Content uploaded by M. Clara P. Amorim
Author content
All content in this area was uploaded by M. Clara P. Amorim on Feb 05, 2018
Content may be subject to copyright.
RESEARCH PAPER
The Role of Agonistic Sounds in Male Nest Defence in the
Painted Goby Pomatoschistus pictus
Ricardo Pereira*, Stefania Rismondo*, Manuel Caiano*, Silvia S. Pedroso†, Paulo J. Fonseca* &
Maria Clara P. Amorim†
* Departamento de Biologia Animal e Centro de Biologia Ambiental, Faculdade de Ci^
encias da Universidade de Lisboa, Lisboa, Portugal
†Unidade de Investigac
ß
~
ao em Eco-Etologia, Instituto Superior de Psicologia Aplicada –Instituto Universit
ario, Lisboa, Portugal
Correspondence
Maria Clara P. Amorim, Unidade de
Investigac
ß
~
ao em Eco-Etologia, Instituto
Superior de Psicologia Aplicada –Instituto
Universit
ario, Rua Jardim do Tabaco 34,
1149-041 Lisboa, Portugal.
E-mail: amorim@ispa.pt
Received: April 1, 2013
Initial acceptance: May 19, 2013
Final acceptance: September 25, 2013
(M. Manser)
doi: 10.1111/eth.12180
Abstract
Animals often vocalize during territorial challenges as acoustic signals may
indicate motivation and fighting ability and contribute to reduce aggres-
sive escalation. Here, we tested the function of agonistic sounds in territo-
rial defence in the painted goby. Pomatoschistus pictus, a small vocal marine
fish that defends nests during the breeding season. We first measured the
number of times a male approached, avoided, explored, entered and
exited two unattended nests associated with either conspecific agonistic
sounds or a control: silence or white noise. Acoustic stimuli were played
back when the male approached a nest. In a second experimental set, we
added visual stimuli, consisting of a conspecific male in a small confine-
ment aquarium near each nest. Even though we found no effect of the
visual stimuli, the sound playbacks induced similar effects in both experi-
mental conditions. In the sound vs. silence treatment, we found that
when males approached a nest, the playback of conspecific sounds usually
triggered avoidance. However, this behaviour did not last as in longer peri-
ods males visited nests associated with agonistic sounds more often than
silent ones. When the control was white noise, we found no significant
effect of the playback treatment in male behaviour. Although we cannot
exclude the possibility that other sounds may dissuade nest occupation,
our results suggest that agonistic sounds act as territorial intrusion deter-
rents but are insufficient to prevent nest intrusion on their own. Further
studies are needed to test the significance of sound production rate,
spectral content and temporal patterns to deter territorial intrusion in fish.
Introduction
In nature, resources such as those needed for breeding
(e.g. food, water accessibility, vegetation, shelter
availability and nesting sites) are often limited and of
variable quality. Hence, in order to have a chance to
reproduce, individuals (typically males) may compete
intensely over these limited resources (Huntingford &
Turner 1987; Andersson 1994). This competition is
often solved through confrontations that frequently
start with a mutual assessment phase, characterized
by low-level agonistic displays and are only expected
to escalate to overt aggression if asymmetries between
individuals’ fighting ability and resource ownership
are small (Huntingford & Turner 1987; Briffa & Sned-
don 2007). Low-level aggressive behaviour involved
in such contests is often accompanied by sound
production in a variety of different taxa (insects -
Gerhardt & Huber 2002; fish - Ladich & Myrberg
2006; anurans - Davies & Halliday 1978; birds –de
Kort et al. 2009; and mammals - Clutton-Brock & Al-
bon 1979). In most vertebrates, such as frogs, birds
and mammals, the sounds are the result of the passage
of air through specialized vocal organs (larynx, syr-
inx) driving the vibration of membranes. These
sounds typically exhibit strong frequency and ampli-
tude modulation (Ladich 2004). By contrast, teleost
fishes have evolved a diversity of sound-producing
Ethology 119 (2013) 1–11 ©2013 Blackwell Verlag GmbH 1
Ethology
mechanisms that often generate low-frequency
pulsed sounds lacking complex patterns of frequency
and amplitude modulation (Amorim 2006; Ladich &
Fine 2006; Rice & Bass 2009). This makes fish particu-
lar suitable models to study the function of acoustic
signals with conceptually simple playback experi-
ments. Nevertheless, carrying out sound playback
experiments with fish can be technically challenging
due to the lack of available commercial underwater
loudspeakers that can reproduce fish sounds accu-
rately (Fonseca & Maia Alves 2012).
Sounds produced by fishes during territorial
defence usually function as a complement to visual
behaviour such as colour alterations or aggressive
visual displays, including fin erection or quivering
(Ladich & Myrberg 2006). Though sound seems to
play an important role in fish communication there is
little experimental evidence demonstrating the func-
tion of acoustic signals in fish territorial defence
(Ladich & Myrberg 2006; Vasconcelos et al. 2010).
Experiments with muted specimens and with sound
playback suggest that in an agonistic context acoustic
signals can reveal valuable information about the sen-
der and may help avoid overt confrontation and thus
energy depletion or even injury and death (Valinsky
& Rigley 1981; Riggio 1981; Ladich et al. 1992; Ladich
1998; Raffinger & Ladich 2009; Bertucci et al. 2010;
for a review see Ladich & Myrberg 2006). For exam-
ple, body size can be advertised by the production of
longer sounds with lower dominant frequencies and
higher amplitudes that result from larger body struc-
tures (Myrberg et al. 1993; Lobel & Mann 1995; Con-
naughton et al. 2000; Amorim & Neves 2008; Colleye
et al. 2009; Amorim et al. 2013). Also, fish sounds
may potentially inform the receiver of the sender’s
motivation during a confrontation (Ladich 2004).
Indeed, steroid circulating levels that affect territorial
behaviour and aggressiveness may also influence the
patterns of sonic muscles contraction or of fin stridu-
lation (Fine & Pennyacker 1986; Connaughton et al.
