Female brown-headed cowbirds' (Molothrus ater) social assortment changes in response to male song: a potential source of public information

Abstract

In many species, females' behavior appears to be influenced by that of other females, particularly regarding mate choice. Females theoretically can reduce the costs associated with independent male assessment by observing conspecifics. Studies of brown-headed cowbirds suggest that females pay attention to other females' behavior. Group-housed females modify their song preferences, whereas females housed in pairs do not. What information is available to females in a group environment? To address this question, we studied two groups of juvenile (i.e. hatch-year birds) and adult female cowbirds in a naturalistic group setting. We used a longitudinal ABA design, consecutively introducing and removing males that differed in age, amount of song production and stage of song development, to isolate the male characteristics that related to changes in female behavior. Juvenile and adult females assorted by age class when singing adult males were in the aviary, but not when singing juveniles or silent males of any age class were in the aviary. Results from playback tests confirmed that adult male song alone influenced female age class assortment. Videotape analysis from playback tests revealed that females also wing stroked in response to male song. Other females sometimes approached females who wing stroked and observed them. We hypothesize that group-level changes in social organization and individual females' responses can serve as visual signals for other individuals.

Full text

Page 1

Abstract In many species, females’ behavior appears to
be influenced by that of other females, particularly re-
garding mate choice. Females theoretically can reduce
the costs associated with independent male assessment
by observing conspecifics. Studies of brown-headed
cowbirds suggest that females pay attention to other
females’ behavior. Group-housed females modify their
song preferences, whereas females housed in pairs do
not. What information is available to females in a group
environment? To address this question, we studied two
groups of juvenile (i.e. hatch-year birds) and adult
female cowbirds in a naturalistic group setting. We used
a longitudinal ABA design, consecutively introducing
and removing males that differed in age, amount of song
production and stage of song development, to isolate the
male characteristics that related to changes in female be-
havior. Juvenile and adult females assorted by age class
when singing adult males were in the aviary, but not
when singing juveniles or silent males of any age class
were in the aviary. Results from playback tests con-
firmed that adult male song alone influenced female age
class assortment. Videotape analysis from playback tests
revealed that females also wing stroked in response to
male song. Other females sometimes approached fe-
males who wing stroked and observed them. We hypoth-
esize that group-level changes in social organization and
individual females’ responses can serve as visual signals
for other individuals.
Keywords Cowbird · Preference · Public information ·
Social assortment · Wing stroke
Introduction
In many species, females appear to attend to the behav-
iors of female conspecifics. Most often studied in the
context of mate choice, females’ grouping patterns, pref-
erences and even mate choices can be influenced by
other females (Gibson et al. 1991; Shuster and Wade
1991; Dugatkin 1992; Gibson and Hoglund 1992;
McComb and Clutton-Brock 1994; Jennions and Petrie
1997; Hoglund et al. 1995; White and Galef 1999;
Westneat et al. 2000). By observing conspecifics,
females theoretically can reduce the costs of sampling or
errors associated with choosing a mate and, in some
cases, they can gain information from more knowledge-
able individuals (Losey et al. 1986; Wade and Pruett-
Jones 1990; Gibson and Hoglund 1992; Pruett-Jones
1992; Stohr 1998; Agrawal 2001; Sirot 2001). Female
brown-headed cowbirds (Molothrus ater) appear to be
influenced by other females’ preferences when housed in
a group environment (Freeberg 1998; Freeberg et al.
1999), whereas laboratory studies have been unable to
find evidence of such malleability (King and West 1983;
see also Baptista et al. 1993for parallels in female song
learning in white-crowned sparrows Zonotrichia leuc-
ophrys). Thus, a group setting appears to provide oppor-
tunities for learning that are unavailable to females
living in more restrictive housing. But what is the mech-
anism that can account for this difference? Do females’
responses in a group provide different information than
their responses in pairs?
We were led to consider social assortment as a mea-
sure of responsiveness based on a series of studies look-
ing at social development of cowbirds housed in large
aviaries. In 12 groups in previous studies, we have found
that social organization by age ad by sex is a robust mea-
sure of social influence on the development of male
courtship behavior in cowbirds living in large aviaries
(Freeberg 1999; Smith et al. 2002; King et al. 2002;
White et al. 2002b; Whiteet al. 2002c). In addition, West
et al. (2002) found that social assortment of females var-
ied as a function of the age-sex class of birds with whom
Communicated by: W.A. Searcy
J. Gros-Louis (✉) · D. J. White · A. P. King · M. J. West
Department of Psychology,
Indiana University,
1101 E.10th St., Bloomington, IN 47405 USA
e-mail: jgroslou@indiana.edu
Tel.: +1-812-3394494, Fax: +1-812-8554691
Behav Ecol Sociobiol (2003) 53:163–173
DOI 10.1007/s00265-002-0560-5
ORIGINAL ARTICLE
Julie Gros-Louis · David J. White · Andrew P. King
Meredith J. West
Female brown-headed cowbirds’ (Molothrus ater)
social assortment changes in response to male song:
a potential source of public information
Received: 24 May 2002 / Revised: 30 October 2002 / Accepted: 3 November 2002 / Published online: 11 December 2002
© Springer-Verlag 2002

