Animal Behaviour 114 (2016) 139e146

Contents lists available at ScienceDirect

Animal Behaviour
journal homepage: www.elsevier.com/locate/anbehav

Senderereceiver dynamics in leafhopper vibrational duetting

Anka Kuhelj, Maarten de Groot 1, Andrej Blejec, Meta Virant-Doberlet*
Department of Organisms and Ecosystems Research, National Institute of Biology, Ljubljana, Slovenia

a r t i c l e i n f o
Article history:
Received 16 September 2015
Initial acceptance 22 October 2015
Final acceptance 18 January 2016
Available online 2 March 2016
MS. number: 15-00804R
Keywords:
calling effort
duet coordination
mate searching
rivalry behaviour
signalling adaptability
vibrational communication

A coordinated reciprocal exchange of acoustic signals (duetting) is common in arthropods relying on

substrate-borne vibrational signalling. Communication between partners is under evolutionary pressures
resulting from ecological and sexual selection and reciprocal effects arising from such dynamic
interactions may inuence the sender's and receiver's mating success. We investigated the inuence of
female reply duration on male mate-searching effort in the leafhopper Aphrodes makarovi in which the
female reply is essential for successful location of the female. In a duet, the beginning of a female reply
overlaps the end of the male call and males evaluate only the nonoverlapped duration of the female
reply. In playback experiments we varied the duration of female replies within the natural range. The
duration of a female reply was negatively correlated with the male calling effort. By increasing her reply
duration a female may signicantly reduce the male's direct and indirect costs associated with signalling
and searching, thus, ultimately, affecting male reproductive success. Males showed high adaptability in
signalling behaviour and when female replies were short, searching males shortened the last section of
their advertisement calls. This strategy allows the nonoverlapped part of the female reply to be longer
irrespective of its overall duration. Despite its deceptively simple form, vibrational duetting may entail
more complex interactions than just temporal coordination.
2016 The Association for the Study of Animal Behaviour. Published by Elsevier Ltd. All rights reserved.

Communication between partners is an essential part of

reproductive behaviour (e.g. Shuster & Wade, 2003). Signals used in
sexual communication enable identication (species, sex, condition, receptivity), as well as the location of a potential partner
(Mendelson & Shaw, 2012; Wilkins, Seddon, & Safran, 2013), and
therefore provide cues that are involved in mate choice and play an
important role in promoting and maintaining reproductive isolation (Kraaijeveld, Kraaijeveld-Smit, & Maan, 2011; Ritchie, 2007).
Sexual communication often involves a reciprocal exchange of
signals between partners (Bailey, 2003; Hall, 2009; Lewis &
Cratsley, 2008; Rodrguez & Barbosa, 2014). Duets are characterized by a predictable temporal association between the partners'
airborne or substrate-borne acoustic signals, and temporal coordination between the initiating call and a reply is expressed in reply
latency, as well as in alternation or overlapping of signals (Bailey,
2003; Hall, 2009). The coordinated exchange of airborne or
substrate-borne acoustic signals has been described in many taxa
including arthropods (Bailey, 2003; Uhl & Elias, 2011), amphibians
(Emerson & Boyd, 1999), birds (Hall, 2009) and mammals

* Correspondence: M. Virant-Doberlet, Department of Organisms and Ecosystems

Research, National Institute of Biology, Ve

cna pot 111, SI-1000 Ljubljana, Slovenia.
E-mail address: meta.virant@nib.si (M. Virant-Doberlet).
1
Present address: Slovenian Forestry Institute, Ljubljana, Slovenia.

(Geissmann 2000). While duetting between partners appears to be

relatively rare in airborne sound communication (Cooley &
Marshall, 2001; Emerson & Boyd, 1999; Hall, 2009; Robinson &
Hall, 2002), it is common in arthropods relying on vibrational signalling (Bailey, 2003; Boumans & Johnsen, 2015; Henry et al., 2013;
Rodrguez & Barbosa, 2014). In arthropods, where partners establish only a temporary bond prior to copulation, the function of a
duet has been associated with mate recognition and mate choice
(Rodrguez & Barbosa, 2014), as well as with location of a partner
(Derlink, Pavlov

ci

c, Stewart, & Virant-Doberlet, 2014; Legendre,
Marting, & Cocroft, 2012; Polajnar et al., 2014).
Duetting is a complex and dynamic interaction in which both
partners modify their signals and behaviour according to the partner's reply (de Groot et al., 2012; Polajnar et al., 2014; Rodrguez &
Barbosa, 2014; Rodrguez, Haen, Cocroft, & Fowler-Finn, 2012).
Detailed studies of reciprocal effects arising from signal exchange
that inuence the sender's and receiver's individual mating success
may improve our understanding of the function and evolution of
duets and their role in sexual and ecological selection (Rodrguez &
Barbosa, 2014; Wilkins et al., 2013). Furthermore, they may provide
valuable insights into the mechanisms of animal communication in
general (Bailey, 2003; Hall, 2009).
Here, we investigated the role of the duration of a female reply
in senderereceiver dynamics in the leafhopper Aphrodes makarovi.

http://dx.doi.org/10.1016/j.anbehav.2016.02.001
0003-3472/ 2016 The Association for the Study of Animal Behaviour. Published by Elsevier Ltd. All rights reserved.