1997; Remage-Healey & Bass 2006).
In this study, we aimed at examining the role of fish
agonistic sounds in territorial defence using playback
experiments in a small marine benthic species, the
painted goby (Pomatoschistus pictus). Painted goby
males produce drums and thumps to attract and court
females and only drums during agonistic interactions
(Amorim & Neves 2007, 2008). During the breeding
season (January–May), males compete aggressively
over nesting sites and have few breeding opportuni-
ties, as they only live up to 2 years (Miller 1986).
Males provide exclusive paternal care and use visual
and acoustic behaviour to defend nests that are used
both in mate attraction and to hold the eggs laid by
females (Bouchereau et al. 2003; Amorim & Neves
2007, 2008). As agonistic drums duration and drum
bout duration increase with male size, it is likely that
acoustic signals are involved in male-male assessment
(Amorim & Neves 2008).
We experimentally examined whether males
approached and intruded vacant nests (silent or asso-
ciated with white noise) more readily than ‘occupied’
nests. We mimicked nest occupation by playing back
a sequence of conspecific agonistic sounds whenever
a male approached a nest within a body length. We
checked for several reactions of the subject male to
vacant and ‘occupied’ nests, including approaching,
avoiding, exploring, entering and exiting the nests.
We further checked for nest intrusion behaviour in a
similar-design experiment that used additional visual
stimuli, consisting of a male in a confinement aquar-
ium placed next to the nest. We predicted that males
would be more reluctant to intrude nests associated
with agonistic sound, especially in the presence of a
conspecific male. Specifically, we expected that males
avoided more often and intruded less frequently nests
associated with agonistic sounds.
Materials and Methods
Fish Collection and Maintenance
We captured approx. 120 adult fish from January to
April 2012 at Parede (38°41′N, 9°21′W), during low
spring tides at night with the help of hand nets and
flashlights. We separated fish by gender which we
assessed by examining the external papilla (rounder
in females and longer and pointed in males) and by
the presence of nuptial colours in males and swollen
bellies in ripe females (Bouchereau et al. 2003). We
kept a maximum of eight fish per stock 18 l aquarium
(24 924 932 cm), provided with sand, shelters and
a closed-circuit flow of artificial filtered sea water kept
at 16°C. Fish were maintained under a 12L/12D
photoperiod and fed daily ad libitum with chopped
shrimps. At the end of experiments (May/June 2012),
the fishes were returned to their collection location in
a healthy condition.
Experimental Procedure
Thirty-five-litre glass experimental aquaria (26 951
931 cm) also provided with a sand substrate were
divided into three compartments by transparent or
opaque partitions. Water, conditioned by dedicated
filter, pumps and cooler, was similarly kept at 16°C.
Ethology 119 (2013) 1–11 ©2013 Blackwell Verlag GmbH2
Fish Sounds as Deterrents in Nest Intrusions R. Pereira, S. Rismondo, M. Caiano, S. S. Pedroso, P. J. Fonseca & M. C. P. Amorim
The aquaria were placed above two 3-cm-thick stone
slabs separated by two levels of rubber foam shock
absorbers. This helped to insulate the aquaria from
floor born vibrations and increased acoustic signal-
to-noise ratio, thus reducing possible interference
with our sound playbacks. Twenty-four hours before
an experiment, one male was placed in the central
compartment of the aquarium to allow it to become
resident. Only males that showed nuptial colours
and territorial behaviour in the stock tanks were
used. Each lateral compartment was equipped with
one artificial PVC shelter (5.5 cm length and 3 cm
inner diameter) partially covered with sand and one
custom made speaker (see below). The speaker was
protected by a plastic net to exclude fish from the
speaker area. We also covered both lateral walls of
the experimental aquarium with a fine net to
reduce areas where a fish might interact with its
own reflection.
Playback Treatment
Prior to all trials, we presented a 2 min sound pre-
stimulus during which subject fish were exposed to
conspecific agonistic sounds and a control, white
noise (WN) or silence. The 2 min sound pre-stimulus
intended to inform the subject fish of territorial own-
ership in the lateral compartments before the test per-
iod. Both the playback treatment (silence vs. sound or
WN vs. sound) and the stimuli assignment to the
lateral sides of the aquarium were randomly selected.
When sound was emitted at both sides of the
aquarium, that is, during WN vs. sound treatment, we
insured that the stimuli did not overlap to avoid
acoustic interference and masking. Different males
were used for each playback treatment.
Agonistic sounds used in the playback treatments
consisted of drums from three different males (stan-
dard length: 3.3, 3.3 and 3.8 cm) from our sound
archive (2010). We made three pre-stimulus drum
playback sequences. Each drum pre-stimulus
sequence comprised three different sounds from the
same male (Fig. 1a). Only sounds with a good sig-
nal-to-noise ratio were used. We also selected
sounds that would represent the natural variability
in number of pulses, sound duration and sound fre-
quency. All files had the same duration and acoustic
energy, and the spacing between sounds followed a
typical pattern observed in sound production during
territorial defence in this species (Fig. 1b). Each pre-
stimulus sequence presented a total of 34 sounds,
that is, a sound rate of 17 per min, a typical rate
for a motivated male (Amorim & Neves 2008). An
identical number of WN files were created and pre-
sented similar sound intervals as the drum
sequences and were equalized to the same sound
amplitude. Also, WN did not contain the pulse pat-
tern of the agonistic drums but had similar acoustic
energy and hence a shorter duration.
