| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Behavioral Ecology Vol. 13 No. 5: 664-669
© 2002 International Society for Behavioral Ecology
Conditional strategies in territorial defense: do Carolina wrens play tit-for-tat?
University of North Carolina at Chapel Hill, Chapel Hill, NC 27599, USA
Address correspondence to J. Hyman, who is now at the Department of Biology, Duke University, Box 90338, Durham, NC 27708, USA. E-mail: jhyman{at}duke.edu.
Received 23 May 2001; revised 9 January 2002; accepted 14 January 2002.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Neighboring territorial animals are viewed primarily as intense rivals, but there are also opportunities for cooperation among competitors. Many animals respond less aggressively toward neighbors than to strangers. This phenomenon, termed the "dear enemy" effect, should be stable only when the reduced aggression is reciprocal. Territory owners should use conditional strategies in territorial defense, showing reduced aggression toward neighbors who cooperate by respecting territorial boundaries but increasing aggression toward invading neighbors. In this study I examined the response of territory owners to playbacks of neighbors at shared boundaries before and after intrusions by that neighbor or by strangers. Results showed that territory owners did not increase their aggression toward defecting neighbors but did increase their aggression toward neighbors after a simulated intrusion by a stranger. This surprising result might reflect the long-term relationship between neighboring Carolina wrens and the threat posed by rare intruding strangers.
Key words: communication, cooperation, territoriality, tit-for-tat, Thryothorus ludovicianus, wrens.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Neighboring territorial animals are often viewed primarily as rivals competing for mates and space. Competitive interactions, such as displays and physical contests, may indeed be common, but not all interactions between neighbors are so clearly antagonistic. Indeed, in many cases, animals respond less aggressively toward neighbors than they do to strangers, a phenomenon termed the "dear enemy" effect (Fisher, 1954
). Reduced
aggression toward familiar competitors could be beneficial to a territorial
animal if it decreases the likelihood of escalated contests whose outcome
should be predictable. Reduced aggression could be beneficial for both parties
by allowing neighbors to spend less time and energy defending shared
territorial boundaries (Eason and Hannon,
1994; Logan and Wingfield,
1990).
But reduced aggression could be costly. A lowered response to a neighbor
might be inappropriate if that individual continues to contest territorial
boundaries and attempts to expand his or her territory. Many territorial
animals increase their territory size when neighboring competitors are removed
(Adams and Tschinkel, 2001;
Both and Visser, 2000;
Koskela et al., 1999). Though
larger territories are not necessarily beneficial, expansion in the absence of
neighbors suggests that competition from neighbors limits territory size.
Stable boundaries resulting from reduced aggression toward neighbors come at
the cost of relinquishing the possible benefits of territorial expansion.
Thus, there is a cost to reducing aggression toward neighbors, and the benefit
of reduced aggression can only be realized if the cost is paid by both
territorial combatants. In this way, the territorial dear enemy effect
resembles cooperation.
Cooperation requires an individual to act in a manner potentially costly to
itself for the benefit of another. The evolution of stable cooperation
requires either that cooperators receive some future benefit or that the
apparently cooperative act actually carries no cost ("by-product
mutualism"; Dugatkin,
1997). There is a potential cost to reduced aggression, however,
if it leads to invasions by neighbors or if it means that a territory cannot
be expanded, suggesting that the dear enemy effect is not by-product
mutualism. When cooperation involves a cost, a possible mechanism for
achieving stable co-operation is reciprocal altruism, where pairs of
individuals trade bouts of cooperative behavior with one another
(Dugatkin, 1997;
Trivers, 1971).
Both Getty (1987) and
Godard (1993) have suggested
that the interaction between territorial neighbors may be modeled as a
Prisoner's Dilemma game, in which two individuals have a chance to cooperate
or defect in interactions with each other, and mutual cooperators fare better
than mutual defectors (Dugatkin,
1997). Defectors paired with cooperating individuals fare best,
whereas a cooperator paired with a defector gets the lowest payoff. In the
context of the dear enemy effect, cooperation could mean showing low
aggression toward neighbors and respecting territorial boundaries, whereas
defecting could mean showing high aggression or even attempting to expand a
territory. Dear enemy cooperation could be explained by reciprocal altruism if
territorial neighbors use conditional strategies such as tit-for-tat
(Axelrod and Hamilton, 1981;
Getty, 1987;
Godard, 1993;
Trivers, 1971). In the
tit-for-tat strategy, a subject will cooperate when its partner cooperates and
defect when the partner defects. If territorial neighbors use conditional
strategies in contesting territory boundaries, we would expect an increase in
aggression toward defecting individuals. Godard
(1993) found such a
conditional response in hooded warblers (Wilsonia citrina), which
increase their aggression toward neighbors who appear to cross territorial
boundaries.