140

A. Kuhelj et al. / Animal Behaviour 114 (2016) 139e146

Sexual communication is based on a species-specic vibrational

duet initiated by a male advertisement call to which a sexually
receptive female responds and the beginning of a female reply
overlaps with the last section of the male call (Fig. 1; Bluemel et al.,
2014; de Groot et al., 2012). While in this species a higher calling
rate increases the probability of the male locating the female
(Kuhelj, de Groot, Pajk, Sim

ci

c, & Virant-Doberlet, 2015), it also has
a detrimental effect on the male's survival, due to eavesdropping
predators (Virant-Doberlet, King, Polajnar, & Symondson, 2011) and
indirect costs arising from high energy expenditure (Kuhelj, de
Groot, Pajk, et al., 2015).
A reciprocal exchange of vibrational signals in A. makarovi differs from the majority of other duetting systems studied so far, in
that female replies are often longer than male calls (Bluemel et al.,
2014; de Groot et al., 2012). Furthermore, the duration of the female
reply varies substantially (5e60 s). This communication system
provided an ideal opportunity to explore in more detail the inuence of the duration of the female reply on male signalling and
searching behaviour (Rodrguez & Barbosa, 2014). We hypothesized
(1) that male mate-searching effort (which includes advertising
signalling and walking to locate the female) is correlated with the
duration of the female reply and (2) that in order to optimize his
searching behaviour the male has to adjust his signalling according
to the duration of the female reply. Males have to trigger every
female reply and they are able to evaluate only the nonoverlapped
duration of the female signal which depends also on the duration of
the initiating call (Kuhelj, de Groot, Blejec, & Virant-Doberlet,
2015). Moreover, during his approach to the female, the male is
motionless while calling; walking associated with searching
behaviour is limited to the period of the nonoverlapped female
reply and shortly afterwards (Kuhelj, de Groot, Pajk, et al., 2015).
Consequently, a longer nonoverlapped duration of the female reply
should enable males to walk longer distances before stopping and
calling again. We predicted that longer female replies would be
associated with lower male calling effort and that when the female
reply is short, males would shorten the duration of their calls in

order to obtain a longer nonoverlapped duration of the female

reply.
METHODS
Study Species
Aphrodes makarovi (Hemiptera, Cicadellidae) is a relatively large
representative of the leafhopper family (males around 6.5 mm,
females around 7.5 mm; Bluemel et al., 2014). This species is widely
distributed over the Palaearctic and has also been introduced to
North America. Aphrodes makarovi is a phloem feeder and has been
found on various host plants and in different habitats, often syntopically with other members of this genus (Bluemel et al., 2014).
Field population densities vary greatly between localities as well as
between years (Derlink et al., 2014). Males increase their signalling
space by moving from plant to plant (y/jump-call strategy) and a
reply from a sexually receptive female triggers a more localized
search on a plant. The female reply is essential for successful
mating, since the male does not approach the female if she does not
respond. A female mates only once in her lifetime (Bluemel et al.,
2014). In rivalry situations (trio: two males and one female)
males produce masking signals that overlap either the latter part of
the female reply or, less frequently, the call of another male (Kuhelj,
de Groot, Pajk, et al., 2015; Appendix Fig. A1).
Males of A. makarovi produce long and complex advertisement
calls with a stereotyped structure composed of chirps and regularly
repeated pulses (Bluemel et al., 2014; de Groot et al., 2012; Derlink
et al., 2014; Fig. 1). The nonspecic and highly variable rst section
(makarovi element 0 (Me0)) is followed by species-specic
elements Me1eMe3. The female reply is a series of regularly
repeated pulses and there is no difference in the pulse repetition
time between the last section of the male call (Me3) and the female
reply. Both signals have broadband frequency characteristics;
however, the dominant frequency of the male call usually lies
between 100 and 300 Hz, while in the female reply most energy is

Frequency (kHz)

5
(a)

4
3
2
1

Velocity (relative)

+1
(b)

2s
1

Me1
Me0

Me2

Me3

Figure 1. Representative maleefemale duet in A. makarovi. The (a) spectrogram (FFT, window size 4096 samples, 50% overlap) and (b) the corresponding waveform are shown.
Me0eMe3: elements in male advertisement call as described in Derlink et al., 2014.

A. Kuhelj et al. / Animal Behaviour 114 (2016) 139e146

contained in the frequency range 500e1500 Hz (Bluemel et al.,

2014; de Groot et al., 2012; Fig. 1).
Collection and Maintenance
Last-stage nymphs and young adult leafhoppers were
collected using a sweep net in alfalfa, Medicago sativa, and
stinging nettle, Urtica dioica, at various localities in Slovenia. In
the laboratory, males and females were kept separated by sex and
age in plastic boxes (38
26 cm and 27 cm high) containing up to
20 individuals at 23e28  C, 50e70% humidity and 15:9 h
light:dark photoperiod. They were fed with cut plants placed in
vials lled with water and replaced twice a week. Owing to
morphological similarities between Aphrodes species, the species
identity of males used in behavioural tests was determined prior
to experiments by recording their vibrational signals (Derlink
et al., 2014). Adult A. makarovi males used in behavioural trials
were 3e4 weeks old (time after eclosion) and were put individually in plastic cups (volume 0.5 litres) 1 day before the start of
experiments.
Ethical Note
The study does not involve endangered or protected species and
no permits were required for the eld collection. All experiments
complied with national laws on animal experimentation. Leafhoppers were maintained under optimal conditions, handled with
care and not harmed at any stage.
Experimental Set-up
For all behavioural tests we used an experimental set-up that


