Behavioral Ecology Vol. 10 No. 6: 636-640
© 1999 International Society for Behavioral Ecology
Cooperatively breeding Arabian babblers call differently when mobbing in different predator-induced situations
a Institut für Verhaltensbiologie, Freie Universität Berlin, Haderslebener Strasse 9, 12163 Berlin, Germany b Hazeva Field Study Center, Arava, Israel
Address correspondence to M. Naguib. E-mail: mnaguib{at}zedat.fu-berlin.de .
Received 25 November 1999; revised 22 February 1999; accepted 8 April 1999.
 |
ABSTRACT |
|---|
 TOPFOOTNOTES ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
Cooperatively breeding Arabian babblers (Turdoides squamiceps) have a repertoire of different calls that they use in predator-induced contexts. We investigated their vocal mobbing behavior in two different predator-induced situations. We presented territorial groups of babblers a perched, stuffed owl representing an avian predator and a cat representing a ground predator. Babblers approached in both situations and mobbed the predators with the same two call types. In both predator-induced situations their first call was a short, metallic-sounding "tzwick." In response to the cat, babblers continued to primarily use tzwicks. However, in continued response to the owl, the babblers primarily used long trills. The experiments indicate that differences in use of two common call types during mobbing provide information on differences in predator-induced situations, although no call type per se identified a specific situation. The experiments suggest that the short tzwicks indicate a higher risk or urgency than trills and that combinations of both call types may provide graded information about differences in predator-induced situations. The relatively stable groups of these cooperative breeders might have favored evolution for using different calls in different mobbing situations.
Key words: alarm calling, Arabian babblers, cooperative breeders, mobbing, predator—prey interactions, Turdoides squamiceps.
|
INTRODUCTION |
|---|
|
TOPFOOTNOTESABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
Many animals use specific call types in response to predators. Typical alarm calls are associated with startle or escape reactions and usually are given when predators pose an immediate threat. These alarm calls can be highly specific with respect to the kind of predator or threat (Blumstein and Armitage, 1997
;
Evans et al., 1993;
Klump and Shalter, 1984;
Seyfarth and Cheney, 1980;
Struhsaker, 1967). A
strikingly different reaction is to approach the predator and mob it by
performing conspicuous vocal and/or visual displays
(Altmann, 1956;
Curio, 1978;
Curio et al., 1978;
Kruuk, 1976;
Marler et al., 1992;
Pettifor, 1990;
Stone and Trost, 1991). In
contrast to typical alarm calls, call types given during mobbing behavior have
not been described as providing specific information on the predator-induced
situation.
Call specificity, with respect to a particular kind of predator or threat,
might be adaptive when different predators require different flight reactions
(Seyfarth and Cheney, 1980) so
that group members that have not yet detected the predator can respond
appropriately. In contrast, species such as sciurid rodents, which have a
similar reaction to a broad range of different predators, use different
combinations of alarm calls to indicate the degree of risk
(Blumstein, 1995;
Loughry and McDonough, 1988).
Unlike alarm calls, which commonly are directed at other potential prey (e.g.,
Klump et al., 1986), mobbing is
often a group display that involves approach toward the predator. Thus, in
contrast to typical alarm calls, selection on mobbing calls is unlikely to
result from benefits a caller might gain by allowing the often familiar group
members to escape appropriately. However, often not all individuals in a group
participate in mobbing. Group members that do not join the mobbers then
potentially could extract information about the kind of predator or threat
and, over time, on the frequency at which the predator visits the area.
Specific calls then might increase group mates' future vigilance with respect
to particular kinds of predators or risk. In birds that live in relatively
stable groups, as cooperative breeders do, evolution might in theory favor
calling that provides such information to other group members. Alternatively,
different mobbing calls could influence detected predators in different ways
(e.g., different calls types might be more efficient in distracting different
predators).
In this paper, we present a study on use of calls during mobbing in two
different predator-induced contexts in Arabian babblers (Turdoides
squamiceps). These birds are cooperative breeders with groups usually
consisting of a single breeding pair and several mostly related helpers.
Arabian babblers respond vocally to a wide range of predators by using a
variety of calls. Flying raptors usually are indicated by loud barks, whereas
other predators such as mammals, perched birds of prey, and snakes are
approached and mobbed intensively (Zahavi
and Zahavi, 1997). Babblers mob snakes by approaching closely and
performing conspicuous vocal and visual displays. Babblers also mob mammalian
or perched avian predators vocally, but visual displays are less conspicuous
compared to their reaction to snakes
(Zahavi and Zahavi, 1997).