During the test period, every time a male
approached a nest and reached within one body
length from the entrance of the nest we immediately
played three sounds of the correspondent playback
treatment (sound or WN). This stimulus consisted of
the first three sounds of the sound pre-stimulus
sequence used in a particular trial. Playback experi-
ments were performed using custom made devices
composed by an underwater speaker and a driver
(Fonseca & Maia Alves 2012), which are able to
reproduce low-frequency pulsed fish sounds with
great accuracy, such as the ones emitted by the
painted goby (cf. Fig. 1 in Fonseca & Maia Alves
2012). These were fed through a D/A converter (Edi-
rol UA-25, Roland, Japan; 16 bit, 8 kHz) controlled by
Adobe Audition 3.0 (Adobe Systems Inc., Mountain
View, CA, USA), which allowed independent play-
backs on two channels. The amplitude of the
sound playback (drum or WN) was measured with a
(a)
(b)
(c)
Fig. 1: Oscillogram of three different drums used in a playback
sequence (a). A similar sequence of only three drums was played back
during trials whenever a subject male approached a nest within a body
length. A longer sequence of 2 min containing the same three drums
was used in the sound pre-stimulus in which the interval between
sounds followed a natural pattern. In (b), approximately 1 min of the
2 min sound pre-stimulus is schematized depicting the rate of 17
sounds per min used in the playback sequence. Note that an agonistic
sound sequence comprised three different sounds from the same male.
(c) Illustrates a played back drum recorded in the middle of the central
compartment. Note that this drum is from a different playback
sequence than the one depicted in (a).
Ethology 119 (2013) 1–11 ©2013 Blackwell Verlag GmbH 3
R. Pereira, S. Rismondo, M. Caiano, S. S. Pedroso, P. J. Fonseca & M. C. P. Amorim Fish Sounds as Deterrents in Nest Intrusions
hydrophone (Br€
uel & Kjær 8104, Br€
uel & Kjær, Nae-
rum, Denmark, sensitivity 205 dB re1V/lPa; fre-
quency response within 1 dB from 0.1 Hz to
180 kHz), connected to a Br€
uel & Kjær 2238 Mediator
Sound Level Meter, Naerum, Denmark) and adjusted
to mimic that of a painted goby male at 1–2 cm dis-
tance measured in previous experiments (approx.
130 dB SPL, re. 1 lPa; Amorim et al. 2013). An addi-
tional hydrophone (High Tech 94 SSQ, High Tech
Inc., Gulfport, MS, USA; sensitivity 165 dB re1V/
lPa; frequency response within 1 dB from 30 Hz to
6 kHz) was kept on the central compartment of the
experimental aquarium, to register any sound pro-
duced during trials and to monitor sound playback.
Played back sounds reached the central compartment
with a good signal-to-noise ratio and quality (Fig. 1c).
Playback Experiments
We carried out two experiments that differed in
the presence of visual stimuli during the test period:
‘playback only’ and ‘playback+visual’.
The ‘playback only’ experiment started 15 min after
we turned off all the equipment in the room to reduce
noise background in the experimental aquarium. We
started video/sound recordings, removed the opaque
partitions and exposed the subject fish to a 5 min
visual pre-stimulus. This consisted of a male inside a
small confinement aquarium (8 97922 cm), pro-
vided with sand substrate and high enough to prevent
chemical communication, placed in each lateral com-
partment next to the transparent partition (Fig. 2a).
Stimulus males were matched in size, that is, differ-
ence in standard length ratio was <10%. This visual
pre-stimulus intended to arouse territorial behaviour
in the subject fish and thus to increase responsiveness
during the trial (see Lugli 1997). We finished the
visual pre-stimulus period by replacing the opaque
partitions. We then removed the lateral stimulus
males and allowed a silence period of 8 min which
was followed by the 2-min sound pre-stimulus. The
PVC and acrylic plastic partitions allowed good
transmission of the sound between compartments.
Afterwards, all partitions were removed and the fish
behaviour was observed and recorded for another
5 min (test period).
In the ‘playback+visual’ experiment, there was no
visual pre-stimulus and trials started with the 2-min
sound pre-stimulus, prior to the partitions’ removal,
followed by the 5-min-test period. The stimulus males
stayed between the nest and the side wall of the
experimental aquarium throughout the test period
(Fig. 2b).
We used different males in the ‘playback only’ and
‘playback+visual’ experiments. Fish were measured
for standard length (SL) and weighted (W) after each
test. A total of 12 subject males with SL ranging from
3.4 to 3.9 cm (mean standard deviation [SD] =3.7
0.18 cm) and W from 0.4 to 0.88 g (0.67 0.12 g)
were tested in the ‘playback only’ experiment. From
these, eight and four males were tested in the silence
vs. sound and in the WN vs. sound treatments, respec-
tively. A total of 14 subject males were tested in the
‘playback+visual’ experiment with SL ranging from
3.4 to 3.9 cm (3.7 0.16 cm) and W from 0.48 to
0.88 g (0.7 0.13 g), half of which were tested in
the silence vs. sound and the other half in the WN vs.
sound treatments. All males interacted with the nests
except from two males tested in the ‘playback+visual’
experiment. These were 3.6 cm SL, 0.58 g W (silence
vs. sound treatment) and 3.9 cm SL, 0.62 g W (WN
vs. sound treatment).
Behavioural Recording and Analysis
Fish behaviour was recorded with a video camera
(Sony DCR-HC39, Sony, Tokyo, Japan) positioned
(a)
(b)
Fig. 2: Top view of experimental setup during the visual pre-stimulus
on the ‘playback only’ (a) and during the whole ‘playback+visual’ (b)
experiments. Males in glass confinement aquaria (8 97922 cm) were
placed near the transparent partitions (dashed lines) in ‘playback only’
experiments to provide a visual pre-stimulus before playback proce-
dures. In ‘playback+visual’ experiments visual stimulus males were
placed instead between the nest and the side wall of the experimental
tank (26 951 931 cm) throughout the experiment.
Ethology 119 (2013) 1–11 ©2013 Blackwell Verlag GmbH4
Fish Sounds as Deterrents in Nest Intrusions R. Pereira, S. Rismondo, M. Caiano, S. S. Pedroso, P. J. Fonseca & M. C. P. Amorim
50 cm in front of the experimental tank and subse-
quently analysed with Etholog (v.2.2, Ottoni 2000).
During the pre-stimuli period video recording
captured the entire aquarium, while during the test
period, we focused on the behaviour of the test male.
