Steel tube phones are positioned close to each ear. (B) 3D printed rat acoustic manikin with miniature microphones in each ear canal used for calibrating and validating the setup. (D) Waveforms of a single binaural pulse as recorded from the microphones inside each ear canal of the acoustic manikin (L: left ear, R: right ear) for ±100 μs ITD (top and bottom pair of traces, respectively). The pulse train stimuli shown in (C) are made from sequences of pulses like the one shown. Cross correlation functions of the recorded signals confirmed what is also apparent by inspection of the waveforms, namely, that the ITDs of the acoustically recorded signals corresponded precisely to the electric interaural delays. (E) Frequency spectra of the sound waveforms in (D) for ±100 μs ITD. (F) Acoustic ILDs (y axes) measured from microphones placed in the ear canals of the acoustic manikin for ±100 μs ITD. ILDs were computed as the difference in root mean square (RMS) power of the signals in (D). Waveforms were recorded for ten pulses at each ITD. Each dot therefore represents one trial. A small random x axis scatter was added for visualization. Dotted lines indicate average broadband ILD thresholds of ferrets (Keating et al., 2013) for comparison. Behavioral rat ILD thresholds have not been reported in the literature, but are estimated to be around 2.8–4 dB (Greene et al., 2018).

Close modal

Signals were generated on a Raspberry Pi 3 computer and sent through a USB sound card (StarTech.com, Ontario Canada, part No. ICUSBAUDIOMH) running at 16 bit, and amplifier (Adafruit stereo 3.7 W class D audio amplifier, part No. 987) to miniature high fidelity headphone drivers (GQ-30783‐000, Knowles, Itasca, Illinois, US), which were mounted to hollow stainless steel tubes for sound delivery. The pulses resonated in the tubes to produce pulse-resonant sounds resembling single-formant artificial vowels, with a fundamental frequency corresponding to the click rate. These tube phones were held in place by custom 3D printed ball-in-socket joints and were positioned such that, when the rat started a trial by licking a center start spout, the tips of the tubes were located right next to each ear, allowing for near-field stimulation [Fig. 1(B)]. Note that this mode of sound delivery is similar to that produced by “open” headphones, such as those commonly used in previous studies on binaural hearing in humans and animals, e.g., Keating et al. (2013).

The acoustic setup was validated using a custom, 3D printed acoustical manikin rat head with miniature microphones in each ear canal [Fig. 1(C)]. Waveforms for a single binaural pulse recorded at each ear are shown in Fig. 1(D), and their frequency spectra are shown in Fig. 1(E). As might be expected, Fig. 1(E) reveals interaural spectral differences at some frequencies due to manufacturing tolerances, but these do not result in systematically varying ILDs, and Fig. 1(F) confirms that any residual ILDs were negligible and did not covary with ITD conditions. The crosstalk between the ears over the full 500 Hz to 20 kHz signal range was −20 dB. This setup therefore provided robust and naturalistic ITDs but no useable ILD cues to the animals.

Rats were trained to perform a two-alternative forced choice sound lateralization task using established methods (Itskov et al., 2012; Keating et al., 2013). Rats were put on a schedule of 5 days of testing, during which drinking water was a positive reinforcer, followed by two days off with ad lib water. Drinking bottles were removed ∼16 h prior to the next testing period. Rats performed two sessions per day, each lasting ∼30 min, corresponding to ∼200 trials and ∼15 ml of water consumed.

One of the walls of a perspex cage was fitted with three brass water spouts, mounted ∼7 cm from the floor and separated by ∼6 cm [Fig. 1(B)]. Contact with the spouts was detected by touch detectors (Adafruit industries, USA, part No. 1362). Initiating a trial at the center spout triggered the release of one small drop of water through a solenoid valve on a fraction (1/7) of trials, followed by presentation of the sound stimulus. Correct lateralization of the stimulus by licking the left or right response spouts triggered four small drops of water as positive reinforcement. Incorrect responses triggered no water delivery, a negative feedback sound, and a 15–30 s timeout during which no new trial could be initiated. After an incorrect trial, the same stimulus was repeated as a “correction trial” to prevent animals from developing idiosyncratic biases favoring one side (Keating et al., 2013). Correction trials were excluded from analysis. Rats were initially trained to lateralize 300 Hz trains containing both ILDs of ±6 dB and ITDs of ±175 μs. Once they performed at ≥80% correct, ILD cues were removed, and variable ITDs were gradually introduced. Training to a high level of performance with variable ITD-only stimuli took between 14 and 20 days (28–40 sessions).