During mobbing, babblers use primarily two call types: trills and
"tzwicks" (Figure
1). Here, we examined babblers' use of these most frequent call
types in two different predator-induced contexts. We exposed groups of
babblers to a cat representing a ground predator and an owl representing an
avian predator. The results allowed us to assess whether babblers use
different mobbing calls in differing predator contexts.
|
|
METHODS |
|---|
|
TOPFOOTNOTESABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
General
We conducted field experiments from 27 February to 6 March 1997 at the Shezaf Nature Reserve in the Arava Valley and around the Hazeva Field Study Center, approximately 30 km south of the Dead Sea in Israel. Groups of babblers defend year-round territories of 0.2-1.0 km2 or of up to several kilometers in length along dry river beds (wadis) with few trees and widely spaced bushes. Groups of babblers in this area have been studied more or less continuously since 1971 (details in Ostreiher, 1997
;
Zahavi, 1990;
Zahavi and Zahavi, 1997). Size
of our study groups varied from two to nine birds, with an average of five
individuals per group. Babblers in this area are well habituated to presence
of humans, so approaching them closely did not cause appreciable changes in
their behavior.
Presentation of an owl
We conducted the experiments throughout the day by presenting a stuffed
short-eared owl (Asio flammeus) to 11 groups. To 10 of these groups
we also presented the cat, as described in the next section, in a balanced
design with about 10 min between the end of the first treatment to the
beginning of the second treatment. Vocal and other behavioral reactions always
ended when we removed the predator and babblers had ceased to respond before
the beginning of the second treatment. The owl was fixed in a natural position
on a short branch of a small section of a tree. In each trial we carried the
owl in a box to a location within 15-30 m of the babblers. We then either
positioned the owl in a tree at about 2 m height, or, if no convenient tree
was close by, we set it on top of its box, about 50 cm above ground. Each
trial lasted for 2 min after the babblers had given the first call.
We monitored the babblers' movements and recorded vocal responses using Sennheiser ME 66 microphones and a SONY TC-D5M stereo tape recorder.
Presentation of a cat
We presented a caged live cat (Felis communis) to 13 groups, 10 of
which also were used in the experiment with the owl. To control for effects of
a specific cat, a different individual was captured in the research station on
each of the mornings of the 2 experimental days. The cat was kept in a wire
cage (80 x 40 x 40 cm) until the experiments ended in the
afternoon. We transported the cat in a Landrover which we used to move between
groups and made sure that the cat was always supplied with sufficient water.
Just before and immediately after each trial, we covered the cage with a sheet
to prevent babblers from seeing the cat and to reduce the time the cat was
exposed to the sun.
In each case we presented the cat at about 15-30 m from the babblers. We positioned the covered cage on the ground at least 2 m away from dense bushes. At the beginning of a trial we removed the sheet. Usually the cat sat or stood in the cage and visually scanned its environment without displaying high activity. The cat vocalized occasionally, but we could not detect changes in babblers' behavior with regard to the cat's activities. We recorded babblers responses as described for the experiments with the owl.
Analysis of responses
We measured vocal responses in 18 experiments (10 with the owl and 8 with
the cat) because in 6 experiments babblers did not respond, presumably because
they did not detect the predator. Groups often were feeding by moving along
their territories and sometimes were about to move away from the predator
model before we uncovered it. Because not all groups responded to both models,
we had to restrict the statistical analysis to the six groups that responded
to both models. This permitted use of pairwise comparisons and avoided
statistical problems associated with pooling dependent and independent data.
We standardized variables in the time domain by setting the first call to a
latency of zero and only analyzed responses in the subsequent 2 min after the
first call. This procedure excluded confounding effects of differences in the
latency of subjects to respond (vocally), as this latency can be influenced by
a variety of factors other than those associated with our predators. Babblers
that were about to move in our direction may have detected the predators
earlier than babblers that were not moving toward us. In addition, some
individuals often act as sentinels by perching on top of a bush, which may
have resulted in earlier detection of the predator compared to groups in which
all individuals were feeding in the vegetation.