We tallied the frequency of occurrence of the fol-
lowing nest-related behavioural parameters: approach
(move within a body length of the nest), avoidance
(abruptly swims away from the nest), explore (swim
around and on top of the nest), enter and exit the
nest. Avoidance, explore and ‘initial enter’ were con-
sidered when they occurred within 5 s of the male’s
approach to the nest and hence approximately within
5 s of the playback stimulus. Enter and exit refer to
the number of times a male entered, following an
approach and exited a nest with no such time restric-
tion. Notice that in the case a male left the vicinity of
the nest and later approached it again and entered,
we considered that the first approach did not result in
nest intrusion (enter). The total time spent on each
side of the aquarium and the first side explored by the
fish were also registered. Trials were considered valid
when the tested male approached at least one of the
nests.
Statistical Analyses
Playback effects were analysed based on the different
behaviours exhibited by the subject males. We car-
ried out repeated measures ANOVAs to compare each
male behaviour (approach, avoidance, explore, enter
and exit) in relation to control nests and ‘sound-
associated’ nests, that is, using playback treatment
(control vs. sound) as the dependent variable (i.e.
the repeated measures variable), as males could
interact with both nest types. We also used experi-
ment type: ‘playback only’ and ‘playback+visual’ as a
factor (independent variable). We analysed sepa-
rately the data for the playback treatments silence vs.
sound and WN vs. sound. The two males that did not
interact with the nests were excluded from this
analysis, and hence, 24 males were considered in
total.
We tested male preference for the sound or control
sides of the experimental aquarium with Wilcoxon
non-parametric tests, where we considered the total
time spent on each side during the test period. We
carried out Binomial tests to examine whether males
first swam to the sound (or the control, i.e. silence or
WN) side of the aquarium more often than expected
by chance alone. The intention was to determine
whether there was a first preference based on the pre-
stimuli playback treatment (silence vs. sound or WN
vs. sound). We performed further Binomial tests to
examine whether the first nest occupation was made
randomly relative to nest type (test or control). In
these analyses, we pooled the ‘playback only’ and
‘playback+visual’ data because the presence of the
visual stimuli had no effect on the male’s behaviour
towards the nest (see results). We also used the full
data set, that is, the 26 tested males in these analyses
(except for the first nest occupation), as the two males
that did not interact with nests still showed active
exploration of the full aquarium.
Statistical analyses were conducted with Statistica
(10, Statsoft Inc., Tulsa, USA), and all data were trans-
formed when necessary to meet assumptions of the
used parametric tests. When there was no normality
of the transformed data, non-parametric tests were
used.
Results
During the acoustic pre-stimulus, when the partitions
were in place, subject males did not exhibit any mea-
surable reaction to the sound playback. In contrast,
during the visual pre-stimulus (in the ‘playback only’
experiment), tested males repeatedly swam up and
down the partition in front of the visual stimulus
males. Upon the removal of the partitions, the subject
male readily moved towards one side of the experi-
mental aquarium, this behaviour being more obvious
when an additional visual stimulus was present. In
the absence of conspecifics, males showed a general
explorative behaviour swimming around the aquar-
ium. However, when confined conspecific males were
present (‘playback+visual’ experiment), subject males
often interacted with the opponent by performing
quivering lateral and frontal displays while darkening
the chin.
In the treatment, silence vs. sound fish showed a
tendency (p =0.07) to approach more often nests
associated with sound than silent nests (Table 1,
Fig. 3a). Figure 4 depicts the number of times males
approached nests associated with different playback
treatments during both experiment types (‘playback
only’ and ‘playback+visual’) and hence the number of
received playback stimuli. Interestingly, males
avoided nests associated with agonistic drums more
frequently than silent nests (Table 1, Fig. 3b). In spite
of this, the frequency with which males entered both
nest categories within 5 s of an approach showed no
significant differences (Table 1), but in longer periods,
males entered and exited nests associated with sound
more frequently than silent ones (Table 1, Fig. 3c, d).
Hence, the first reaction of an approaching male
Ethology 119 (2013) 1–11 ©2013 Blackwell Verlag GmbH 5
R. Pereira, S. Rismondo, M. Caiano, S. S. Pedroso, P. J. Fonseca & M. C. P. Amorim Fish Sounds as Deterrents in Nest Intrusions
towards an ‘occupied’ nest (i.e. associated with
sounds) was often avoidance, but this behaviour
would not last and could be followed by a nest visit.
We found no significant differences between the
exploring behaviour towards silent or ‘occupied’
nests (Table 1). In addition, we found no effect of
experiment type for any variable (Table 1).
Regarding the treatment WN vs. sound, there were
no significant effects of the variables playback treat-
ment and experiment type for the nest-related behav-
iours approach, explore, enter (any time frame) and
exit performed by males towards the nest (Table 2).
Nevertheless, we found a significant effect of experi-
ment type on avoidance behaviour as males avoided
nests more often when there was no visual stimulus
nearby (Table 2, Fig. 5).
After the pre-stimuli, males swam to one side of the
aquaria (sound vs. control) at random, that is, swim-
ming towards the sound side was not different than
expected by chance (Binomial test, p >0.05). Fish
swam to the sound side 10 times out of 15 in the
silence vs. sound treatment and 6 of 11 in the WN vs.