When sensory signals are large relative to noise, the task is easy and performance will be consistent, but when sensory signals are close to threshold, performance will be near chance. This relationship is captured by the sigmoidal cumulative Gaussian function Φ (Schnupp et al., 2005) which we fitted to our data to quantify ITD sensitivity,

pR=Φ(ITDα),
(1)

where pR denotes the probability of choosing the right (R) spout, ITD denotes the interaural time difference (positive if the right ear leads, in ms), and α is the ITD sensitivity parameter with units of 1/ms that captures the change in the proportion of R responses a given change in ITD would induce.

This model was extended to account for possible additive lapses of attention and idiosyncratic biases towards one ear or one spout. The γ term denotes the lapse rate and it compresses the range of the psychometric sigmoid to [γ/2, 1 – γ/2], which is equivalent to scaling by 1 − γ and shifting by γ/2. An ear bias exists if chance (50%) performance occurs at an ITD value some small value β away from zero. The parameter δ captures the increased probability of choosing the R spout due to an idiosyncratic preference. The extended model is

pR=Φ(ITDα+β)(1γ)+γ2+δ
(2)

and maximum likelihood estimates for the parameters α, β, γ, and δ were derived from the data using gradient descent [scipy.optimize minimize() (Jones et al., 2001)].

For stimuli that the animals could only lateralize with difficulty or not at all (i.e., ITD sensitivity α is close to zero) the parameters of the sigmoid model become poorly constrained by the data. Therefore, two alternative models were additionally fitted: a null model that assumes that α = 0 and the rate of R responses is simply a constant,

pR=0.5+δ
(3)

and a linear model that assumes α > 0 but does not fit a sigmoid because the proportion of R responses does not asymptote over the tested range of ITDs,

pR=ITDα+δ.
(4)

A χ2 deviance test was used to choose the best model from Eqs. (2), (3), and (4) for each condition, and ITD sensitivity was in all cases defined as the slope of the modeled psychometric function around zero. ITD sensitivity was either zero if the null model gave the best fit, α for the linear model, or

slope=φ(0)α(1γ)
(5)

for the sigmoidal model. Equation (5) is obtained by differentiating Eq. (2) and setting ITD = 0, and φ(0) is the Gaussian normal probability density at zero (∼0.3989). The ITD sensitivity metric is interpretable as the increase in the proportion of R spout choices for each μs increase in ITD.

D-prime values were estimated using the standard formula d′ = z(hit rate) – z(false alarm rate), where the hit rate is the proportion of “right” responses for a given positive ITD and the false alarm rate is the proportion of “right” responses for the corresponding negative ITD.

Figures 2 and 3 show the psychometric curves obtained for each rat (rows of panels) at each click rate (columns of panels) using rectangular or Hanning window click trains, respectively. There is considerable variability between individual rats: Rats #2 and #4 are examples of particularly good and poor performers, respectively. There are also clear, systematic effects of pulse rate and envelope. ITD sensitivity declined rapidly as pulse rates exceeded a few hundred Hz. Similar sharp declines in ITD sensitivity with pulse (Laback et al., 2007; van Hoesel et al., 2009; Chung et al., 2016) or AM (Joris and Yin, 1995; Bernstein, 2001) rates above 500 Hz have been reported in previous physiological and behavioral studies on humans and other species.

Fig. 2.

(Color online) Psychometric curves for rats localizing rectangular window click trains by ITD. Rows: individual rats. Columns: stimulus click rates. X-axes: stimulus ITD. Y-axes: proportion of responses to the right (R) spout. Negative ITDs mean the stimulus in left ear is leading. Dots indicate the observed proportion of R responses at each ITD tested. The fractions in small print show raw number of R responses / total number of trials at the corresponding ITD value. Error bars: 95% Wilson confidence intervals for the underlying probability of choosing the R spout. Solid lines: fitted psychometric models, as described in Sec. 2.4. Sigmoid fits are shown with dark lines, linear fits in a lighter shade, null model fits in a very light shade. Dotted diagonals: slopes of the fitted psychometric at ITD = 0.