Page 2

they were housed. Females showed segregation by age
class when housed with males (see also Smith et al.
2002; King et al. 2002) but not when housed only with
females. In conditions with adult males, females had
more male neighbors, approached males more often, and
assorted more strongly by age than when housed with ju-
venile males (White, unpublished data). Furthermore, in
the breeding season, the females in conditions with adult
males copulated more and produced more fertile eggs as
compared to the females who were housed with juvenile
males.
Taken as a whole, the results of West et al. (2002) in-
dicated that females are sensitive to the variables of sex
and age and correlated behaviors when housed in flocks
and that this sensitivity has reproductive consequences.
The design of the West et al. study did not allow us to
determine proximate influences on social responsiveness
because the females remained in the same social groups
all year. For this reason, in this study, we turned to a lon-
gitudinal ABA design to examine how different types of
male stimulation affected social organization of females
in the fall.
In addition to documenting changes in females’ group
behavior, we sought to identify behavioral responses that
may reflect females’ assessment of males. Do females
respond in particular ways that are publicly informative
about males? Do females perhaps also perform species-
typical signals, seen in other contexts, when females in-
teract in groups? For example, we know from laboratory
studies that females respond to a small percentage of
songs with rapid flicks of their wings, termed wing
strokes (West and King 1988a). If a female exhibits a be-
havioral response such as this in a more social context,
not only does the male receive feedback, but so might
females who are in close proximity.
To explore the nature of female responsiveness to
males, we studied two groups of juvenile (i.e. hatch-year
birds) and adult females in outdoor aviaries during the
fall. We consecutively introduced and removed males
that differed in age, amount of song production and stage
of song development, to isolate which characteristics of
males related to changes in female behavior. Widowski
et al. (1990,1992) successfully used a similar within-sub-
jects experimental design to determine the male charac-
teristics that influenced female reproductive condition in
cotton-top tamarins (Saguinus oedipus). Because we
needed to compare the relative influence of particular
male characteristics on female behavior, we used the
same females to avoid a confound due to inter-female
variability in responsiveness (King and West 1989). We
asked how females responded to male presence and, in
the absence of males, how females responded to male
song as determined by playback tests.
Methods
General
Subjects
We used 62 wild-caught brown-headed cowbirds from Monroe
County, Indiana as subjects in experiments conducted between
7 September and 14 December 2001. We trapped all of the juve-
niles between 12 July and 19 August 2001. We determined the sex
and age of the juveniles based on the mature plumage that had
emerged (Selander and Giller 1960). We judged juveniles to be be-
tween 40 and 50 days old. We trapped four adult males (>2 years)
between 5 and 10 July 2001. We trapped the other adult birds in
the previous 2 years and housed them with birds of different age-
sex classes for various experiments. We used unique combinations
of colored plastic leg bands to identify individuals.
The subjects of the experiments were 13 juvenile females and
16 adult females. We randomly assigned six juveniles and eight
adults to one aviary (aviary 1), and seven juveniles and eight
adults to the other (aviary 2). We consecutively added and re-
moved other individuals, 6 adult females, 18 adult males, and 9 ju-
venile males, over the course of five experiments.
The two aviaries were large outdoor enclosures (9.1×21.4×
3.4 m) containing grass, trees, indoor and outdoor perches, and
feeding stations. A shelter building separated the aviaries and thus
birds were visually but not acoustically isolated from one another.
Birds experienced weather conditions and were able to see and
hear wild local conspecifics and predators. We provided the birds
daily with vitamin-treated water, a modified Bronx zoo diet for
blackbirds, and a mixture of white millet, red millet, and canary
seed. In addition, the birds foraged freely.
Procedure
The study consisted of 15 rounds of data collection within a series
of five experiments to observe changes in female social organiza-
tion and behavior in response to the introduction of males with
different characteristics into the aviaries. Each round represented a
different phase within an experiment and thus will be explained
below for each experiment. However, here we provide a general
overview of the procedure.
In the first experiment, we observed only one aviary. We first
took baseline data to document levels of female age class assort-
ment in the absence of males. We then introduced males to the fe-
males to observe changes in female age class assortment in re-
sponse to male presence. For the final four experiments, we used a
modified ABA design in the two aviaries (baseline-experimental
introductions-reversal of introductions-return to baseline) to docu-
ment the immediate effects of introducing males with different
characteristics on female age class assortment. We first collected
baseline data to document female age class assortment before
male introductions into the aviaries. We sequentially added and re-
moved males who were either ‘singing’ or ‘non-singing’ (see be-
low) using a counter-balanced design to control for order effects.
Lastly, we removed the males and collected data to document
whether females returned to baseline levels of age class assort-
ment. For each social manipulation, we added or removed birds
one day prior to the commencement of data collection. We did not
use the same males in more than one experiment.
In rounds 12–15, we performed playback tests within the avi-
aries to determine more precisely what role male song had on fe-
male age class assortment. First we performed playbacks in the
presence of ‘non-singing’ juvenile males (as controls) and then in
the absence of males.
Data collection
To measure age class assortment, we recorded near neighbor (NN)
associations in 7-min sampling blocks. We defined near neighbors
164