In this study I attempted to replicate the study of Godard
(1993) in a species with
year-round territoriality, the Carolina wren (Thryothorus
ludovicianus). Carolina wrens show neighbor/stranger discrimination
(Shy and Morton, 1986) and the
dear enemy effect (Hyman,
2001), reacting more aggressively toward strangers than toward
neighbors. Carolina wrens often expand their territories to encompass areas
formerly held by neighbors that have disappeared
(Morton and Shalter, 1977),
and territory size is density dependent
(Haggerty and Morton, 1995;
Simpson, 1982). These
observations suggest that mutual acceptance of territorial boundaries in
Carolina wrens requires birds to forgo the possible benefits of larger
territories.
The stability of Carolina wren territories and the long-term relationships
between neighbors contrast with the short-term territories of hooded warblers,
defended only during the breeding season. The strategies that territory owners
might use in response to neighbors and strangers could vary greatly depending
on their population demographics. To examine the generality of Godard's
(1993) study, I used playback
experiments to examine the response of territorial male Carolina wrens to
invasions by neighbors and strangers. Following the methods of Godard
(1993), I presented
territorial males with intrusions of either strangers or familiar neighbors.
Playbacks of a neighbor presented well within the subject's territory
simulated an intrusion, or defection, by that neighbor, but playbacks of a
stranger simulated an intrusion by an unknown individual. Measuring the
responses of subjects to playbacks of neighbors at shared territorial
boundaries before and after intrusions allowed me to test if territory owners
increase their aggression toward defecting neighbors. A simulated intrusion by
a stranger served to determine whether higher aggression toward neighbors
results from a specific response to a defecting neighbor or from high level of
aggression in general. If the subjects increase their aggression specifically
toward defectors, I expected to see an increase in aggression only after the
simulated intrusion of a neighbor. If subjects increase their aggression after
an intrusion as a result of general stimulation, I expected increased
aggression toward neighbors after either type of intrusion.
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
I conducted this study at the Mason Farm Biological Reserve in Chapel Hill, North Carolina, USA. I performed experiments in April and May 1999, during the breeding season of Carolina wrens. All territorial birds included in this study as subjects or neighbors were members of mated pairs and had been present on their territory for several months. I estimated territory boundaries by extensive mapping of the locations of singing birds.
I recorded song with a Sennheiser K3U/ME88 ultradirectional microphone and a Marantz PMD221 recorder. I used songs recorded from males in 1998 and 1999 to construct playback tapes. Clear examples of songs were chosen after examining them with a Uniscan II real-time spectrum analyzer and digitized at 16 kHz and 16-bit accuracy on a 68030 Macintosh computer using Audiomedia hardware and software (Digidesign, Palo Alto, California). All songs were adjusted to the same maximal amplitude and rerecorded on a Marantz PMD221 recorder. Playback tapes were played from a Marantz PMD221 recorder connected to an Amplivox amplifier and a Realistic horn speaker (frequency response 1.5-2.5 kHz ± 3dB, 2.5-8 kHz ± 2.5 dB). All playbacks were conducted from 0700-0900 h Eastern Standard Time.
Playback tapes were 45 s long and consisted of 7 repetitions of the same
song type, delivered at a rate of approximately 10 songs/min. These tapes
simulated natural singing behavior, as males typically repeat the same song
type 5-250 times and at a rate of 5-15 songs per minute
(Haggerty and Morton, 1995).
Neighbors' territories directly adjoined the territory of the subject.
Stranger's songs were recorded from birds with territories > 5 territory
diameters distant.
Song exemplars used as neighbor or stranger song were not matched with
respect to song type. I did not determine the song repertoires of subjects and
neighbors; therefore, I do not know if stranger song types were present in the
repertoires of neighbors. Song-type sharing between neighboring male Carolina
wrens can be high (66%; Haggerty and
Morton, 1995), and song-type sharing is also likely to be > 50%
over the short distance between subjects and strangers in this experiment
(Morton, 1987). Therefore, it
is likely that stranger song types could have been in the repertoires of
neighbors.