Fig. 2.

(Color online) Psychometric curves for rats localizing rectangular window click trains by ITD. Rows: individual rats. Columns: stimulus click rates. X-axes: stimulus ITD. Y-axes: proportion of responses to the right (R) spout. Negative ITDs mean the stimulus in left ear is leading. Dots indicate the observed proportion of R responses at each ITD tested. The fractions in small print show raw number of R responses / total number of trials at the corresponding ITD value. Error bars: 95% Wilson confidence intervals for the underlying probability of choosing the R spout. Solid lines: fitted psychometric models, as described in Sec. 2.4. Sigmoid fits are shown with dark lines, linear fits in a lighter shade, null model fits in a very light shade. Dotted diagonals: slopes of the fitted psychometric at ITD = 0.

Close modal
Fig. 3.

(Color online) Psychometric curves for rats localizing Hanning window click trains by ITD. Plotted as in Fig. 2.

Fig. 3.

(Color online) Psychometric curves for rats localizing Hanning window click trains by ITD. Plotted as in Fig. 2.

Close modal

Figure 4 summarizes the ITD sensitivities across conditions. In addition to illustrating inter-subject variability and the overall decline in ITD sensitivity with increasing pulse rate, Fig. 4 also shows that ITD discrimination was consistently better for rectangular than for Hanning windowed stimuli. For rectangular windows, the mean sensitivity remained significantly above zero at all click rates, but for Hanning windows it declined to zero for four out of five rats at 900 Hz, and for all rats above 900 Hz. (Note that the model selection described in Sec. 2.4 only assigns non-zero ITD sensitivities to psychometrics where a χ2 deviance test rejects the zero slope null model at p < 0.05.)

Fig. 4.

(Color online) Summary of ITD sensitivity across click rates and window types. Faded lines show the ITD sensitivity of individual rats for rectangular (continuous lines) and Hanning (broken lines) windows. Dark lines show mean performance across animals.

Fig. 4.

(Color online) Summary of ITD sensitivity across click rates and window types. Faded lines show the ITD sensitivity of individual rats for rectangular (continuous lines) and Hanning (broken lines) windows. Dark lines show mean performance across animals.

Close modal

A repeated measures analysis of variance confirmed that ITD sensitivity exhibited a strong and highly significant dependence on click rate (F = 19.12, df = 4, p < 10−5, ηpartial2 = 0.93) as well as on click-rate by window type interactions (F = 7.414, df = 4, p = 0.0014, ηpartial2 = 0.74). Thus, consistent with previous work in other species (Stecker and Hafter, 2002), rectangular windowed stimuli produced onset ITD cues that facilitated localization across all click rates tested, whereas Hanning windowed stimuli, with their very gentle on- and offset slopes, generated only ongoing ITD cues. Arguably, our 50 Hz pulse rate condition, which is near the “fusion boundary rate” where human listeners no longer perceive click trains as individual clicks but instead as a continuous complex tone, might be considered as “consisting of nothing but onsets” irrespective of the window. However, the same cannot be said for our Hanning windowed 300 Hz click trains, which all our rats were able to localize with fairly high sensitivity, demonstrating that rats are able to use ongoing envelope ITDs. Sensitivity to ongoing ITDs has also recently been demonstrated in gerbils (Tolnai et al., 2018).

With 50 Hz rectangular click trains, our rats had a median 75% correct threshold of 65 μs ITD and a mean of 77 μs. Median and mean ITD thresholds corresponding to d′ = 1 were 46 and 56 μs, respectively. Our best performing rat (#2) exhibited 75% correct thresholds as low as ∼29 μs (compare Fig. 2) and d′ = 1 thresholds as low as 21 μs. These values are similar to the ∼10–60 μs range of 75% correct ITD discrimination thresholds reported for normal hearing human subjects tested with noise bursts (Klumpp and Eady, 1956) and pure tones (Zwislocki and Feldman, 1956), or the ∼40 μs ITD thresholds reported for normal hearing ferrets tested with noise bursts (Keating et al., 2013). These values also compare well to behavioral measures and theoretical estimates of ITD thresholds for other small mammals; these are reported to be ∼30 μs for cats (Wakeford and Robinson, 1974), ∼23–45 μs for guinea pigs (Greene et al., 2018), ∼50–60 μs for rabbits (Ebert, Jr. et al., 2008), ∼55 μs for chinchilla (Koka et al., 2011), and 12–96 μs for Mongolian gerbils (Tolnai et al., 2017).