Page 3

as two birds within approximately 30 cm of one another. Observ-
ers identified one bird (the "target") and recorded the identity of
any bird within 30 cm (the "near neighbor") and then located an-
other target bird. The same individuals could not be recorded as
near neighbors again in the same block unless they moved away
and re-associated. For example, if A and B were within 30 cm of
one another, they were counted as near neighbors only once in a
data block (A as B’s neighbor or B as A’s neighbor, but not both)
unless they moved apart and came to within 30 cm of one another
again during that block. Individuals could serve as the target mul-
tiple times in the same block.
To document singing behavior, we collected data ad libitum on
all songs produced by males during the 7-min sampling blocks
(Altmann 1974). A solo bout was defined as a series of ten undi-
rected songs in a minute. If a male sang a solo bout, we noted it
and stopped collecting song production data for that male during
that block.
Three to four observers sequentially collected typically four
blocks of data each over a period of 3–9 days for every round of
data collection (see below for the exact numbers of days and data
blocks within each round). Due to weather conditions, the num-
bers of days varied between experiments, but were roughly equal
within experiments. We started each round of data collection on
the morning following the introduction of new birds into an avi-
ary. We collected a total of 94.4 h of data totaling 32,266 NN
points.
Observers spoke the color band acronyms of birds that were
NN into omni-directional, wireless lapel microphones (Telex WT
150; Telex Communications). A receiver (Telex FMR 150) trans-
mitted each observer’s vocal signal to a Pentium III 500-MHz
IBM-compatible computer running Microsoft Windows 1998. We
used voice recognition software (IBM ViaVoice Millenium Pro
Edition) and Microsoft Word 2000 to transcribe the speech into
text. We exported data into a database (4th Dimension v. 6.5.1;
ACI) that we programmed to compare incoming bird identification
bands and data codes with a list of possible acronyms to detect
and correct errors automatically (White et al. 2002a).
Statistical analysis
Due to small sample sizes and heterogeneity of variance, we per-
formed non-parametric statistical analyses throughout. In addition,
because of small samples sizes, we combined the results from the
two aviaries for statistical analysis. We used each female’s propor-
tion of NN associations with adult females in Wilcoxon signed-
ranks tests to document changes in individual females’ assortment
within and between rounds in each experiment. Using this experi-
mental design, each female served as her own control. We report
the mean change (±SE) in the proportion of NN associations for
females within each age class for each comparison to illustrate
changes in female assortment. Although we combined data from
the two aviaries, we provide separate figures for each aviary
(Fig. 1a, b) to illustrate the variance. See Table 1 for an overview
of the experimental manipulations performed in each aviary.
Experiment 1: presence of adult males
We had documented in a previous study that females or-
ganized themselves by age class in the presence of adult
males, but not in their absence (West et al. 2002). We
wanted to replicate and extend these findings to deter-
mine whether females were influenced by mere visual
contact with adult males or whether adult males needed
to be physically present in the aviary with females.
Therefore, the purpose of the first experiment was to
investigate how the presence of adult males influenced
female age class assortment.
Methods
Procedure
We collected NN association data in aviary 1 for 7 days,
for a total of 47 blocks, to determine the level of female
age class assortment in the absence of adult males. We
then placed three adult males in a mobile cage
(2.5×1.5×1.4 m) just outside of the aviary. The cage was
within 2 m of where the females frequently perched. We
collected 31 data blocks during the 5-day collection peri-
od. We recorded no singing from the males until one
male sang a solo bout at the end of the last 7-min data
block. In the final round of this experiment, we intro-
165
Fig. 1 The percentage of juvenile and adult female near neighbor
associations with adult females for each round of the five experi-
ments in a aviary 1 and b aviary 2. Percentages were calculated by
averaging each female’s mean number of near neighbor associa-
tions with adult females. Asterisks mark the introduction of ‘sing-
ing’ adult males into the aviary and the presentation of adult male
song during playback trials. AM Adult male, AF adult female,
PLBK playback, NS no song (baseline), ■ ■ juvenile female, ● ● adult
female. See text for further explanation