sound treatment. However, when considering only
the males that entered a nest during trials (N =19), it
was significantly more likely that the first nest that
males entered was a nest associated with agonistic
sound rather than a control nest (Binomial test,
p=0.002). The first nest to be intruded was a nest
associated with agonistic sound 16 of 19 times: 6 of 8
times when the control was WN and 10 of 11 times
Table 1: Effects of playback treatment (PBK: silence vs. sound) and
experiment type (EXP: ‘playback’ only or ‘playback+visual’) on nest-
related behaviours: approach, avoidance, explore, initial enter (upon 5 s
of approach), enter (no time restraints) and exit
Dependent variable Factor Fp
Approach PBK F
1,12
=4.64 0.07
Exp F
1,12
=0.00 0.23
PBK 9Exp F
1,12
=1.16 0.35
√Avoidance PBK F
1,12
=5.37 0.04
Exp F
1,12
=2.15 0.17
PBK 9Exp F
1,12
=2.15 0.17
Explore PBK F
1,12
=0.02 0.89
Exp F
1,12
=2.44 0.14
PBK 9Exp F
1,12
=0.94 0.35
Log-initial enter PBK F
1,12
=0.24 0.64
Exp F
1,12
=0.07 0.80
PBK 9Exp F
1,12
=1.64 0.22
Log-enter PBK F
1,12
=4.84 0.048
Exp F
1,12
=0.30 0.59
PBK 9Exp F
1,12
=1.91 0.19
Log-exit PBK F
1,16
=4.47 0.05
Exp F
1,16
=0.11 0.74
PBK 9Exp F
1,16
=2.01 0.18
Data were transformed when necessary to meet the ANOVA assumptions.
(a) (b)
(c) (d)
Fig. 3: Comparison of the frequency of the subject male behaviour towards the nest between playback treatments (silence and sound): (a) approach,
(b) avoidance, (c) enter and (d) exit. Avoidance was measured in a time window of 5 s after approach, whereas enter and exit were measured with
no time restraints. Dots and error bars are means and 95% confidence intervals, respectively. Data were transformed when necessary to meet the
ANOVA assumptions. (*)p <0.10; *p<0.05.
Ethology 119 (2013) 1–11 ©2013 Blackwell Verlag GmbH6
Fish Sounds as Deterrents in Nest Intrusions R. Pereira, S. Rismondo, M. Caiano, S. S. Pedroso, P. J. Fonseca & M. C. P. Amorim
when the control was silence. The time until first nest
occupation averaged 91.2 s for nests associated with
agonistic sound (N =16) but only 50.3 s for control
nests (N =3). Males also spent more time in the
sound than in the control side of the aquarium
(Wilcoxon matched-pairs signed-ranks test: T=83.0,
N=26, p =0.02; Fig. 6).
Discussion
In fish, sounds play an important role in establishing
and defending territories allowing to assess the condi-
tion of competitors and to decide whether they should
possibly risk injury by engaging in a fight (Ladich
2004). In this study, we used a small marine nest-
reproducing fish as a model and observed the effect of
playing back conspecific aggressive sounds, which
mimicked the presence of a vocalizing nest-holder, on
nest occupation.
In the silence vs. sound playback treatment, we
found that although subject males first headed ran-
domly to a side of the aquarium (sound or control
side), they tended to approach nests associated with
conspecific sounds more frequently than silent ones.
As expected, a frequent first reaction of males
approaching nests associated with agonistic sounds
Fig. 4: Mean standard deviation of the number of times a male
approached a nest associated with different stimuli during the ‘playback
only’ and the ‘playback+visual’ experiments. Note that the number of
approaches to a nest also corresponds to the number of received stim-
uli. Sound1 and sound2 correspond to the sound playbacks performed
during the silence vs. sound and WN vs. sound treatments, respectively.
Table 2: Effects of playback treatment (PBK: WN vs. sound) and experi-
ment type (Exp: ‘playback’ only or ‘playback+visual’) on nest-related
behaviours: approach, avoidance, explore, initial enter (upon 5 s of
approach), enter (no time restraints) and exit
Dependent variable Factor Fp
Log-approach PBK F
1,8
=0.04 0.84
Exp F
1,8
=0.10 0.76
PBK 9Exp F
1,8
=0.24 0.64
Log-avoidance PBK F
1,8
=0.01 0.92
Exp F
1,8
=33.41 0.00
PBK 9Exp F
1,8
=0.01 0.92
Explore PBK F
1,8
=0.02 0.88
Exp F
1,8
=1.82 0.21
PBK 9Exp F
1,8
=0.60 0.46
Initial enter PBK F
1,8
=0.08 0.78
Exp F
1,8
=0.08 0.78
PBK 9Exp F
1,8
=2.11 0.18
Enter PBK F
1,8
=0.73 0.42
Exp F
1,8
=0.49 0.42
PBK 9Exp F
1,8
=0.00 0.97
Exit PBK F
1,8
=1.07 0.33
Exp F
1,8
=0.36 0.56
PBK 9Exp F
1,8
=0.02 0.89
Data were log-transformed when necessary to meet the ANOVA
assumptions.
Fig. 5: Avoidance behaviour exhibited by subject males towards the
nest according to different experiment types: ‘playback only’ and ‘play-
back+visual’. Dots and error bars are means and 95% confidence inter-
vals, respectively. *** p<0.001.
Fig. 6: Mean standard deviation of the amount of time spent by
tested males on each side of the aquarium according to playback treat-
ment, that is, control vs. sound. *p<0.05.