In conclusion, even though free field localization experiments with low frequency tones have led to the suggestion that rats may be unable to use ITDs to localize sounds (Wesolek et al., 2010), we have shown that Wistar rats tested with pulse-resonance sounds clearly can lateralize both onset and ongoing envelope ITDs with a sensitivity and a reliance on onsets that are broadly similar to that seen in humans and other mammalian species.

Funding for this work was provided by Health and Medical Research Fund (HK) Grant No. 06172296 and Shenzhen Science Technology and Innovation Committee Grant No. JCYJ20180307124024360 to J.W.H.S., and a Postdoctoral Researchers International Mobility Experience (P.R.I.M.E.) fellowship with funds from the German Federal Ministry of Education and Research and the People Programme (Marie Curie Actions) of the European Union's Seventh Framework Programme (No. FP7/2007-2013) under REA Grant Agreement No. 605728 to N.R.K.

1.
Bernstein
,
L. R.
(
2001
). “
Auditory processing of interaural timing information: New insights
,”
J. Neurosci. Res.
66
,
1035
1046
.
2.
Brugge
,
J. F.
,
Reale
,
R. A.
,
Jenison
,
R. L.
, and
Schnupp
,
J.
(
2001
). “
Auditory cortical spatial receptive fields
,”
Audiol. Neuro-otol.
6
,
173
177
.
3.
Chung
,
Y.
,
Hancock
,
K. E.
, and
Delgutte
,
B.
(
2016
). “
Neural coding of interaural time differences with bilateral cochlear implants in unanesthetized rabbits
,”
J. Neurosci.
36
,
5520
5531
.
4.
Ebert
,
C. S.
, Jr.
,
Blanks
,
D. A.
,
Patel
,
M. R.
,
Coffey, C.
S.
,
Marshall, A.
F.
, and
Fitzpatrick, D.
C.
(
2008
). “
Behavioral sensitivity to interaural time differences in the rabbit
,”
Hear. Res.
235
,
134
142
.
5.
Greene
,
N. T.
,
Anbuhl
,
K. L.
,
Ferber
,
A. T.
,
DeGuzman
,
M.
,
Allen
,
P. D.
, and
Tollin
,
D. J.
(
2018
). “
Spatial hearing ability of the pigmented Guinea pig (Cavia porcellus): Minimum audible angle and spatial release from masking in azimuth
,”
Hear. Res.
365
,
62
76
.
6.
Heffner
,
H. E.
, and
Heffner
,
R. S.
(
1985
). “
Sound localization in wild Norway rats (Rattus norvegicus)
,”
Hear. Res.
19
,
151
155
.
7.
Itskov
,
P. M.
,
Vinnik
,
E.
,
Honey
,
C.
,
Schnupp
,
J.
, and
Diamond
,
M. E.
(
2012
). “
Sound sensitivity of neurons in rat hippocampus during performance of a sound-guided task
,”
J. Neurophysiol.
107
,
1822
1834
.
8.
Jones
,
E.
,
Oliphant
,
T.
, and
Peterson
,
P.
(
2014
). “
SciPy: Open source scientific tools for Python
,” http://www.scipy.org/ (Last viewed 4/17/2019).
9.
Joris
,
P. X.
, and
Yin
,
T. C.
(
1995
). “
Envelope coding in the lateral superior olive. I. Sensitivity to interaural time differences
,”
J. Neurophysiol.
73
,
1043
1062
.
10.
Kavanagh
,
G. L.
, and
Kelly
,
J. B.
(
1986
). “
Midline and lateral field sound localization in the albino rat (Rattus norvegicus)
,”
Behav. Neurosci.
100
,
200
205
.
11.
Keating
,
P.
,
Nodal
,
F. R.
,
Gananandan
,
K.
,
Schulz
,
A. L.