Page 4

duced the three adult males from the mobile cage into
the aviary with the females to determine the effect of the
presence of males on age class assortment. Thus, in this
round, females could interact with adult males rather
than merely being in visual contact with them.
Once in the aviary, two of the three adult males sang
solo bouts in 17 of the 56 7-min data blocks. We ana-
lyzed the NN association data separately for blocks with
song (n=17) and blocks without song (n=39).
Results
The juvenile and adult females assorted by age class on-
ly when adult males were present in the aviary, particu-
larly when they were singing. Baseline levels of age
class assortment showed that the juvenile and adult fe-
males (JF and AF respectively) did not associate prefer-
entially with one age class (mean proportion of JF-JF as-
sociations=0.55±0.02; AF-AF mean=0.49±0.02; Fig. 1a).
The juvenile and adult females did not increase their NN
associations with females of the same age class in re-
sponse to the visual presence of males outside of the
aviary (JF-JF mean change from baseline=−0.05±0.03:
Wilcoxon T=3, n=6, P>0.5; AF-AF=0.01±0.03: Wilcoxon
T=20, n=8, P>0.8). When we placed males inside the
aviary, both the juvenile and the adult females associated
significantly more with individuals of the same age class
than when males were outside of the aviary (JF-JF mean
increase=0.07±0.02: Wilcoxon T=21, n=6, P<0.04; AF-
AF=0.09±0.03: Wilcoxon T=35, n=8, P<0.02).
Male song appeared to have the strongest effect on fe-
male social organization. Based on observations during
which males were singing (mean no. of solo bouts =
6.3±4.9), the juvenile females significantly increased
their NN associations with juvenile females compared to
when males were not singing (JF-JF mean increase=
0.17±0.03: Wilcoxon T=21, n=6, P<0.04). Adult females
also tended to associate more with adult females in re-
sponse to male song (AF-AF mean increase=0.08±0.03:
Wilcoxon T=31, n=8, P<0.08).
Experiment 2: effect of singing males
In experiment 1, we had observed that females assorted
by age class when adult males were present in the aviary
but more so during blocks when the males were singing.
Furthermore, females responded immediately to male
song; we saw effects within the 7-min blocks when song
was present. Because the males were singing throughout
the experiment, it was hard to isolate the effects of male
presence and male song. The purpose of the second ex-
periment was to determine whether the presence of
males, or singing males specifically, was responsible for
female age class assortment by introducing ‘singing’ or
‘non-singing’ males into each aviary.
Methods
Procedure
We selected ‘singing’ and ‘non-singing’ males based on
4 days of observations between 10 and 17 October 2001
conducted in the home aviary of adult males. We collect-
ed data ad libitum to note all songs produced by males
(Altmann 1974). The number of songs produced by 40
adult males ranged from 0 to 157. We defined ‘singing’
males as those males who sang often, compared to other
males in their home aviary (median no. of songs = 130).
‘Non-singing’ males were those males who never sang in
their home aviary. In the fall, unlike in the spring, sing-
ing behavior has no obvious relationship to dominance
(Dufty 1986).
166
Table 1 Summary of experimental manipulations, number of days and number of data blocks per experiment. AM adult male, AF adult
female, JM juvenile male, PB playback
ExperimentRound No. of data blocksDaysManipulation
Aviary 1Aviary 2
11
2
3
47
31
56
7
5
5
No AM present
3 AM outside aviary
3 AM inside aviary
21
2
3
4
25
45
41
41
8
5
5
4
No AM
3 singing AM
3 non-singing AM
No AM
No AM
3 non-singing AM
3 singing AM
No AM
3
4
1
1
2
19
45
57, 58 (av. 1,2)
7
5
5
3 AF
3 singing JM
6 singing AM
3 AF
3 singing AM
6 singing JM
51
2
3
4
5
44
37
26
12
12
4
3
3
3
3
Non-singing JM
PB AM song with JM
PB JM song with JM
PB AM song without JM
PB JM song without JM
Non-singing JM
PB JM song with JM
PB AM song with JM
PB JM song without JM
PB AM song without JM