Thirteen subjects each received two playback treatments. One treatment measured the response of the subject to a neighbor before and after a simulated territorial intrusion by that neighbor (NNN trials); the other treatment measured the response of the subject to a neighbor before and after a simulated intrusion by a stranger (NSN trials). I randomized the order of presentation such that half of the subjects received the NNN trial first and half received the NSN trial first. For each subject, the NNN trial and the NSN trial were separated by 7-10 days. Both treatments followed a protocol with three playbacks and three observation periods. The first playback simulated the presence of a neighbor at a boundary shared with the subject. The speaker was placed at a position 5 m beyond the last mapped singing post for the subject and clamped to a sapling 1.5-2 m off the ground. I chose a point beyond the last mapped singing post to assure that the location designated for a playback at the boundary did not simulate a neighbor invading the subjects' territory. Playbacks began when both the subject and neighbors were silent. I used song exemplars from 13 different neighbors for playback, such that each subject was presented with songs from a different neighbor to avoid pseudoreplication. Subjects' responses were recorded for 15 min from the start of the playback.
After 25 min, I presented a second playback near the center of the subject's territory, simulating a deep intrusion. In the simulated neighbor invasions, each male heard the same playback tape that had been used in the first boundary playback. I simulated stranger invasions using song exemplars from 13 different strangers for playback, such that each subject was presented with songs from a different stranger to avoid pseudoreplication. I conducted the playback and observation period as above. In some cases, it was necessary to wait longer than 25 min before the subject and neighbors were silent.
After another 25 min, I presented a third playback, again simulating the presence of a neighbor at the shared territorial boundary. I conducted the playback and observation period as above. For each subject, I used the same neighbor song exemplars as above to simulate a neighbor at a shared boundary. Once again, it was necessary in some cases to wait longer than 25 min for the subject to become silent before the playback began.
I recorded the response variables latency to approach, closest approach,
latency to sing, and number of songs. These variables tended to be correlated
and were combined using principal component analysis to calculate a composite
score for statistical analysis. I used the first principal component scores in
a Wilcoxon signed-ranks test to compare playbacks and treatments. Power
analyses were performed following Cohen
(1977).
To determine if invasions by neighbors and strangers had different effects
on subject birds, I examined the response of a subject to a neighbor at a
shared boundary before and after an invasion by that neighbor and a stranger.
In this way, for both treatments, I was able to determine if subjects
responded differently to a neighbor at a territory boundary after a simulated
intrusion. These methods differed from those of Godard
(1993) in several respects,
primarily in that the intrusions I used were shorter in duration. I examine
the influence of these methodological differences in the Discussion.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Neighbor invasion trials
There was little difference in a subject's response to playbacks of neighbors at a boundary before and after a simulated intrusion by that neighbor (Figure 1). The mean responses suggest that after the simulated intrusion, subjects responded to a neighbor with a slower latency to approach, similar closest approach and latency to sing, and a lower song output. The principal component loadings for the response variables are in Table 1. Comparing the scores of the first principal component showed no difference in the subjects' response to neighbors before and after a simulated neighbor intrusion (Figure 2; Wilcoxon matched-pairs test: Z = -0.078, p = .937). Comparing the response to a neighbor before and during an invasion clearly shows that subjects responded more strongly to the invasion (Figure 2; Wilcoxon matched-pairs test: Z = -2.197, p = .028).
|
|
|
Stranger invasion trials
Unlike the NNN trials, there were significant differences in the subjects'
response to neighbors at the boundary before and after an intrusion by a
stranger. After the intrusion by a stranger, subjects responded to neighbors
with shorter latency to approach, closer approach, a shorter latency to sing,
and more songs (Figure 3).
Comparing the first principal component scores, I found that subjects
responded significantly more strongly to neighbors after an intrusion by a
stranger (Figure 2; Wilcoxon
matched-pairs test: Z = -2.981, p = .003).
|
Godard (1993) found an
increase in response after both neighbor and stranger invasions, as might be
expected. However, the magnitude of the change in response was much greater
after the neighbor invasion. Surprisingly, I found no increase in aggression
after neighbor invasions, but to facilitate a comparison with Godard's
(1993) results, I also
analyzed these results by comparing the magnitude of the mean difference in
response to neighbors before and after invasions in the two trials. Comparing
these two means, I found a significantly greater change in response after the
stranger invasions than after neighbor invasions (Wilcoxon matched-pairs test:
Z = 2.197, p = .028).