Because the different measures on the use of the two types of calls were correlated with each other (Table 1), we used six measures of response in a principal components analysis (SPSS 6.1 for windows, unrotated factor solution) and used the scores on the first component as a combined measure of response for each trial for further statistical analysis. The data set appeared to be suitable for such an analysis (Kaiser-Meyer-Olkin measure of sampling adequacy = 0.819, Bartlett test of sphericity = 50.86, p < 0.000), making repeated tests on correlating variables superfluous. The first principal component had an Eigenvalue of 4.3 and explained 72% of variance in the data. All variables loaded heavily on the first principal component. The variables and their loadings were (1) latency to the first trill (0.935), (2) duration of trilling within the first 2 min (-0.671), (3) duration in which tzwicks were given within the first 2 min (0.962), (4) number of tzwicks in the first 30 s (0.808), (5) number of trills in the first 30 s (-0.771), and (6) number of tzwicks given before the first trill was given (0.899). We measured the duration in which trills or tzwicks were given by measuring the lengths of bouts in which each call type was given. Calls were included in the same bout if they were separated by <10 s. Bouts of trills and tzwicks sometimes overlapped so that their combined bout length could add up to more than the 120 s observation time.
|
We then used the scores on the first principal component in a Wilcoxon
matched-pairs signed-rank test. For this test and a sign test on the number of
call types in the first 10 s, we followed the exact procedure for small sample
sizes (Mundry and Fischer,
1998; Siegel and Castellan,
1988).
|
RESULTS |
|---|
|
TOPFOOTNOTESABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
Presentation of predators
In 16 of the 18 trials in which subjects responded vocally, babblers gave both call types, tzwicks and trills, within the first 2 min. In all but three trials, the babblers' first vocal response was a tzwick, regardless of which predator was presented. In these three trials, all in response to the owl, the first vocalization was a trill. In six trials (three in response to the cat and three in response to the owl), additional calls (barks) were given occasionally, and in three trials we recorded other types of calls, which were given at low volume.
Pairwise comparison of the scores on the first principal component indicated that babblers used the two primary types of calls differently in response to the owl than in response to the cat (T+ = 21, p =.031, n = 6; Wilcoxon matched-pairs signed-rank test, two-tailed). In all six groups, babblers gave more tzwicks before they started to trill in response to the cat than they did in response to the owl (Figures 1a and 2). Furthermore, in the first 30 s babblers gave more tzwicks in response to the cat than they did in response to the owl, but they trilled more in response to the owl than they did in response to the cat (Figures 1b and 2). They also differed in the number of call types given within the first 10 s in response to the cat compared to their response to the owl (p =.032, n = 6, Sign test, two-tailed). Within the first 10 s all groups gave only tzwicks in response to the cat, but they all additionally trilled in response to the owl.
|
In summary, babblers gave both call types in both situations but used more tzwicks in response to the cat and more trills in response to the owl.
|
DISCUSSION |
|---|
|
TOPFOOTNOTESABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
The experiments indicate that Arabian babblers did not use different types of calls when mobbing in the two different predator-induced contexts. In all groups, babblers approached and gave calls of both types, tzwicks and trills, in both predator-induced situations. Interestingly, however, they used these two call types differently in the two situations.
Such differences in use of discretely different call types in different
predator-induced situations to our knowledge has not yet been described for
situations in which the prey approaches the predator and mobs it. Studies on
use of different calls in different predator-induced situations have focused
on typical alarm calls associated with flight reactions rather than with
approach to the predator (Blumstein and
Armitage, 1997; Evans et al.,
1993; Seyfarth and Cheney,
1980). In fact, if different predators warrant specific rapid
flight reactions, natural selection is expected to favor calls that transfer
information about the kind of predator or the required flight reaction,
provided that the caller benefits by warning others. When potential prey
approaches a predator and performs conspicuous displays, the context for the
evolution of vocal signals is different compared to typical alarm calling.
Thus, our findings force us to expand the view on evolution of predator
context—specific calling beyond typical alarm calls that are associated
with flight reactions. Differences in use of call types during mobbing
presumably have no immediate benefits in the way presumed for alarm calling.