Ethology 119 (2013) 1–11 ©2013 Blackwell Verlag GmbH 7
R. Pereira, S. Rismondo, M. Caiano, S. S. Pedroso, P. J. Fonseca & M. C. P. Amorim Fish Sounds as Deterrents in Nest Intrusions
was avoidance, a behaviour that was significantly less
likely to occur on approaching silent nests. However,
contrary to our predictions, after the first avoidance
reaction, males entered ‘occupied’ nests more fre-
quently than silent ones, probably because they could
not associate the sound to a male nest-holder. Alto-
gether, the results suggest that agonistic sounds have
a significant role in territorial defence in the painted
goby and may act as a deterrent to nest intrusion until
the absence of a nest-holder physically defending the
nest is confirmed. Agonistic sounds have been shown
to be important in territorial defence in a few other
fish species. Schuster (1986) observed that in the
dwarf gourami Colisa lalia territorial defence was
approx. sixfold more effective when attacks chasing
intruders were accompanied by sounds than when
they were silent. Valinsky & Rigley (1981) muted
juveniles of skunk loaches Yasuhikotakia horae and
showed that muted individuals had fewer chances to
chase intruders from their shelters. Consistent with
our results, Ladich & Myrberg (2006) and Vasconcelos
et al. (2010) suggest that fish agonistic sounds can act
as a ‘keep-out’ signal towards intruders. For example,
nest-holding Lusitanian toadfish (Halobatrachus
didactylus) males defend their territories with loud
sounds (boatwhistles), which often elicits fleeing
behaviour from the intruder, thus decreasing the
chances of escalated fights and nest takeovers (Va-
sconcelos et al. 2010). Myrberg (1997) also showed
with a series of elegant field experiments that play-
backs of the chirp sound of bicolour damselfish males
(Stegastes partitus) made in unguarded territories inhi-
bit territorial intrusion by neighbouring conspecific
males for a significantly longer time than when there
is no sound playback. Consistent with our study, Myr-
berg also found that the deterrent effect of the sound
was short-term when transmitted in the absence of
the resident (Myrberg 1997). Similarly, in birds acous-
tic signals can slow down, but not prevent, territory
intrusion (reviewed in Nowicki et al. 1998). For
example, great tit males (Parus major) take longer to
intrude recently unoccupied territories if they are
associated with conspecific song playback than silent
territories (Krebs 1976).
The fact that males visited more often and were
more likely to first intrude nests associated with con-
specific sounds, and spent more time in the sound side
of the aquarium, deserves further attention. For
example, males could be selecting for good-quality
nests already chosen by other males thus saving the
time and energy costs of searching for a suitable nest-
ing site (Lindstr€
om & Pampoulie 2005). Also, as
females of the congeneric P. minutus prefer males that
already have eggs in the nest (Forsgren et al. 1996),
the search for such nests could be an added value if
mate selection in the studied species follows the same
pattern.
In the trials using white noise as a control, we regis-
tered no significant effect of the playback treatment in
any male nest behaviour suggesting that males may
not distinguish conspecific sound from white noise.
However, the lack of significant differences in the
reaction towards nests associated with agonistic sound
or white noise should be seen with caution as it is
likely due to the low sample size. In the bicolour dam-
selfish, Myrberg (1997) found that white noise had
no effect as a territorial deterrent. Consistently,
Schwarz (1974) found that agonistic sounds reduced
aggression in the cichlid, Archocentrus centrarchus,in
contrast with control noise that rendered no effect in
male–male agonistic interactions. These and other
studies (reviewed in Ladich & Myrberg 2006) support
our suggestion that low sample sizes hampered
the ability to find a treatment effect in this set of
experiments.
Fish are believed to mainly use the temporal
patterns of calls for communication (Ladich 2004;
Amorim 2006). Therefore, we removed the specific
temporal pattern from the white noise stimuli used in
our experiments to avoid relaying to the receiver
information likely contained in the pulse pattern. As a
consequence, and to keep the same acoustic energy
among the sound stimuli, a parameter that might
affect the behaviour by changing the overall stimula-
tion of the hearing pathway, the white noise stimuli
were shorter than the agonistic drums. However, by
doing this, we were modifying another possible rele-
vant parameter, the duration of the stimuli, which
thus did not corresponded to the duration of the aver-
age drum and might have influenced the fish
response. For example, McKibben & Bass (1998) car-
ried out phonotactic experiments with plainfin mid-
shipman females (Porichthys notatus) and concluded
that duty cycle (sound on time, i.e. acoustic energy)
and also sound duration affected female preference.
The decrease in duty cycle and sound duration of mat-
ing-like signals decreased the likelihood of phonotaxis
possibly because playback sounds decreased their
resemblance to the long tonal mate attraction hums
made by nesting males.
Although test and visual stimuli painted goby
males often engaged in mutual visual displays, the
presence of confined males had in general no effect
on the male’s nest-related behaviour. Most available
studies have shown that agonistic sounds usually
need to be associated with signals of other modali-
Ethology 119 (2013) 1–11 ©2013 Blackwell Verlag GmbH8
Fish Sounds as Deterrents in Nest Intrusions R. Pereira, S. Rismondo, M. Caiano, S. S. Pedroso, P. J. Fonseca & M. C. P. Amorim
ties to elicit appropriate behavioural responses (Lad-
ich 2004). For example, Bertucci et al. (2010)
experimentally separated or coupled visual and
acoustic signals to test the role of sounds produced
during male–male aggressive interactions in a cichlid
fish, Metriaclima zebra, and observed that acoustic
signals alone never triggered aggression. However,
when associated with visual signals agonistic sounds
significantly lowered the levels of aggressive behav-
iour observed when visual signals were presented
alone. The general lack of effect of additional visual
stimuli further points to a prevalent and indepen-
dent role of acoustic signals in territorial defence in
our study species, a role scarcely documented in
fish. However, we cannot rule out that the lack of
effects induced by the visual stimulus male might
be due to test fish perceiving that the male in
the confinement aquarium was not a threat and
could not defend the available nest. Nevertheless,
Myrberg (1997) found in the aforementioned field
experiments with the bicolour damselfish that
the visual presence of a confined resident near the
shelter (a male in a confinement bottle) was a more
effective deterrent to intrusion than sound playbacks
alone.
The function of sounds emitted during agonistic
behaviour is not fully understood in fishes and the
relative paucity of studies have shown that agonistic
sounds can have different roles depending on species
or individual status (reviewed in Raffinger & Ladich
2009). For example, in the satinfin shiner Cyprinella
analostana, a small cyprinid where males aggressively
defend territories while producing single knocks or
series of knocks (Stout 1963), playbacks of rapid series
of knocks can increase or inhibit males’ aggression
depending on the receiver’s social status. When domi-
nant males interacting with a mirror received play-
backs of agonistic sounds, they increased aggression,
in contrast to submissive males where the playback
induced the opposite effect (Stout 1975).