Comparison of invasions by neighbors and strangers
There was no clear difference in the way subjects responded to playbacks of
neighbors or strangers at the center of the territory
(Figure 4). There was a trend
to approach strangers' playbacks more slowly, but also to approach more
closely. Comparing the first principal component scores, however, there was no
significant difference in the subject's response to neighbors and strangers at
the center of the territory (Figure
2; Wilcoxon matched-pairs test: Z = -0.863, p =
.388).
|
As mentioned above, it was sometimes necessary to wait longer than 25 min after the invasion playback for the subject to be silent and for the next playback to begin. I considered the possibility that males responding for longer periods of time (longer than 25 min) might also respond more strongly to subsequent playbacks. However, there was no significant correlation between the strength of a subject's response to a playback at the center of his territory and his subsequent response to a neighbor at the boundary after the invasion (NNN trials rs = .266; NSN trials rs = .378). Thus, the strength of a subject's response to an invasion did not predict the strength of response to subsequent playbacks of neighbors.
The responses of the neighbors to playbacks on the territories of the subjects were noted as well. Previous experiments have demonstrated no consistent difference in the way neighbors respond to playbacks of themselves or of strange birds (Hyman, unpublished data). In the present study, neighbors frequently counter-sang with subjects responding to playback, but in no trial simulating a playback at a boundary did a neighbor respond more strongly that the intended subject. In this sense, there was no evidence of any defensive coalitions being established. Thus, the responses of neighbors were not considered to have had any consistent effect on the behavior of subjects.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
I examined the response of territorial males to two kinds of simulated invasions: The NNN trials simulated an intrusion by a neighbor deep into the subject's territory. The simulated invasions elicited a high level of aggression, but there was no continuing effect in the form of a significant change in the subject's response to the neighbor at the boundary after the intrusion. Thus, subjects evidently did not retaliate against defecting neighbors.
The NSN trials simulated an intrusion by a stranger. If Carolina wrens use their discrimination abilities to establish dear enemy cooperation with neighbors, intrusions by strangers should have no effect on the subjects' relationships with neighbors. Contrary to this prediction, subjects responded significantly more strongly to neighbors at the boundary after an intrusion by a stranger.
The statistical power of the NNN trials to detect a difference of the size
found in the NSN trial is high (power = 0.81, following
Cohen, 1977), consistent with
the fact that a significant effect was detected with the same birds and the
same sample size in the NSN trials. Therefore, I have confidence that there
was not an undetected difference in the subjects' responses to neighbors after
an invasion. At the least, the magnitude of the increase in response after
stranger invasions is much greater than any possible increase in response to
neighbor invasions.
There was no significant difference in subjects' initial responses to
invasions by neighbors or strangers. The power of this test to detect a
difference of the size seen in the NSN trials was not as high (power = 0.66).
The lack of a difference in response to neighbor or stranger invasions does
not imply that Carolina wrens are not dear enemies, however. A differential
response to neighbors and strangers is clearly observed at a boundary
(Hyman, 2001), but not at the
center of a territory. This is consistent with the idea that neighbors at a
boundary represent less of a threat than do strangers. The relative threat of
a neighbor increases after he invades the territory, and an invading neighbor
could represent as great a threat as an invading stranger
(Stoddard, 1996).
The only other study to examine responses of subjects to cooperating or
defecting neighbors, found that hooded warblers responded more strongly to a
neighbor after the neighbor intruded, but not after a stranger intruded
(Godard, 1993). It appears
that hooded warblers use strategies like titfor-tat, responding specifically
to defecting neighbors with higher aggression. Carolina wrens thus differed
from hooded warblers in a similar experiment in that wrens did not show a
simple tit-for-tat response by becoming more aggressive toward defecting
neighbors. Instead, they increased their response to neighbors after invasions
by strangers. Carolina wrens discriminate between the songs of neighbors and
strangers (Hyman, 2001;
Shy and Morton, 1986). Thus,
it is unlikely that the increase in response to neighbors after stranger
invasions is caused by confusion about the identity of the intruder.
Additionally, the differential response after invasions by neighbors and
strangers indicates a clear ability to discriminate.