Callers might use the calls to indicate their own assessment of the situation,
and differential use of calls during mobbing might be used by group members to
obtain information on the predator-induced situation. The composition of
call-type sequences then could reflect gradual differences in threat that are
easier to discern than when such information is conveyed by variation of only
one call type (Leger et al.,
1979; Macedonia,
1990; Owings and Virginia,
1978), particularly when different individuals assess the
situation differently and thus use different call types
(Todt and Naguib, in press).
Clearly, mobbing calls might affect predators, too, but we so far have no
argument for why the predator leads to direct selection for different mobbing
calls, unless some calls are more effective in deterring specific predators.
Babblers' intensive vocal mobbing of snakes
(Zahavi and Zahavi, 1997),
which are unable to hear within that frequency range, clearly indicate that
calling is not necessarily directed at the predator.
Because our models differed in several parameters, we cannot conclude that
primary use of specific call types indicates a specific type of predator. The
almost exclusive use of tzwicks as the first call suggests that tzwicks
indicate higher urgency, risk, or uncertainty—such as when the predator
is first detected and when the situation has not yet been fully evaluated or
when the predator poses an immediate threat. The continuous high rate of
tzwicks in response to the cat supports this interpretation because the cat
was alive and generally is more of an immediate threat than an owl. The owl,
in contrast, differed in two crucial ways. First, the stuffed model might have
been perceived by babblers after a short period as being less threatening
(Curio, 1993). However, in
recent experiments with an imitation of an owl that could move, babblers also
responded by primarily using trills (Mundry et al., unpublished data),
indicating that at least the lack of movement of our model was not a cause for
high rates of trills. Second, short-eared owls primarily hunt at night, so the
immediate threat was comparatively low. Mobbing influences owls to move their
daytime roosts, as shown by Pavey and Smyth
(1998), so that mobbing in
this case has long-term effects. Also, the structure of calls is consistent
with the argument that the two different call types reflect different degrees
of risk or urgency. Calls that indicate immediate threat have to be short to
allow a quick warning of conspecifics or a quick flight reaction of the
caller. Long calls, such as the babblers' trills, are not well suited for
symbolizing immediate threat but are more effective in transferring
information that requires no rapid reaction.
Differences in the use of call types when mobbing in different
predator-induced situations in babblers contrasts with vocal mobbing described
for other passerines studied so far. A major difference between babblers and
other passerines in which mobbing has been studied is that babblers live in
relatively stable groups of familiar and mostly related individuals. Thus, it
is tempting to speculate that these stable groups of cooperative breeders
might have played a role in providing the context for selection to favor vocal
specificity in mobbing behavior. For instance, distant group mates may obtain
information about the kind of threat, so that they can assess the situation
before approaching or even without approaching at all. Callers may benefit
from this behavior, for instance, when individuals switch roles, as they do
when acting as sentinels. Specific calls additionally might increase group
mates' future vigilance with respect to particular kinds of predators, as
argued earlier in this paper. Zahavi
(1990) and Zahavi and Zahavi
(1997) further suggested that
mobbing in babblers is a signal associated with display of social status.
Although our experiments were not designed to test these ideas, it is
conceivable that conspicuous mobbing is likely to have consequences for social
relations between group members.
In summary, the clear differences in calling during mobbing in different
predator-induced situations complements findings of situationally specific
alarm calling. Differential use of mobbing calls in different predator-induced
contexts provides a new aspect for the general alarm call system, which
includes mobbing calls as well as typical alarm calls
(Klump and Shalter, 1984).
|
ACKNOWLEDGEMENTS |
|---|
|
TOPFOOTNOTESABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
We are grateful to Frank Veit for assistance in the field and Matthias Dietz for assistance during preliminary experiments with captive babblers in Berlin. In addition, we thank Amotz Zahavi for his support and many critical discussions and Joe Waas and three anonymous referees for fruitful comments on the manuscript. The study was supported by a grant of the German-Israeli Foundation for Scientific Research and Development (G.I.F.: research project No. I 063-114.91/94) provided to Amotz Zahavi and Dietmar Todt.
|
FOOTNOTES |
|---|
|
TOPFOOTNOTESABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
L. Schrader is now at the ETH in Zürich, Switzerland.
|
REFERENCES |
|---|
|
TOPFOOTNOTESABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
Altmann SA, 1956. Avian mobbing behavior and predator recognition. Condor 58:241-253.
Blumstein D, 1995. Golden marmot alarm calls. I. The production of situationally specific vocalizations. Ethology 100:113-125.