Page 5

We first collected NN association data on females in
both aviaries for 8 days to document baseline levels of
age class assortment. After 8 days, we concurrently in-
troduced three ‘singing’ adult males into aviary 1 and
three ‘non-singing’ adult males into aviary 2. We ob-
served individuals for 5 days and collected 45 blocks of
NN data in each aviary. We then switched the adult
males housed in the two aviaries and again observed the
birds for 5 days and collected 41 blocks of NN data in
each aviary. Lastly, we removed the adult males from
both aviaries and collected 41 blocks of data over a 4-
day period to determine whether in the absence of the
males, females returned to baseline levels of age class
assortment.
Results
The presence of ‘singing’ adult males generally resulted
in female age class assortment, whereas the presence of
‘non-singing’ adult males did not. During the baseline
period, females did not assort by age class (JF-JF=
0.55±0.02; AF-AF=0.51±0.03). The juvenile and adult
females showed a significant increase in NN associations
with individuals of the same age class compared to base-
line levels when we introduced ‘singing’ adult males
(mean no. of solo bouts across all data blocks = 7.8±2.1)
(JF-JF mean increase=0.11±0.02: Wilcoxon T=91, n=13,
P<0.0005; AF-AF=0.05±0.02: Wilcoxon z=2.28, n=16,
P<0.03). When we introduced ‘non-singing’ adult males
(n=0 songs each across all data blocks) into the aviaries,
NN associations for the juvenile and adult females
returned to levels that were not significantly different
from baseline (JF-JF mean change=−0.01±0.02: Wil-
coxon T=31, n 13, P>0.5; AF-AF=0.01±0.02: Wilcoxon
z=0.78, n=16, P>0.4). In addition, NN association pat-
terns remained at levels indistinguishable from baseline
in the final phase of the experiment when we removed
the ‘non-singing’ adult males from the aviaries (JF-JF
mean change=0.01±0.02: Wilcoxon T=52, n=13, P>0.5;
AF-AF=−0.02±0.02: Wilcoxon z=1.03, n=16, P>0.3).
Experiment 3: presence of adult females
We conducted a third experiment to control for the possi-
bility that the mere introduction of any novel birds could
influence female age class assortment. We introduced
three adult females into each aviary to determine whether
they influenced the social organization of juvenile and
adult females, as did the adult males.
Methods
Procedure
We randomly selected six adult females from an all-adult
female aviary and randomly assigned three to each test
aviary. We collected 19 blocks of NN data over a 7-day
period. We compared levels of age class assortment after
the female introductions to baseline data taken at the end
of the previous round.
Results
When we introduced the adult females into the aviaries,
the juvenile and adult females did not show a significant
increase in the proportion of their NN associations with
females of their own age class compared to baseline lev-
els (JF-JF mean increase=−0.003±0.01: Wilcoxon T=24,
n=13, P>0.5; AF-AF=−0.001±0.02: Wilcoxon z=0.05,
n=16, P>0.8).
Experiment 4: effect of juvenile versus adult males
The purpose of this experiment was to compare the
effects of juvenile and adult males on female age class
assortment. We had seen an effect of singing adult males
on female assortment in experiments 1 and 2. Therefore,
we wanted to determine whether females responded to
singing males in general, regardless of their age.
Methods
Procedure
We selected ‘singing’ males using the same procedure as
described in experiment 2. We based our selections on
5 days of observations of juveniles in an all-juvenile avi-
ary and 9 days of observations of adults in an all-adult
aviary between 10 October and 7 November 2001. The
number of songs produced by 26 juvenile males ranged
from 0 to 150 (median no. of songs=89), whereas the
number of songs produced by 40 adult males ranged
from 0 to 259 (median no. of songs=160).
We first introduced three ‘singing’ juvenile males
(median no. of songs=130) into aviary 1 and three ‘sing-
ing’ adult males (median no. of songs=189) into aviary
2. We collected 45 blocks of NN associations over a
5-day period. For the reversal phase, we introduced new
males rather than switching the males between aviaries
because of low levels of singing by the original males.
We selected three new juvenile and three new adult
males from the all-juvenile and all-adult male aviaries.
We placed three new ‘singing’ adult males (median no.
of songs=163) into aviary 1 and three new ‘singing’
juvenile males (median no. of songs=71) into aviary 2.
During the first two days of observation, none of the
juvenile males produced any songs, and two of the three
adult males sang 10 and 20 songs, respectively, in 3 of
32 blocks. Therefore, we decided to add three more
males to each condition. We selected three more ‘sing-
ing’ juvenile and adult males (median no. of songs=71
and 54, respectively). We observed the birds for an addi-
167