My methods followed those of Godard
(1993), but with a few
differences. First, Godard simulated a neighbor at a boundary with a playback
located 10 m inside the subject's territory. My boundary playbacks were
located 5 m beyond the subject's territorial boundary. Second, Godard
performed two intrusion playbacks per trial, whereas I simulated only a single
intrusion. Finally, during the intrusions, Godard played songs until the
subject approached within 10 m, whereas in the present study, all intrusion
playbacks lasted 45 s. These differences make it likely that subjects in the
present study were provoked to a lesser degree, which could explain the lack
of increased aggression toward intruding neighbors, but not the increased
level of aggression toward neighbors after an intrusion by a stranger.
Ecological differences between Carolina wrens and hooded warblers might
account for the behavioral differences in response to neighbors and strangers.
Carolina wrens and hooded warblers differ considerably in terms of the length
of interaction of neighbors and the likelihood of encountering strangers.
Neighboring territorial hooded warblers interact over the course of 3-4 months
on their breeding grounds (Evans Ogden and
Stutchbury, 1994). Territory owners might encounter strangers
regularly because young birds settle territories in the spring, and males
unsuccessful in attracting a mate may switch territories (Wiley RH, personal
communication). In year-round territorial Carolina wrens, young birds settle
on territories in their first fall
(Haggerty and Morton, 1995).
Thus, Carolina wrens in spring have been neighbors from 7 months to several
years, and intruding adult strangers in spring are rare. The increased length
of interaction between neighboring Carolina wrens could select for a more
forgiving strategy, overlooking the occasional intrusion as soon as the
neighbor returns to his territory. For Carolina wrens, the intrusion of an
unknown competitor might cause such high alarm that an increased aggressive
response carries over toward all potential rivals, including neighbors. In
this sense, the increase in response would not be a specific adaptation
directed toward neighbors, but rather, subjects reacting aggressively toward
neighbors could still be reacting to the earlier invasion by a stranger.
Additionally, all territory owners in a neighborhood might react differently
after an invasion by a stranger. An invasion by a stranger could suggest that
a territory owner is at severe risk for territory loss. If neighbors come to
probe the subject's ability to defend his territory, subjects might need to
reassert their territorial or social status, as either neighbors or mates
could be eavesdropping on these interactions
(McGregor and Dabelsteen,
1996; Naguib et al.,
1999; Otter et al.,
1999). However, I found no evidence that the strength of subjects'
reaction toward stranger invasions predicted his subsequent response to
neighbors at a boundary, though it is possible that responses to invasions by
neighbors and strangers differed in ways I did not measure.
There is also a difference in the type of threat posed by invading
neighbors and strangers. Carolina wrens and hooded warblers also differ in
their extrapair breeding behavior. In multiple studies of hooded warblers,
18-47% of females had at least one extrapair young, and the extrapair father
is often a neighbor (Evans Ogden and
Stutchbury, 1994). Carolina wrens apparently do not engage in
extrapair copulations (Haggerty et al.,
2001). Short-term invasions by neighbors could result in loss of
paternity in hooded warblers, but not in Carolina wrens. For hooded warblers,
the threat of cuckoldry by neighbors might represent a greater threat than
invasions by strangers. For Carolina wrens, invading strangers might be a more
significant threat because they present an unprecedented threat to usurp a
male's territory. In both species, the pattern of aggression might simply
reflect a continuing aggressive response after the most threatening
intrusion.
The acceptance of mutual territorial boundaries in Carolina wrens requires
territory owners to forgo the possible benefits of larger territories. If we
define this scenario as a game where cooperation means showing low aggression
at boundaries and defection means showing high aggression and trying to expand
territory boundaries, the payoffs would be as follows. A wren able to expand
his territory (defector) without challenge from a neighbor (cooperator)
clearly has a higher payoff than any male who cooperates. Mutually cooperating
males achieve dear enemy benefits, while continually defecting males would be
in a constant fight over territory boundaries. Finally, mutually defecting
males would be expected to do better than a cooperator paired with a defector
because the cooperator would inevitably lose the benefits of holding a
territory if intruders are unchallenged. Thus, in this scenario, the payoffs
fit the conditions for the Prisoner's Dilemma. The lack of increased
aggression toward defecting neighbors does not necessarily indicate that
neighboring Carolina wrens are not in a Prisoner's Dilemma. Rather, the
failure to increase aggression toward intruding neighbors could indicate that
Carolina wrens either do not use tit-for-tat at all or use more forgiving
strategies than hooded warblers, such as generous tit-for-tat
(Nowak and Sigmund, 1992),
which allows for occasional unchallenged defections by neighbors.