enabled bilateral playback stimulation (described in de Groot, Cokl,
& Virant-Doberlet, 2011; de Groot et al., 2012; Kuhelj, de Groot,
Pajk, et al., 2015). The top of the nettle plant (height approximately
25 cm) was cut off and the bottom of the stem was inserted into a
vial lled with water to prevent withering. All leaves, except the
pair at the apex and another pair approximately 12 cm down the
stem, were removed. Each of the lower leaves was attached to a
separate vibration exciter and a new plant was taken every second
day.
We applied female vibrational replies (described below) to the
leaf via the conical tip of the 5 cm metal rod (4 mm in diameter)
screwed rmly into the head of a vibration exciter (Minishaker type
4810, Brel & Kjr, Nrum, Denmark) and xed to the tip of the
leaf with Blu-tack. The vibration exciter was driven from the
computer via a Sound Blaster X-Fi Surround 5.1 pro sound card
(Creative, Singapore) by Cool Edit Pro2 (Syntrillium Software,
Phoenix, AZ, U.S.A.). Female replies and male vibrational signals
were registered from the reective tape placed on the nettle stem
1 cm below the branching point of the lower leaves with a laser
vibrometer (PDV-100, Polytech, GmbH, Waldbronn, Germany) and
stored in a computer using the above-mentioned sound card and
Cool Edit Pro2 software at the sampling rate 48 kHz and 16-bit
resolution. We regularly checked that the frequency characteristics of playback signals recorded on the plant corresponded to the
input signal by comparing spectral properties of the applied and
recorded vibrational signals. Male behaviour together with recorded vibrational signals were simultaneously lmed with a 3CCD
video camcorder (Canon DM XM2).
Experimental Protocol
A single male was placed on a leaf at the apex of the nettle plant
and to induce him to call we played a prerecorded duet once.

141

Afterwards, the male received a manually triggered female reply

(see below) to every advertisement call he made. To simulate a
natural duet we timed the onset of each female reply with the
typical decrease in the amplitude at the end of the male call, since
this corresponds to the timing of the female response often
observed in a natural A. makarovi duet (de Groot et al., 2011, 2012).
Males are not able to detect a female reply while calling and in
order to trigger an appropriate behavioural response of the male, a
female reply has to continue immediately after the end of the
initiating call (Kuhelj, de Groot, Blejec, et al., 2015). The amplitude
of stimulation was adjusted to the level of natural male calls and
female replies registered at the point of recording (male call:
0.3 mm/s; female reply: 0.2 mm/s). The side from which we applied
the female reply changed randomly. The trials continued for 15 min
after stimulation with a duet or until the male localized the
vibration exciter, whichever came rst.
We tested males in two experimental series that each included
12 males (24 males altogether). In both series males were tested
with the same four playback treatments in which we varied the
duration of the female reply within the natural range of durations
determined previously (Bluemel et al., 2014; de Groot et al., 2012).
The female signal used in playbacks was chosen from the signal
library and was composed of natural pulses assembled by Cool Edit
Pro 2. This composed female reply and manual playback application
trigger the same male signalling and searching behaviour as
observed in natural mating behaviour (de Groot et al., 2012). In the
control treatment F10 the duration of the female reply was 10.4 s
which corresponded to the average nonoverlapped value. The F5
treatment represented a shorter-than-average female reply (duration 5.2 s), while two values (20.8 and 41.6 s) corresponded to the
longer-than-average female responses (F20 and F41 treatments).
Males were tested only once per day and each male was tested once
with each female reply. The treatment order was randomized for
each male.
We monitored the following behavioural parameters: calling
rate (number of calls/min of trial), duration of the replyecall
interval (time between the end of the female reply and the
beginning of the following male call), number of males searching
(dened as leaving the apex of the nettle plant and walking during
the female reply), number of males locating the source (i.e. vibrated
leaf), searching time (time needed to locate the source after the
onset of searching) and number of males producing masking
signals.
To get an insight into the dynamics of the exchange of male and
female vibrational signals we measured the duration of the last
section (Me3) within the calls (Fig. 1).
To compare the effect of the duration of the female reply on
male and female signalling effort, we calculated the cumulative
calling duration needed to locate the female, which was dened as
the sum of durations of all individual calls produced during the
trial. In A. makarovi, male longevity is also negatively correlated
with the cumulative calling duration (Kuhelj, de Groot, Pajk, et al.,
2015). The corresponding cumulative signalling duration for
females was obtained by summing the total durations of the
playback replies.

Analyses
We included in the analyses only males that produced advertisement calls since calling indicates that males were motivated to
nd a mate. Only males that located the source of the female reply
were included in the analyses of searching time and cumulative
signalling duration. We compared the numbers of males and values
obtained in treatments with short and long female replies with the

142

A. Kuhelj et al. / Animal Behaviour 114 (2016) 139e146

numbers and values obtained in the control F10 treatment (i.e.