This study has revealed that acoustic communica-
tion is important in territorial defence in the painted
goby and likely other fishes as agonistic sounds slo-
wed down territorial intrusions. Additional experi-
ments are required to test the significance of sound
rates, spectral and temporal patterns to deter territo-
rial intrusion in fish and to ascertain the specificity of
the acoustic features that can prime territorial behav-
iour. Other studies not involving direct assessment of
the function of particular sound features on territorial
defence have, however, shed light on signal parame-
ters that might be used by fish to encode-decode rele-
vant information. For example, sound pressure level,
sound frequency or sound emission rate were demon-
strated to give information on fish size and condition
and therefore resource holding potential (Ladich
1998; McKibben & Bass 1998; Amorim et al. 2010).
However, the best approach to assess the function of
signals is the use of playback experiments. This has
been common practice to infer the territorial function
of acoustic signals and has shown in other taxa the
importance of particular acoustic traits in male-male
assessment (Davies & Halliday 1978; de Kort et al.
2009). Associating nests with acoustic stimuli differing
in the above-mentioned features could shed light to
whether agonistic sounds only inform the intruder of
the presence of a nest-holder or if they are also used
in male-male assessment.
Acknowledgements
We are thankful to Susana Varela for comments on
the manuscript. This research was funded by the
Science and Technology Foundation, Portugal (project
PTDC/MAR/68868/2006, pluriannual program UI&D
331/94 and UI&D 329). The methods for animal
collection, housing, handling and experimental proto-
cols comply with Portuguese animal welfare laws,
guidelines and policies. The authors have no conflict
of interest.
Literature Cited
Amorim, M. C. P. 2006: Diversity of sound production in
fish. In: Communication in Fishes, Vol. 1 (Ladich, F.,
Collin, S. P., Moller, P. & Kapoor, B. G., eds). Science
Publishers, Enfield, N.H., pp. 71—104.
Amorim, M. C. P. & Neves, A. S. M. 2007: Acoustic signal-
ling during courtship in the painted goby, Pomatoschistus
pictus. J. Mar. Biol. Assoc. UK 87, 1017—1023.
Amorim, M. C. P. & Neves, A. S. M. 2008: Male painted
gobies (Pomatoschistus pictus) vocalise to defend
territories. Behaviour 145, 1065—1083.
Amorim, M. C. P., Sim~
oes, J. M., Mendonc
ßa, N., Bandarra,
N. M., Almada, V. C. & Fonseca, P. J. 2010: Lusitanian
toadfish song reflects male quality. J. Exp. Biol. 213,
2997—3004.
Amorim, M. C. P., Pedroso, S. S., Bolgan, M., Jord~
ao, J.
M., Caiano, M. & Fonseca, P. J. 2013: Painted gobies
sing their quality out loud: acoustic rather than visual
signals advertise male quality and contribute to mating
success. Funct. Ecol. 27, 289—298.
Andersson, M. 1994: Sexual Selection. Princeton Univ.
Press, Princeton, New Jersey.
Bertucci, F., Beauchaud, M., Attia, J. & Mathevon, N.
2010: Sounds modulate males’ aggressiveness in a cich-
lid fish. Ethology 116, 1179—1188.
Ethology 119 (2013) 1–11 ©2013 Blackwell Verlag GmbH 9
R. Pereira, S. Rismondo, M. Caiano, S. S. Pedroso, P. J. Fonseca & M. C. P. Amorim Fish Sounds as Deterrents in Nest Intrusions
Bouchereau, J. L., Houder, V., Marques, A. & Rebelo, J. E.
2003: A new distribution record and the reproductive
strategy of Pomatoschistus pictus adriaticus (Pisces: Gobii-
dae) in the Mediterranean Sea. J. Mar. Biol. Assoc. UK
83, 1157—1161.
Briffa, M. & Sneddon, L. U. 2007: Physiological constraints
on contest behaviour. Funct. Ecol. 21, 627—637.
Clutton-Brock, T. H. & Albon, S. D. 1979: The roaring of
red deer and the evolution of honest advertisement.
Behaviour 69, 145—169.
Colleye, O., Fr
ed
erich, B., Vandewalle, P., Casadevall, M.
& Parmentier, E. 2009: Agonistic sounds in the skunk
clownfish Amphiprion akallopisos: size-related variation
in acoustic features. J. Fish Biol. 75, 908—916.
Connaughton, M. A., Fine, M. L. & Taylor, M. H. 1997:
The effects of seasonal hypertrophy and atrophy on
fiber morphology, metabolic substrate concentration
and sound characteristics of the weakfish sonic muscle.
J. Exp. Biol. 200, 2449—2457.
Connaughton, M. A., Taylor, M. H. & Fine, M. L. 2000:
Effects of fish size and temperature on weakfish distur-
bance calls: implications for the mechanism of sound
generation. J. Exp. Biol. 203, 1503—1512.
Davies, N. B. & Halliday, T. R. 1978: Deep croaks and fight
assessment in toads Bufo bufo. Nature 274, 683—685.
Fine, M. L. & Pennyacker, K. R. 1986: Hormonal basis for
sexual dimorphism of the sound-producing apparatus of
the oyster toadfish. Exp. Neurol. 92, 289—298.
Fonseca, P. J. & Maia Alves, J. 2012: A new concept in
underwater high fidelity low frequency sound genera-
tion. Rev. Sci. Instrum. 83, 055007.
Forsgren, E., Karlsson, A. & Kvarnemo, C. 1996: Female
sand gobies gain direct benefits by choosing males with
eggs in their nests. Behav. Ecol. Sociobiol. 39,91—96.
Gerhardt, H. C. & Huber, F. 2002: Acoustic Communica-
tion in Insects and Anurans: Common Problems and
Diverse Solutions. The Univ. of Chicago Press, Chicago
and London.
Huntingford, F. & Turner, A. K. 1987: Animal Conflict.
Chapman & Hall, London.
de Kort, S. R., Eldermire, E. R. B., Cramer, E. R. A. & Vehr-
encamp, S. L. 2009: The deterrent effect of bird song in
territory defense. Behav. Ecol. 20, 200—206.