Ydenberg et al. (1988)
proposed the asymmetric war of attrition rather than reciprocal altruism to
explain the dear enemy effect. In this model, extended fighting occurs when
unfamiliar pairs of competitors both think they are likely to win a contest.
Familiar neighbors rarely engage in escalated fights because they make
educated assessments of the likely outcome of a fight. This model, or
additionally, the "fighting-to-learn" model proposed by Getty
(1989), are not necessarily
alternatives to reciprocal altruism models of the dear enemy effect. These
models provide proximate explanations for why neighbors stop fighting over
shared boundaries, but they do not take into account mutual benefits that
accrue from a reduction in aggression between established neighbors. The
mechanism that results in lower aggression toward a neighbor does not
necessarily account for the evolution of reciprocally lowered aggression
between neighbors. The Prisoner's Dilemma provides a framework to understand
when reciprocal restraint can evolve.
It is worth emphasizing that dear enemies in some species have not agreed
to a truce in all aspects of their behavior. Radio-tracking studies show that
territorial neighbors make frequent foraging trips into neighboring
territories (Jansen, 1999;
Zach and Falls, 1979), and DNA
fingerprinting reveals that neighbors are competitors for extrapair
copulations in most species (Kempenaers et
al., 1999; Langefors et al.,
1998; Westneat,
1990). These areas of competition do not prevent neighbors from
attaining dear enemy cooperation, suggesting that the benefits of dear enemy
cooperation must be strong. Possible benefits of dear enemy cooperation,
including decreased fighting (Eason and
Hannon, 1994; Logan and
Wingfield, 1990), establishing defensive coalitions
(Elfstrom, 1997;
Getty, 1987), and avoiding the
appearance of territory instability that may attract floaters
(Beletsky, 1992), could lead to
increased reproductive success (Beletsky
and Orians, 1989). The widespread existence of the dear enemy
effect might be explained by the variety of benefits available to familiar,
cooperative neighbors. However, the strategies that territorial animals use to
achieve dear enemy cooperation might differ considerably, depending on the
possible benefits and on the population dynamics that underlie the formation
of territorial neighborhoods.
|
ACKNOWLEDGEMENTS |
|---|
I thank Haven Wiley, Helmut Mueller, Steve Nowicki, Ken Lohmann, Alan Feduccia, Cindy Hogan, Will Mackin, Jon Micancin, Amy Skypala, Joanna Vondrasek, and Barbara Ballentine for their help with the manuscript. This work was supported in part by a Wilson Award from the Department of Biology, UNC-Chapel Hill.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Adams ES, Tschinkel WR, 2001. Mechanisms of population regulation in the fire ant Solenopsis invicta: an experimental study. J Anim Ecol 70: 355-369.
Axelrod R, Hamilton WD, 1981. The evolution of
cooperation. Science 211:
1390-1396.
Beletsky L, 1992. Social stability and territory acquisition in birds. Behaviour 123: 290-313.[Web of Science]
Beletsky LD, Orians GH, 1989. Familiar neighbors
enhance breeding success in birds. Proc Natl Acad Sci USA
86: 7933-7936.
Both C, Visser ME, 2000. Breeding territory size affects fitness: an experimental study on competition at the individual level. J Anim Ecol 69: 1021-1030.
Cohen J, 1977. Statistical power analysis for the behavioral sciences: New York: Academic Press.
Dugatkin LA, 1997. Cooperation among animals: An evolutionary perspective. New York: Oxford University Press.
Eason P, Hannon SJ, 1994. New birds on the block—new neighbors increase defensive costs for territorial male willow ptarmigan. Behav Ecol Sociobiol 34: 419-426.[Web of Science]
Elfstrom ST, 1997. Fighting behaviour and strategy of rock pipit, Anthus petrosus, neighbours: cooperative defense. Anim Behav 54: 535-542.[Web of Science][Medline]
Evans Ogden LJ, Stutchbury BJ, 1994. Hooded warbler. In: The birds of North America, no. 110 (Poole A, Gill F, eds). Washington, DC: Academy of Natural Sciences.
Fisher J, 1954. Evolution and bird sociality. In: Evolution as a process (Huxley J, Hardy A, Ford E, eds). London, Allen and Unwin; 71-83.