female reply of average duration).
Since signalling and behavioural parameters were often not
normally distributed (ShapiroeWilk test: P < 0.05) data were analysed with nonparametric tests. We compared values using the
Wilcoxon matched-pairs signed-ranks test which takes into account male individual identity. One- and two-tailed Fisher's exact
tests were used to compare the numbers of searching males,
numbers of males locating the source and numbers of males producing masking signals. We used the Wilcoxon rank sum test to
compare the durations of the replyecall intervals in males that
located or failed to locate the source in the allotted time and the
cumulative signalling durations for males and females.
To analyse the dynamics of reciprocal exchange of male and
female vibrational signals, males were divided into two categories
according to their behaviour: males that called but did not search
(calling males) and males that both called and searched for the
source of the female reply (searching males). For each trial we
took the Me3 duration of the rst call in the trial as a reference,
since males were not yet able to evaluate the duration of the female reply. Since Me3 durations in the rst call did not differ
signicantly between treatments and male categories (Appendix
Table A1), we determined the proportion of calls produced during the rest of the trial, in which the Me3 section was shorter than
in the rst call. For each male category within the treatment we
determined whether the proportion of these calls deviated
signicantly from a random distribution (0.5) using a two-tailed,
one-sample t test.
All statistical analyses were conducted using R version 2.15.2 (R
Development Core Team, 2012).
RESULTS
The proportion of searching males was high in all treatments
(Fig. 2a). In comparison with duets that included the female reply of
average duration (F10 treatment), the number of males locating the
source of the shortest and the longest female replies was signicantly lower (Fig. 2b). Males needed signicantly more time to
locate the source in the F5 and F20 treatments than the source of the
average female reply (Wilcoxon signed-ranks test: F5: V 2, N 7,
P 0.047; F20: V 1, N 7, P 0.031; Fig. 2c). In males that located
the source, the calling rate decreased with increasing female reply
duration (Wilcoxon signed-ranks test: F5: V 1, N 7, P 0.031;
F20: V 26, N 7, P 0.047; F41: V 21, N 7, P 0.031; Fig. 2d).
The duration of the replyecall interval did not differ between
treatments (Wilcoxon signed-ranks test with Bonferroni correction) and searching males usually called 2e3 s after the end of the
female reply (Appendix Fig. A2).
In comparison with the F10 treatment, the cumulative calling
duration needed to locate the source of the shorter-than-average
female reply (F5 treatment) was signicantly longer, while it was
signicantly shorter when males received the longest reply (Wilcoxon rank sum test: F5: W 99, N1 14, N28, P 0.002; F41:
W 9, N1 14, N2 7, P 0.002; Fig. 3a). In contrast, when female
replies were shorter than average, the cumulative signalling duration of the female replies in the F5 treatment did not differ from those
in the F10 treatment, while it was signicantly longer in both treatments with the longer-than-average female replies (F20 and F41
treatments; Wilcoxon rank sum test: F20: W 119, N1 14, N2 10,
P 0.004; F41: W 82, N1 14, N2 7, P 0.015; Fig. 3b). Furthermore, while in duets with the average female reply, the cumulative
signalling duration for males and females did not differ (Wilcoxon
rank sum test: W 122, N1 N2 14, P 0.280); it was signicantly higher in males when the female reply was short (Wilcoxon
rank sum test: W 62, N1 N2 8, P 0.002) and signicantly

higher in females when males received longer-than-average female

replies (Wilcoxon rank sum test: W 24, N1 N2 10, P 0.045
and W 0, N1 N2 7, P 0.002 for 20.8 and 41.6 s reply,
respectively).
When males received the short (F5) and average (F10) female
replies, in searching males, the proportion of calls with the Me3
section shorter than in the rst call was signicantly higher than
random (Table 1). In contrast, in the presence of the
longer-than-average female replies, the searching males did not
consistently shorten the Me3 section. Although under our experimental conditions, in which the female reply was timed to the end
of the male call, by shortening their calls males never achieved a
longer nonoverlapped duration of the female reply, taken together
these results suggest that males searching for the source of short
replies adjusted their signalling according to the duration of the
female reply.
In all treatments, some males showed rivalry behaviour
expressed as masking signals that overlapped the latter part of the
female reply and the number of males producing these signals in
treatments with longer-than-average female replies was signicantly higher than in the F10 treatment (Fig. 4).
DISCUSSION
Communication between partners is under strong evolutionary
pressures resulting from ecological and sexual selection (Maan &
Seehausen, 2011; Mendelson & Shaw, 2012; Rodrguez et al.,
2013), duetting being no exception (Bailey, 2003; Hall, 2009;
Rodrguez & Barbosa, 2014). In agreement with our hypotheses,
the duration of a female reply had profound effects on male matesearching behaviour and inuenced the likelihood of nding the
female. Overall, our results show the importance of female signals
in senderereceiver dynamics which, ultimately, affects male
reproductive success.
In general, longer replies should provide males with better
mate choice and searching cues (de Groot et al., 2011; Rodrguez &
Barbosa, 2014; Rodrguez et al., 2012). The results of the present
study show that a duet structure, in which the beginning of a
female reply overlaps the end of the initiating male call, may offer
males additional control over the reciprocal interaction. In
A. makarovi, males obtain the information needed to increase their
mating probability by evaluating the nonoverlapped duration of
the female reply (Kuhelj, de Groot, Blejec, et al., 2015) and they can
increase this duration irrespective of the latency and overall
duration of the female reply by shortening the last section of their
calls. In this species, it has been suggested that, in a natural mating
sequence, the duration of a male call may be inuenced by the
duration of the previous female reply (de Groot et al., 2012). In the
present study, males showed a surprisingly high adaptability in
their signalling behaviour that requires from males that they
retain the information at least about the duration of their last call
and of the corresponding female reply and adjust the duration of
the following call accordingly. The crucial role of the nonoverlapped duration of the female reply in the male's calling effort
and reproductive success may also favour learning in the context
of sexual behaviour. Learning and memory related to various
behavioural contexts, including sexual behaviour, are well established in insects, although the available information is limited to
only a few model systems (e.g. Dukas, 2006; Mery, 2013; Paur &
Gray, 2011).
Sexual selection on male phenotypes is often driven by female
preference for more complex or longer signals (Ryan & KeddyHector, 1992) and in the treehopper Enchenopa binotata female
preference is the strongest source shaping male vibrational signals
(Rodrguez, Ramaswamy, & Cocroft, 2006; Sullivan-Beckers &

(a)

1
0.8
0.6
0.4
0.2
0

F10

N=

F5
22

20

0.8
0.6
0.4
0.2
0

F41
21

N=

F10

F5

F20

F41

14

10

F5
22

F20
21

F41
21

(d)

300

F10
20

N=

600

*
*

Calling rate (per min)

Searching time (s)

900 (c)

F20
21

143

(b)

Proportion of calling males

locating the female

Proportion of searching males

A. Kuhelj et al. / Animal Behaviour 114 (2016) 139e146

0
N=

F10

F5

F20

F41

14

10

Cumulative signalling duration (s)