Page 6

tional 5 days for a total of 58 and 57 blocks of data re-
spectively for aviaries 1 and 2. We compared levels of
age class assortment in the presence of ‘singing’ juvenile
and adult males to baseline levels obtained at the end of
the previous round.
Results
The juvenile and adult females assorted by age class in
the presence of ‘singing’ adult males but not in the pres-
ence of ‘singing’ juvenile males. When we introduced
the ‘singing’ juvenile males (mean no. of solo bouts
across all data blocks=4.7±1.7) into the aviaries, the ju-
venile and adult females allocated roughly equal propor-
tions of NN associations to females of each age class
that were not significantly different from baseline levels
(JF-JF mean change=−0.06±0.01: Wilcoxon T=4, n=13,
P>0.5; AF-AF=0.007±0.02: Wilcoxon z=0.31, n=16,
P>0.6). When we introduced ‘singing’ adult males
(mean no. of solo bouts across all data blocks=2.5±0.65)
into the aviaries, the juvenile females tended to increase
their NN associations with other juveniles but the change
was not significant (mean increase=0.04±0.02: Wilcoxon
T=70, n=13, P>0.09). The adult females allocated a sig-
nificantly greater proportion of their NN associations to
adult females compared to baseline levels in the presence
of ‘singing’ adult males (mean increase=0.08±0.02:
Wilcoxon z=3.15, n=16, P<0.002).
‘Singing’ adult males had a greater effect on NN as-
sociations than did ‘singing’ juvenile males. Juvenile
and adult females increased their within-age class NN
associations in response to the presence of ‘singing’
adult males compared to ‘singing’ juvenile males (JF-JF
mean increase=0.10±0.01: Wilcoxon T=20, n=13,
P<0.001; AF-AF mean increase=0.08±0.01: Wilcoxon
z=3.21, n=16, P<0.0002).
Experiment 5: effect of juvenile and adult male song
in the absence of singing males
The differential effect of adult males and juvenile males
on female age class assortment led us to question what
age-related cues were driving social reorganization
among females. Males’ size and morphological charac-
teristics did not seem to be a reliable cue; males varied
slightly in size and/or plumage coloration. By contrast,
juvenile and adult males clearly differed in their stage of
song development. In addition, juvenile and adult males
may have differed in their non-vocal behavior associated
with singing. Females could have been responding dif-
ferently to juvenile and adult males because of differ-
ences in their vocal or their non-vocal behavior. We de-
signed a series of playback tests to distinguish between
these two possibilities.
Methods
Subjects
The original two groups of females in aviary 1 and
aviary 2 served as subjects with the exception of one
juvenile female in aviary 1 who died during the course of
playback tests. In addition, we introduced six ‘non-sing-
ing’ juvenile males, three in each aviary, to serve as con-
trols for male presence during some of the playback tests
(see Procedure).
Playback stimuli. We created two CDs of juvenile male
song and two CDs of adult male song to be used as play-
back stimuli. We used songs that were recorded from
adult and juvenile males during the first week of October
1999. We recorded songs using a Sennheiser RF con-
denser microphone and a Sony TCD-D10 PRO II DAT
recorder at a distance of approximately 30 cm. We se-
lected, based on recording quality, a sample of 23 songs
from four juvenile males and 23 songs from four adult
males to use in playback sequences.
We only used songs from four juvenile and four adult
males because we wanted to simulate the introduction of
‘singing’ and ‘non-singing’ males that we had conducted
in previous experiments. The four adult males were se-
lected based on the amount of song they produced such
that they were representative of the distribution of total
song production by adult males (few–many). Their songs
were judged to be typical Indiana songs, containing com-
mon whistle types. The four juvenile males were hatch-
year birds, and therefore, we did not have any data on
their singing history. Adult and juvenile song differed in
that all of the adult songs contained both note clusters
and whistles, whereas the juvenile song was more vari-
able in its elements, structure and rhythm (see West and
King 1988b; Smith et al. 2000for a description).
To create playback sequences, we first digitized
the 46 selected songs by playing them on a Panasonic
SV-3700 professional DAT recorder that was connected
to an Apple 500 MHz G3 Powerbook running Sound
Edit 16 version 2.0.7. We digitized the calls at a sam-
pling rate of 44.1 kHz and created a sequence of 23
songs covering a 7-min period. Within the 7 min, we
created one solo bout, while the remainder of the songs
occurred at different intervals so that playback sequences
reflected a natural rate of singing by males at this time of
year. We used the same juvenile male and adult male
songs to make two different temporal sequences of song
for the juvenile and adult male playback stimuli.
Playback setup. We placed a speaker (AIWA LCX-350)
in each aviary 2 days before the experiment for a few
hours each day to habituate the birds to its presence. On
the days of playback tests, we placed the speakers in the
aviaries half an hour before the commencement of the
tests. To enhance the credibility of the playback stimuli,
we chose a position for the speakers where it would be
feasible for a bird to be perched out of view of the group.
168