Krebs, J. R. 1976: Bird song and territorial defence. New
Scientist 70, 534—536.
Ladich, F. 1998: Sound characteristics and outcome of
contests in male croaking gouramis (Teleostei).
Ethology 104, 517—529.
Ladich, F. 2004. Sound production and acoustic communica-
tion. In: The Senses of Fishes. Adaptations for the Recep-
tion of Natural Stimuli (van der Emde, G., Mogdans, J. &
Kapoor, B. G., eds). Narosa, New Delhi, pp. 210—230.
Ladich, F. & Fine, M. L. 2006. Sound-generating mecha-
nisms in fishes: a unique diversity in vertebrates. In:
Communication in Fishes, Vol. 1 (Ladich, F., Collin, S.
P., Moller, P. & Kapoor, B. G., eds). Science Publishers,
Enfield, N.H., pp. 3—43.
Ladich, F. & Myrberg, A. A. 2006. Agonistic behavior and
acoustic communication. In Communication in fishes,
Vol. 1 (Ladich, F., Collin, S. P., Moller, P. & Kapoor, B.
G., eds). Science Publishers, Enfield, N.H., pp.
122—148.
Ladich, F., Bischof, C., Schleinzer, G. & Fuchs, A. 1992:
Intra- and interspecific differences in agonistic vocaliza-
tion in croaking gouramis (Genus: Trichopsis, Anaban-
toidei, Teleostei). Bioacoustics 4, 131—141.
Lindstr€
om, K. & Pampoulie, C. 2005: Effects of resource
holding potential and resource value on tenure at nest
sites in sand gobies. Behav. Ecol. 16,70—74.
Lobel, P. S. & Mann, D. 1995: Spawning sounds of the
damselfish, Dascyllus albisella (Pomacentridae), and rela-
tionship to male size. Bioacoustics 6, 187—198.
Lugli, M. 1997: Response of male goby, Padogobius mar-
tensii, to aggressive sound playback following pre-
experimental visual stimulation. Behaviour 134,
1175—1188.
McKibben, J. R. & Bass, A. H. 1998: Behavioral assessment
of acoustic parameters relevant to signal recognition
and preference in a vocal fish. J. Acoust. Soc. Am. 104,
3520—3533.
Miller, P. J. 1986: Gobiidae. In: Fish of the North-eastern
Atlantic and Mediterranean, Vol. III (Whitehead, P. J. P,
Bauchot, M. L., Hureau, J. C., Nielsen, J. & Tortonese,
E., eds). UNESCO, Paris, pp. 1019—1985.
Myrberg, A. A. Jr 1997: Sound production by a coral reef
fish (Pomacentrus partitus): evidence for a vocal, territo-
rial “keep-out” signal. Bull. Mar. Sci. 60, 1017—1025.
Myrberg, A. A. Jr, Ha, S. J. & Shamblott, H. 1993: The
sounds of bicolor damselfish (Pomacentrus partitus): pre-
dictors of body size and a spectral basis for individual
recognition and assessment. J. Acoust. Soc. Am. 94,
3067—3070.
Nowicki, S., Searcy, W. A. & Hughes, M. 1998: The terri-
tory defense function of song in song sparrows: a test
with the speaker occupation design. Behaviour 135,
615—628.
Ottoni, E. B. 2000: EthoLog 2.2: a tool for the transcription
and timing of behavior observation sessions. Behav.
Res. Methods Instrum. Comput. 32, 446—449.
Raffinger, E. & Ladich, F. 2009: Acoustic threat displays
and agonistic behaviour in the red-finned loach
Yasuhikotakia modesta. J. Ethol. 27, 239—247.
Remage-Healey, L. & Bass, A. H. 2006: A rapid
neuromodulatory role for steroid hormones in the
control of reproductive behavior. Brain Res. 1126,
27—35.
Rice, A. N. & Bass, A. H. 2009: Novel vocal repertoire and
paired swimbladders of the three-spined toadfish,
Batrachomoeus trispinosus: insights into the diversity of
the Batrachoididae. J. Exp. Biol. 212, 1377—1391.
Ethology 119 (2013) 1–11 ©2013 Blackwell Verlag GmbH10
Fish Sounds as Deterrents in Nest Intrusions R. Pereira, S. Rismondo, M. Caiano, S. S. Pedroso, P. J. Fonseca & M. C. P. Amorim
Riggio, R. 1981: Acoustical correlates of aggression in the
bicolor damselfish, Pomacentrus partitus. M.Sc. Thesis,
University of Miami, Miami, USA.
Schuster, S. 1986: Untersuchungen zur Bioakustik des
Zwergfisches Colisa lalia (Perciformes, Belontiidae).
Zool. Beitr. 29, 295—306.
Schwarz, A. 1974: The inhibition of aggressive behavior by
sound in cichlid fish, Cichlasoma centrarchus. Z. Tierpsy-
chol. 35, 508—517.
Stout, J. F. 1963: The significance of sound production
during the reproductive behaviour of Notropis analost-
anus (Family Cyprinidae). Anim. Behav. 11,83—92.
Stout, J. F. 1975: Sound communication during the
reproductive behavior of Notropis analostanus (Pisces:
Cyprinidae). Am. Midl. Nat. 94, 296—325.
Valinsky, W. & Rigley, L. 1981: Function of sound produc-
tion by the skunk loach Botia horae (Pisces, Cobitidae).
Z. Tierpsychol. 55, 161—172.
Vasconcelos, R. O., Sim~
oes, J. M., Almada, V. C., Fonseca,
P. J. & Amorim, M. C. P. 2010: Vocal behaviour during
territorial intrusions in the Lusitanian toadfish: boat-
whistles also function as territorial ‘keep-out’ signals.
Ethology 116, 155—165.
Ethology 119 (2013) 1–11 ©2013 Blackwell Verlag GmbH 11
R. Pereira, S. Rismondo, M. Caiano, S. S. Pedroso, P. J. Fonseca & M. C. P. Amorim Fish Sounds as Deterrents in Nest Intrusions