Getty T, 1987. Dear enemies and the Prisoner's Dilemma: why should territorial neighbors form defensive coalitions? Am Zool 27: 327-336.
Getty T, 1989. Are dear enemies in a war of attrition? Anim Behav 37: 337-339.
Godard R, 1993. Tit for tat among neighboring hooded warblers. Behav Ecol Sociobiol 33: 45-50.
Haggerty TM, Morton ES, 1995. Carolina wren. In: The birds of North America, no. 188 (Poole A, Gill F, eds). Washington, DC: Academy of Natural Sciences.
Haggerty TM, Morton ES, Fleischer RC, 2001. Genetic monogamy in Carolina wrens (Thryothorus ludovicianus). Auk 118: 215-219.[Web of Science]
Hyman J, 2001. Territorial communication in a songbird, the Carolina wren (PhD dissertation). Chapel Hill: University of North Carolina.
Jansen A, 1999. Homeranges and group territoriality in Chowchilla. Emu 99: 280-290.[Web of Science]
Kempenaers B, Congdon C, Boag P, Robertson R, 1999.
Extrapair paternity and egg hatchability in tree swallows: evidence for
genetic compatibility hypothesis? Behav Ecol
10: 304-311.
Koskela E, Mappes T, Ylonen H, 1999. Experimental manipulation of breeding density and litter size: effects on reproductive success in the bank vole. J Anim Ecol 68: 513-521.
Langefors A, Hasselquist D, Von Schantz T, 1998. Extra-pair fertilizations in the sedge warbler. J Avian Biol 29: 134-144.
Logan CA, Wingfield JC, 1990. Autumnal territorial aggression is independent of plasma testosterone in mockingbirds. Horm Behav 24: 568-581.[Medline]
McGregor PK, Dabelsteen T, 1996. Communication networks. In: Ecology and evolution of acoustic communication in birds (Kroodsma DE, Miller EH, eds). Ithaca, New York: Cornell University Press; 409-425.
Morton ES, 1987. The effects of distance and isolation on song-type sharing in the Carolina wren. Wilson Bull 99: 601-610.
Morton ES, Shalter MD, 1977. Vocal response to predators in pairbonded Carolina wrens. Condor 79: 222-227.[Web of Science]
Naguib M, Fichtel C, Todt D, 1999. Nightingales
respond more strongly to vocal leaders of simulated dyadic interactions.
Proc R Soc Lond B 266:
537-542.
Nowak MA, Sigmund K, 1992. Tit for tat in heterogeneous populations. Nature 355: 250-253.[Web of Science]
Otter K, McGregor PK, Terry Andrew MR, Burfurd Fiona RL, Peake T,
Dabelsteen T, 1999. Do female great tits assess males by
eaves-dropping? A field study using interactive playback. Proc R Soc
Lond B 266:
1305-1309.
Shy E, Morton ES, 1986. The role of distance, familiarity, and time of day in Carolina wren responses to conspecific songs. Behav Ecol Sociobiol 19: 393-400.[Web of Science]
Simpson BS, 1982. Communication with complex vocal repertoires by a territorial passerine, the Carolina wren (PhD dissertation). Chapel Hill: University of North Carolina.
Stoddard PK, 1996. Vocal recognition of neighbors by territorial passerines. In: Ecology and evolution of acoustic communication in birds (Kroodsma DE, Miller EH, eds). Ithaca, New York: Cornell University Press; 356-374.
Trivers R, 1971. The evolution of reciprocal altruism. Q Rev Biol 46: 35-57.
Westneat D, 1990. Genetic parentage in the indigo bunting: a study using DNA fingerprinting. Behav Ecol Sociobiol 27: 67-76.
Ydenberg RC, Giraldeau LA, Falls JB, 1988. Neighbours, strangers and the asymmetric war of attrition. Anim Behav 36: 343-347.
Zach R, Falls JB, 1979. Foraging and territoriality of male ovenbirds in a heterogeneous habitat. J Anim Ecol 48: 33-52.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  J. Chaves-Campos, Y.-m. Araya-Ajoy, C. A. Lizana-Moreno, and K. N. Rabenold The effect of local dominance and reciprocal tolerance on feeding aggregations of ocellated antbirds Proc R Soc B, November 22, 2009; 276(1675): 3995 - 4001. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||