Page 7

We positioned the speaker close to perches that the birds
frequently used. There were enough perches within one
meter of the speaker that potentially all of the birds
could position themselves in close range of the speaker.
We broadcast the playback stimuli with an AIWA
LCX-350 portable audio system. We selected a volume
level for broadcasting that produced realistic renditions
of the songs in the playback sequence and then measured
sound pressure levels to ensure that the amplitude for all
of the songs was similar to naturally occurring song. The
mean sound pressure levels of the 23 songs in each of
the playback sequences were 85.1±1.1 dB for juvenile
song and 88.3±0.83 dB for adult male song, as measured
at 1 m using a K & B 2209 sound pressure meter,
A-weighting, impulse reading.
Experimental procedure. We introduced three juvenile
males who had been in aviary 1 and had not sung during
experiment 4 into aviary 2. Similarly, we introduced
three juvenile males who had been in aviary 2 and had
not sung during experiment 4 into aviary 1. We did not
use new ‘non-singing’ males from the home aviaries of
all males because we wanted to be sure that the males
would not sing during the playback tests. We first col-
lected NN association data for 4 days to document the
baseline levels of female age class assortment in the
presence of the juvenile male controls.
We randomly assigned the playback stimuli (either
juvenile or adult male song) to each aviary for the first
3-day playback round. We then alternated the stimuli be-
tween the aviaries for each 3-day round that followed.
Subjects heard one playback sequence per day during the
3 days of playback tests. We presented playback stimuli
in the first block of four 7-min data blocks.
We next performed the reversal phase of the experi-
ment for replication by switching which playback stimuli
the females heard first in each aviary. During this round,
we alternated temporal playback sequences each day dur-
ing the 3-day round to reduce habituation to the sequences.
As a final test of the effect of male song on female
age class assortment, we removed the juvenile males
from each aviary and repeated the playback tests using
juvenile and adult male song in both aviaries. We fol-
lowed the same playback procedure as described above.
Playback observations. Three observers sequentially col-
lected NN association data for four 7-min blocks each
day during the first round of playback tests (with the
playback presented in the first of the four blocks as not-
ed above). For the last three rounds of playback tests, we
changed the data collection protocol to try to document
additional behavior of the females. Two observers simul-
taneously collected data during the four 7-min data
blocks each day. One observer recorded NN associa-
tions, while the other observer recorded approach data ad
libitum (Altmann 1974). Approaches documented the
initiation of NN associations. When one bird came to
within 30 cm of a second bird, we recorded the band of
the approaching bird and the band of the bird that was
approached. In the first of the last three rounds of play-
back tests, the third observer collected four 7-min blocks
of NN association data when the two observers finished
their four blocks of simultaneous data collection. In the
final two rounds of playback tests, the third observer
recorded the females’ behavior using a Sony Hi-8 DCR-
TRV330 video camera with a Sony ECM-737 electret
condenser microphone while the two observers were re-
cording data simultaneously.
Statistical analysis. The 7-min block of data occurring
during playback and the 7-min block of data collection
immediately following playback served as the experi-
mental data set (called ‘song blocks’). The final two
7-min data blocks served as data blocks to calculate lev-
els of age class assortment in the absence of song (‘post-
song blocks’). For those rounds in which the second and
third observers entered the aviaries sequentially after the
first observer conducted the playback experiment and
collected four blocks of data, their NN association data
were included with the ‘post-song’ data.
We used Wilcoxon signed-ranks tests to compare
females’ NN associations in the song blocks to their NN
associations obtained during the baseline round prior to
the playback tests. In addition, we compared NN associ-
ations in ‘song’ and ‘post-song’ data blocks within each
round. We also compared the proportion of each female’s
approaches to other females in ‘song’ blocks and ‘post-
song’ blocks. As with the first four experiments, we
combined data from the two aviaries for statistical ana-
lyses although the data from the aviaries are presented
separately in Fig. 1a and b.
We scored the behavioral responses of females that
we videotaped ad libitum (Altmann 1974) during the ex-
periment. We were constrained in our sampling proce-
dure because we had to be able to see the bands of the
birds for identification. Therefore, we opportunistically
sampled those females who were in the best position for
videotaping. Throughout the experiment, females often
moved in and out of view of the camera. Because fe-
males were not equally sampled, we were unable to con-
duct extensive statistical tests on the behavioral respons-
es of females. We report the behavioral changes that we
observed for descriptive purposes.
Results
Near neighbor associations
We observed changes in female age class assortment in
response to the playback of adult male song but not in
response to juvenile male song. During the baseline peri-
od, the juvenile and adult females in both aviaries did
not assort by age class in the presence of the ‘non-sing-
ing’ juvenile male controls (JF-JF=0.53±0.01; AF-AF=
0.51±0.01). When we performed the playback of adult
male song, the females associated significantly more
with females from the same age class compared to
baseline association patterns in the days prior to the start
169