, and
King
,
A. J.
(
2013
). “
Behavioral sensitivity to broadband binaural localization cues in the ferret
,”
JARO
14
,
561
572
.
12.
Kidd
,
S. A.
, and
Kelly
,
J. B.
(
1996
). “
Contribution of the dorsal nucleus of the lateral lemniscus to binaural responses in the inferior colliculus of the rat: Interaural time delays
,”
J. Neurosci.
16
,
7390
7397
.
13.
Klumpp
,
R.
, and
Eady
,
H.
(
1956
). “
Some measurements of interaural time difference thresholds
,”
J. Acoust. Soc. Am.
28
,
859
860
.
14.
Koka
,
K.
,
Jones
,
H. G.
,
Thornton
,
J. L.
,
Lupo
,
J. E.
, and
Tollin
,
D. J.
(
2011
). “
Sound pressure transformations by the head and pinnae of the adult Chinchilla (Chinchilla lanigera)
,”
Hear. Res.
272
,
135
147
.
15.
Koka
,
K.
,
Read
,
H. L.
, and
Tollin
,
D. J.
(
2008
). “
The acoustical cues to sound location in the rat: Measurements of directional transfer functions
,”
J. Acoust. Soc. Am.
123
,
4297
4309
.
16.
Laback
,
B.
,
Majdak
,
P.
, and
Baumgartner
,
W.-D.
(
2007
). “
Lateralization discrimination of interaural time delays in four-pulse sequences in electric and acoustic hearing
,”
J. Acoust. Soc. Am.
121
,
2182
2191
.
17.
Lingner
,
A.
,
Wiegrebe
,
L.
, and
Grothe
,
B.
(
2012
). “
Sound localization in noise by gerbils and humans
,”
JARO
13
,
237
248
.
18.
Patterson
,
R. D.
(
2015
). “
Pulse-resonance sounds
,” in
Encyclopedia of Computational Neuroscience
, edited by
D.
Jaeger
and
R.
Jung
(
Springer
,
New York
).
19.
Schnupp
,
J. W.
,
Dawe
,
K. L.
, and
Pollack
,
G. L.
(
2005
). “
The detection of multisensory stimuli in an orthogonal sensory space
,”
Exp. Brain Res.
162
,
181
190
.
20.
Stecker
,
G. C.
, and
Hafter
,
E. R.
(
2002
). “
Temporal weighting in sound localization
,”
J. Acoust. Soc. Am.
112
,
1046
1057
.
21.
Tolnai
,
S.
,
Beutelmann
,
R.
, and
Klump
,
G. M.
(
2017
). “
Exploring binaural hearing in gerbils (Meriones unguiculatus) using virtual headphones
,”
PLoS One
12
,
e0175142
.
22.
Tolnai
,
S.
,
Beutelmann
,
R.
, and
Klump
,
G. M.
(
2018
). “
Interaction of interaural cues and their contribution to the lateralisation of Mongolian gerbils (Meriones unguiculatus)
,”
J. Comp. Physiol. A: Neuroethol. Sens. Neural. Behav. Physiol.
204
,
435
448
.
23.
van Hoesel
,
R. J. M.
,
Jones
,
G. L.
, and
Litovsky
,
R. Y.
(
2009
). “
Interaural time-delay sensitivity in bilateral cochlear implant users: Effects of pulse rate, modulation rate, and place of stimulation
,”
JARO
10
,
557
567
.
24.
Wakeford
,
O. S.
, and
Robinson
,
D. E.
(
1974
). “
Lateralization of tonal stimuli by the cat
,”
J. Acoust. Soc. Am.
55
,
649
652
.
25.
Wesolek
,
C. M.
,
Koay
,
G.
,
Heffner
,
R. S.
, and
Heffner
,
H. E.
(
2010
). “
Laboratory rats (Rattus norvegicus) do not use binaural phase differences to localize sound
,”
Hear. Res.
265
,
54
62
.
26.
Zwislocki
,
J.
, and
Feldman
,
R.
(
1956
). “
Just noticeable differences in dichotic phase
,”
J. Acoust. Soc. Am.
28
,
860
864
.