Blumstein D, Armitage KB, 1997. Alarm calling in yellow-bellied marmots I: the meaning of situationally variable alarm calls.Anim Behav 53:143-171.
Curio E, 1978. The adaptive significance of avian mobbing I. Telenomic hypothesis and predictions. Z Tierpsychol 48:175-183.
Curio E, 1993. Proximate and developmental aspects of antipredator behavior. Adv Study Behav 22:135-238.
Curio E, Ernst U, Vieth W, 1978. The adaptive significance of avian mobbing II. Cultural transmission of enemy recognition in blackbirds: effectiveness and some constraints. Z Tierpsychol 48:184-202.
Evans CS, Evans L, Marler, P 1993. On the meaning of alarm calls: functional reference in an avian vocal system. Anim Behav 39:785-796.
Klump GM, Kretzschmar E, Curio E, 1986. The hearing of an avian predator and its prey. Behav Ecol Sociobiol 18:317-323.
Klump GM, Shalter MD, 1984. Acoustic behaviour of birds and mammals in the predator context. I. Factors affecting the structure of alarm signals. II. The functional significance and evolution of alarm signals. Z Tierpsychol 66:189-226.
Kruuk H, 1976. The biological function of gulls' attraction towards predators. Anim Behav 24:146-153.
Leger DW, Owings DH, Boal LM, 1979. Contextual information and differential responses to alarm whistles in California ground squirrels. Z Tierpsychol 49:142-155.
Loughry WJ, McDonough CM, 1988. Calling and vigilance in California ground squirrels: a test of the tonic communication hypothesis.Anim Behav 36:1533-1540.
Macedonia JM, 1990. What is communicated in the antipredator calls of lemurs: evidence from playback experiments with ringtailed and ruffed lemurs. Ethology 86:177-190.
Marler P, Evans SE, Hauser MD, 1992. Animal signals: motivational, referential, or both? In: Nonverbal vocal communication (Papousek H, Jürgens U, Papousek M, eds). Cambridge: Cambridge University Press;66 -86.
Mundry R, Fischer J, 1998. Use of statistical programs for nonparametric tests of small samples often leads to incorrect P values: examples from Animal Behaviour. Anim Behav 56:256-259.[Web of Science][Medline]
Owings DH, Virginia RA, 1978. Alarm calls of California ground squirrels (Spermophilus beecheyi). Z Tierpsychol 46:58-78.
Ostreiher R, 1997. Food division in the Arabian babbler nest: adult choice or nestling competition? Behav Ecol 8:233-238.
Pavey CR, Smyth AK, 1998. Effects of avian mobbing on roost use and diet of powerful owls, Ninox strenus. Anim Behav 55:313-318.[Web of Science][Medline]
Pettifor RA, 1990. The effects of avian mobbing on a potential predator, the European kestrel, Falco tinnunculus.Anim Behav 39:821-827.
Seyfarth RM, Cheney DL, 1980. Vervet monkey alarm calls: semantic communication in a free-ranging primate. Anim Behav 28:1070-1094.
Siegel S, Castellan NJ, 1988. Nonparametric statistics for the behavioral sciences, 2nd ed. New York: McGraw-Hill.
Stone E, Trost CH, 1991. Predators, risks and context for mobbing and alarm calls in black-billed magpies. Anim Behav 41:633-638.
Struhsaker TT, 1967. Auditory communication among vervet monkeys (Cercopithecus aethiops). In: Social communication among primates (Altman SA, ed). Chicago: University of Chicago Press; 281-324.
Todt D, Naguib M, in press. Vocal interactions in birds: the use of song as a model in communication. Adv Study Behav.
Zahavi A, 1990. The Arabian babbler. In:Cooperative breeding in birds (Stacey PB, Koenig WD, eds). London: Cambridge University Press;103 -130.
Zahavi A, Zahavi A, 1997. The handicap principle: a missing piece of Darwin's puzzle. Oxford: Oxford University Press.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  M. Griesser Mobbing calls signal predator category in a kin group-living bird species Proc R Soc B, August 22, 2009; 276(1669): 2887 - 2892. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Krama and I. Krams Cost of mobbing call to breeding pied flycatcher, Ficedula hypoleuca Behav. Ecol., January 1, 2005; 16(1): 37 - 40. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||