Figure 2. The effect of duration of the female reply on signalling and searching behaviour of A. makarovi males. Duration of female reply: control treatment F10: 10.4 s; F5 treatment:
5.2 s; F20 treatment: 20.8 s; F41 treatment: 41.6 s. (a) Proportion of searching males; (b) proportion of calling males locating the source; (c) searching time; (d) calling rate in males
locating the source. Proportions and values obtained in the control F10 treatment are shown in black. (a, b) Determined proportion (black or white circle) together with 95%
condence interval for proportions is shown. Asterisks indicate values that are signicantly lower than in the F10 treatment (one-tailed Fisher's exact test: P < 0.05). (c, d) Box and
whisker plots show the median (black or white line), the 25e75% interquartile range (boxes) and the lowest and the highest data points still within 1.5 times the interquartile range
(whiskers). Asterisks indicate a signicant difference from the F10 treatment (Wilcoxon signed-ranks test: P < 0.05). N number of males.

(a)
500

(b)
500

**

**

**

400

400

300

300

200

200

100

100

0
N=

F10

F5

F20

F41

14

10

0
N=

F10

F5

F20

F41

14

10

Figure 3. Cumulative signalling duration needed to locate the source of the female reply for (a) males and (b) females. Duration of female reply: control treatment F10: 10.4 s; F5
treatment: 5.2 s; F20 treatment: 20.8 s; F41 treatment: 41.6 s. Values obtained in the control F10 treatment are shown in black. Box and whisker plots show the median (black or
white line), the 25e75% interquartile range (boxes), the lowest and the highest data points still within 1.5 times the interquartile range (whiskers) and outliers (circles). Asterisks
indicate values that differ signicantly from the F10 treatment (Wilcoxon rank sum test: *P < 0.05; **P < 0.01). N number of males.

144

A. Kuhelj et al. / Animal Behaviour 114 (2016) 139e146

Table 1
Proportion of advertisement calls with Me3 section shorter than in the rst call in
the trial
Treatment

Proportion of calls with shorter Me3 section

Calling males

F5
F10
F20
F41

Searching males

Mean

Mean

5
2
4
2

0.495
0.000a
0.337
0.875a

17
18
17
18

0.692*
0.701*
0.490
0.378

Duration of female reply: F10 treatment: 10.4 s; F5 treatment: 5.2 s; F20 treatment:
20.8 s; F41 treatment: 41.6 s. Calling males: males that called, but did not search for
the source of a female reply. Searching males: males that both called and searched
for the source of a female reply. N number of males. Only males that produced
more than one call are included. Asterisks indicate values that deviate signicantly
from a random (0.5) distribution (two-tailed, one-sample t test: df N  1,
*P < 0.05).
a
Not tested due to the small number of males.

Proportion of males producing

masking signals

0.8
*
*
0.6

0.4

0.2

0
N=

F10

F5

F20

F41

20

22

21

21

Figure 4. The proportions of A. makarovi males producing masking signals. Duration of

female reply: control treatment F10: 10.4 s; F5 treatment: 5.2 s; F20 treatment: 20.8 s;
F41 treatment: 41.6 s. Proportion obtained in the control F10 treatment is shown in
black. Determined proportion (black or white circle) together with 95% condence
interval is shown. Asterisks indicate values that are signicantly different from the F10
treatment (two-tailed Fisher's exact test: P < 0.05). N number of males.

Cocroft, 2010). Females can express their preferences not only by

selectively duetting with some males (Bailey, 2003; Rodrguez
et al., 2006), but also by producing longer replies (Rodrguez &
Barbosa, 2014; Rodrguez et al., 2012). In agreement with our
hypothesis, short female replies resulted in higher male calling
effort and increased searching time. Taken together, our results
suggest that by increasing the duration of her replies from short to
average, an A. makarovi female may, for no additional personal cost,
signicantly reduce the male's calling and searching costs. In this
species vibrational calling imposes direct costs due to eavesdropping spiders (Virant-Doberlet et al., 2011), as well as indirect costs
due to high energy expenditure (Kuhelj, de Groot, Pajk, et al., 2015);
reduced costs may contribute more to a male's lifetime reproductive success. It has been suggested that, in natural duets, during a
mating sequence A. makarovi females vary their reply duration
according to the duration of the male call to which they are

responding (de Groot et al., 2012). Future studies should elucidate

the extent of the observed large variation in female reply duration
in this species (Bluemel et al., 2014; de Groot et al., 2012) related to
mate choice (Rodrguez & Barbosa, 2014).
However, contrary to our expectations, there was no overall
benecial effect of longer-than-average female replies on male
searching behaviour. Moreover, producing longer-than-average
replies means that the female would have to invest signicantly
more in signalling than the male. In addition, females may also be
affected by direct costs due to eavesdropping predators (VirantDoberlet et al., 2011). Taken together, the results of the present
and previous studies (de Groot et al., 2011, 2012; Kuhelj, de Groot,
Pajk, et al., 2015) suggest that the durations of the male call and
the female reply, the resulting costs and benets and the processing of directional information in the central nervous system
may be tightly coordinated. Neuronal mechanisms underlying
vibrational directionality are virtually unexplored (Virant

Doberlet, Cokl,
& Zorovi
c, 2006; Zorovi
c, 2011) and we can only
speculate that neuronal processing of long vibrational signals may
mer, 2014; Reinhold, 2011). However,
be difcult (Hartbauer & Ro
reduced success in locating the source of long female replies in
the allotted time may also result from increased expression of
rivalry behaviour, since males do not move while producing
masking signals. The data obtained in the present study were not
sufcient to reveal any correlation between the production of
masking signals, success in locating the source and searching
time.
Unexpectedly, the results of the present study revealed
behaviour that may underlie complex rivalry interactions. Males
retained the information about the presence of a rival acquired
from one presentation of a prerecorded duet at the beginning of
each trial and produced more masking signals in the presence of
long female replies. Alternative mating tactics have been documented in many acoustic communication systems (Gerhardt &
Huber, 2002; Hammond & Bailey, 2003; Villarreal & Gilbert,
2014) including those based on vibrational signals (VirantDoberlet et al., 2014). Temporal variation in the social environment (e.g. level of competition and number of mating opportunities) over short timescales favours male behavioural plasticity
(Bretman, Gage, & Chapman, 2011). Males of A. makarovi overlap
female signals produced in reply to the call of another male
(Kuhelj, de Groot, Pajk, et al., 2015) as well as the female reply
received in response to their own calls (present study). While it is
not yet clear whether long female replies offer some advantage to
the eavesdropping rival males relying on satellite behaviour, it
should be noted that masking signals are as energetically
demanding to produce as advertisement calls (Kuhelj, de Groot,
Pajk, et al., 2015). In addition, our results indicate that care
should be taken when male signalling is induced by the playback
of a maleefemale duet, since in some circumstances even a single
presentation may be sufcient to induce the lasting expression of
alternative mating tactics.
In summary, our study shows that, although in insects duets
have a supercially simple form, as in birds (Hall, 2009; Kovach,
Hall, Vehrencamp, & Mennill, 2014; Templeton et al., 2013),
vibrational duetting in insects entails more complex interactions
than just temporal coordination. In A. makarovi mate-searching
behaviour is associated with high plasticity in adapting male
signalling behaviour to the duration of the female reply. Such
attentiveness to the partner is usually associated with duets in birds
(Hall, 2009; Kovach et al., 2014). The reciprocal exchange of
vibrational signals is a highly complex interaction in which the
female reply plays an important role, not only in male signalling
and searching behaviour but also in the expression of rivalry tactics.
Selective pressures driving the evolution of duetting are likely to

A. Kuhelj et al. / Animal Behaviour 114 (2016) 139e146

depend on various ecological and social interactions (Bailey, 2003;

Hall, 2009), as well as on processes involved in sexual selection and
reproductive isolation (Derlink et al., 2014; Rodrguez & Barbosa,
2014; Wilkins et al., 2013).
Acknowledgments
We thank our colleagues from the Department of Organisms and
Ecosystems Research at the National Institute of Biology for their
help with eld work and maintaining leafhoppers in the laboratory.
Funding was provided by the Slovenian National Research Agency
(PhD fellowship 1000-11-310197 to A.K., research project J1-2181
and research programme P1-0255). We also thank the anonymous referees for their helpful comments.

References
Bailey, W. J. (2003). Insect duets: underlying mechanisms and their evolution.
Physiological Entomology, 28, 157e174. http://dx.doi.org/10.1046/j.13653032.2003.00337.x.
Bluemel, J. K., Derlink, M., Pavlov

ci

c, P., Russo, I.-R. M., King, R. A., Corbett, E., et al.
(2014). Integrating vibrational signals, mitochondrial DNA and morphology for
species determination in the genus Aphrodes (Hemiptera: Cicadellidae).
Systematic Entomology, 39, 304e324. http://dx.doi.org/10.1111/syen.12056.
Boumans, L., & Johnsen, A. (2015). Stoney duets: vibrational sexual mimicry can
explain complex patterns. Journal of Ethology, 33, 87e107. http://dx.doi.org/
10.1007/s10164-015-0423-y.
Bretman, A., Gage, M. J. G., & Chapman, T. (2011). Quick-change artists: male plastic
behavioural responses to rivals. Trends in Ecology & Evolution, 26, 467e473.
http://dx.doi.org/10.1016/j.tree.2011.05.002.
Cooley, J. R., & Marshall, D. C. (2001). Sexual signalling in periodical cicadas Magicicada spp. (Hemiptera: Cicadidae). Behaviour, 138, 827e855. http://dx.doi.org/
10.1163/156853901753172674.


de Groot, M., Cokl,
A., & Virant-Doberlet, M. (2011). Searching behaviour in two
hemipteran species using vibrational communication. Central European Journal
of Biology, 6, 756e769. http://dx.doi.org/10.2478/s11535-011-0056-2.


de Groot, M., Derlink, M., Pavlov

ci

c, P., Pre
sern, J., Cokl,
A., & Virant-Doberlet, M.
(2012). Duetting behaviour in the leafhopper Aphrodes makarovi (Hemiptera:
Cicadellidae). Journal of Insect Behavior, 25, 419e440. http://dx.doi.org/10.1007/
s10905-011-9304-6.
Derlink, M., Pavlov

ci

c, P., Stewart, A. J. A., & Virant-Doberlet, M. (2014). Mate
recognition in duetting species: the role of male and female vibrational signals.
Animal Behaviour, 90, 181e193. http://dx.doi.org/10.1016/j.anbehav.2014.01.023.
Dukas, R. (2006). Learning in the context of sexual behaviour in insects. Animal
Biology, 56, 125e141. http://dx.doi.org/10.1163/157075606777304258.
Emerson, S. B., & Boyd, S. K. (1999). Mating vocalizations of female frogs: control
and evolutionary mechanisms. Brain, Behavior and Evolution, 53, 187e197.
http://dx.doi.org/10.1159/000006594.
Geissmann, T. (2000). Duet songs of the siamang Hylobates syndactylus. I. Structure
and organisation. Primate Report, 56, 33e60.
Gerhardt, H. C., & Huber, F. (2002). Acoustic communication in insects and anurans:
Common problems and diverse solutions. Chicago, IL: University of Chicago Press.
Hall, M. L. (2009). A review of vocal duetting in birds. Advances in the Study of
Behavior, 40, 67e121. http://dx.doi.org/10.1016/S0065-3454(09)40003-2.
Hammond, T. J., & Bailey, W. J. (2003). Eavesdropping and defensive auditory
masking in an Australian bushcricket Caedicia (Phaneropterinae: Tettigoniidae:
Orthoptera).
Behaviour,
140,
79e95.
http://dx.doi.org/10.1163/
156853903763999917.
mer, H. (2014). From microseconds to seconds and minutes e
Hartbauer, M., & Ro
time computation in insect hearing. Frontiers in Physiology, 5, 138. http://
dx.doi.org/10.3389/fphys.2014.00138.
Henry, C. S., Brooks, S. J., Duelli, P., Johnson, J. B., Wells, M. M., & Mochizuki, A.
(2013). Obligatory duetting behaviour in the Chrysoperla carnea-group of
cryptic species (Neuroptera: Chrysopidae): its role in shaping evolutionary
history. Biological Reviews, 88, 787e808. http://dx.doi.org/10.1111/brv.12027.
Kovach, K. A., Hall, M. L., Vehrencamp, S. L., & Mennill, D. J. (2014). Timing isn't
everything: responses of tropical wrens to coordinated duets, uncoordinated
duets and alternating solos. Animal Behaviour, 95, 101e109. http://dx.doi.org/
10.1016/j.anbehav.2014.06.012.
Kraaijeveld, K., Kraaijeveld-Smit, F. J. L., & Maan, M. E. (2011). Sexual selection and
speciation: comparative evidence revisited. Biological Reviews, 86, 367e377.
http://dx.doi.org/10.1111/j.1469-185X.2010.00150.x.
Kuhelj, A., de Groot, M., Blejec, A., & Virant-Doberlet, M. (2015). The effect of timing
of female vibrational reply on male signalling and searching behaviour in the
leafhopper Aphrodes makarovi. PLoS One, 10, e0139020. http://dx.doi.org/
10.1371/journal.pone.0139020.
Kuhelj, A., de Groot, M., Pajk, F., Sim

ci

c, T., & Virant-Doberlet, M. (2015). Energetic
cost of vibrational signalling in a leafhopper. Behavioral Ecology and Sociobiology, 69, 815e828. http://dx.doi.org/10.1007/s00265-015-1898-9.

145

Legendre, F., Marting, P. R., & Cocroft, R. B. (2012). Competitive masking of vibrational signals during mate-searching in a treehopper. Animal Behaviour, 83,
361e368. http://dx.doi.org/10.1016/j.anbehav.2011.11.003.
Lewis, S. M., & Cratsley, C. K. (2008). Flash signal evolution, mate choice and predation in reies. Annual Review of Entomology, 53, 293e321. http://dx.doi.org/
10.1146/annurev.ento.53.103106.093346.
Maan, M. E., & Seehausen, O. (2011). Ecology, sexual selection and speciation.
Ecology Letters, 14, 591e602. http://dx.doi.org/10.1111/j.1461-0248.2011.01606.x.
Mendelson, T. C., & Shaw, K. L. (2012). The (mis)conception of species recognition.
Trends in Ecology & Evolution, 27, 421e427. http://dx.doi.org/10.1016/
j.tree.2012.04.001.
Mery, F. (2013). Natural variation in learning and memory. Current Opinion in
Neurobiology, 23, 52e56. http://dx.doi.org/10.1016/j.conb.2012.09.001.
Paur, J., & Gray, D. A. (2011). Individual consistency, learning and memory in a
parasitoid y Ormia ochracea. Animal Behaviour, 82, 825e830. http://dx.doi.org/
10.1016/j.anbehav.2011.07.017.
Polajnar, J., Eriksson, A., Rossi Stacconi, M. V., Lucchi, A., Anfora, G., VirantDoberlet, M., et al. (2014). The process of pair formation mediated by substrateborn vibrations in small insect. Behavioural Processes, 107, 68e78. http://
dx.doi.org/10.1016/j.beproc.2014.07.013.
R Development Core Team. (2012). A language and environment for statistical
computing. Vienna, Austria: R Foundation for Statistical Computing.
Reinhold, K. (2011). Variation in acoustic signalling traits exhibits footprints of
sexual selection. Evolution, 65, 738e745. http://dx.doi.org/10.1111/j.15585646.2010.01130.x.
Ritchie, M. G. (2007). Sexual selection and speciation. Annual Review of Ecology,
Evolution
and
Systematics,
38,
79e102.
http://dx.doi.org/10.1146/
annurev.ecolsys.38.091206.095733.
Robinson, D. J., & Hall, M. J. (2002). Sound signalling in Orthoptera. Advances in
Insect Physiology, 29, 151e278. http://dx.doi.org/10.1016/S0065-2806(02)
29003-7.
Rodrguez, R. L., & Barbosa, F. (2014). Mutual behavioral adjustement in vibrational
duetting. In R. B. Cocroft, M. Gogala, P. S. M. Hill, & A. Wessel (Eds.), Studying
vibrational communication (pp. 147e169). Berlin, Germany: Springer-Verlag.
http://dx.doi.org/10.1007/978-3-662-43607-3_9.
bel, G., & Symes, L. B.
Rodrguez, R. L., Boughman, J. W., Gray, D. A., Hebets, E. A., Ho
(2013). Diversication under sexual selection, the relative roles of mate preference strength and the degree of divergence in mate preferences. Ecology
Letters, 16, 964e974. http://dx.doi.org/10.1111/ele.12142.
Rodrguez, R. L., Haen, C., Cocroft, R. B., & Fowler-Finn, K. D. (2012). Males adjust
signalling effort based on female mate-preference cues. Behavioral Ecology, 24,
1218e1225. http://dx.doi.org/10.1093/beheco/ars105.
Rodrguez, R. L., Ramaswamy, K., & Cocroft, R. B. (2006). Evidence that female
preferences have shaped male signal evolution in a clade of specialized plantfeeding insects. Proceedings of the Royal Society B: Biological Sciences, 273,
2585e2593. http://dx.doi.org/10.1098/rspb.2006.3635.
Ryan, M. J., & Keddy-Hector, A. (1992). Directional patterns of female mate choice
and the role of sensory biases. American Naturalist, 139, S4eS35. http://
dx.doi.org/10.1086/285303.
Shuster, S. M., & Wade, M. J. (2003). Mating systems and mating strategies. Princeton,
NJ: Princeton University Press.
Sullivan-Beckers, L., & Cocroft, R. B. (2010). The importance of female choice, malemale competition and signal transmission as causes of selection on male
mating signals. Evolution, 64, 3158e3171. http://dx.doi.org/10.1111/j.15585646.2010.01073.x.
n, A. A., Quiros-Guerrero, E., Macas
Templeton, C. N., Mann, N. I., Ros-Chele
Garcia, C., & Slater, P. J. B. (2013). An experimental study of duet integration in
the happy wren Pheugopedius felix. Animal Behaviour, 86, 821e827. http://
dx.doi.org/10.1016/j.anbehav.2013.07.022.
Uhl, G., & Elias, D. O. (2011). Communication. In M. E Heberstein (Ed.), Spider
behaviour: Flexibility and versatility (pp. 128e189). Cambridge, U.K.: Cambridge
University Press.
Villarreal, S. M., & Gilbert, C. (2014). Male Scudderia pistillata katydids defend their
acoustic duet against eavesdroppers. Behavioral Ecology and Sociobiology, 68,
1669e1675. http://dx.doi.org/10.1007/s00265-014-1775-y.


Virant-Doberlet, M., Cokl,
A., & Zorovi
c, M. (2006). Use of substrate vibrations for
orientation: from behaviour to physiology. In S. Drosopoulos, & M. F. Claridge
(Eds.), Insect sounds and communication: Physiology, behaviour, ecology and
evolution (pp. 81e97). Boca Raton, FL: Taylor & Francis.
Virant-Doberlet, M., King, R. A., Polajnar, J., & Symondson, W. O. C. (2011). Molecular
diagnostics reveal spiders that exploit vibrational signals used in sexual
communication. Molecular Ecology, 20, 2204e2216. http://dx.doi.org/10.1111/
j.1365-294X.2011.05038.x.
Virant-Doberlet, M., Mazzoni, V., de Groot, M., Polajnar, J., Lucchi, A.,
Symondson, W. O. C., et al. (2014). Vibrational communication networks:
eavesdropping and biotic noise. In R. B. Cocroft, M. Gogala, P. S. M. Hill, &
A. Wessel (Eds.), Studying vibrational communication (pp. 93e123). Berlin, Germany: Springer-Verlag. http://dx.doi.org/10.1007/978-3-66243607-3_7.
Wilkins, M. R., Seddon, N., & Safran, R. J. (2013). Evolutionary divergence in acoustic
signals: causes and consequences. Trends in Ecology & Evolution, 28, 156e166.
http://dx.doi.org/10.1016/j.tree.2012.10.002.
Zorovi
c, M. (2011). Temporal processing of vibratory communication signals at the
level of ascending interneurons in Nezara viridula (Hemiptera: Pentatomidae).
PLoS One, 6, e26843. http://dx.doi.org/10.1371/journal.pone.0026843.x.

146

A. Kuhelj et al. / Animal Behaviour 114 (2016) 139e146

APPENDIX

5
Frequency (kHz)

(a)
4
3
2
1

Velocity (relative)

+1
(b)

0.5 s
1

Figure A1. Representative masking signal of an A. makarovi male. The (a) spectrogram (FFT, window size 1024 samples, 80% overlap) and (b) the corresponding waveform are
shown.

Table A1
Durations of the Me3 section in the rst male advertisement call in the trial for each
treatment and male category

Reply - call interval (s)

40

Treatment

F5
F10
F20
F41

30

10

N=

Calling males

Searching males

MeanSD

MeanSD

5
2
4
3

9.61.9
9.50.1
8.81.6
8.83.2

17
18
17
18

10.01.8
9.61.9
10.12.4
9.62.1

Duration of female reply: F10 treatment: 10.4 s; F5 treatment: 5.2 s; F20 treatment:
20.8 s; F41 treatment: 41.6 s. Calling males: males that called but did not search for
the source of a female reply. Searching males: males that both called and searched
for the source of a female reply; this category includes males that were searching
for, but did not locate, the source of the female reply in the allotted time, as well as
males that located the source. N number of males in each category. Me3 durations
in the rst call did not differ signicantly between treatments and male categories
(KruskaleWallis test: c27 2.706, P 0.911).

20

Duration of Me3 section in the rst call (s)

F10

F5

F20

F41

18

17

17

18

Figure A2. The replyecall interval in searching A. makarovi males. Duration of female
reply: control treatment F10: 10.4 s; F5 treatment: 5.2 s; F20 treatment: 20.8 s; F41
treatment: 41.6 s. Values obtained in the control F10 treatment are shown in black. Box
and whisker plots show the median (black or white line), the 25e75% interquartile
range (boxes), the lowest and the highest data points still within 1.5 times the interquartile range (whiskers) and outliers (circles). N number of males.