Page 8

of the playback experiment (JF-JF mean increase=
0.11±0.03: Wilcoxon T=85, n =13, P<0.004; AF-AF
mean=0.10±0.02: Wilcoxon z=3.05, n=16, P<0.003). By
contrast, the juvenile and adult females did not increase
their proportion of NN associations with females of the
same age class from baseline levels when we presented
juvenile male song (JF-JF mean change=−0.03±0.02:
Wilcoxon T=24, n=12, P>0.5; AF-AF mean=-−0.01±
0.03: Wilcoxon z=0.78, n=16, P>0.4).
As a more precise analysis of females’ responses to
the broadcast of adult male song, we analyzed changes in
association patterns within the playback test. We found
that within consecutive data blocks females showed rap-
id changes in NN association patterns. Females associat-
ed significantly more with females of the same age class
in song blocks in which we played adult male song,
compared to post-song blocks (JF-JF mean increase=
0.12±0.02: Wilcoxon T=90, n=12, P<0.0005; AF-AF=
0.16±0.04: Wilcoxon z=3.15, n=16, P<0.002).
After removing the juvenile male controls from both
aviaries, we still observed significant changes in female
social organization in response to the playback of adult
male song. For these analyses, we compared data within
playback tests from song blocks and post-song blocks.
Both juvenile and adult females allocated more NN asso-
ciations to other adult females in response to adult male
song compared to when song was absent (JF-JF mean in-
crease=0.11±0.03: Wilcoxon T=90, n=12, P<0.0005;
AF-AF=0.11±0.03: Wilcoxon z=2.69, n=16, P<0.008).
As in the playback tests with the juvenile male con-
trols present in the aviaries, the juvenile and adult fe-
males did not assort by age class in song blocks with ju-
venile song compared to post-song blocks (JF-JF mean
change=0.03±0.03: Wilcoxon T=43, n=12, P>0.5;
AF-AF=0.02±0.03: Wilcoxon z=0.47, n=16, P>0.6).
Approaches
We analyzed female approach data for the two play-
back rounds where juvenile males were not present. We
found that both juvenile and adult females approached
adult females significantly more during song blocks than
post-song blocks (JF-AF mean number of approaches:
1.96±0.30 vs. 1.41±0.21: Wilcoxon T=69, n=12, P<0.02;
AF-AF mean=1.29±0.23 vs. 0.94±0.14: Wilcoxon z=100,
n=15, P<0.03). Furthermore, juvenile females approached
adult females more when they heard adult male song com-
pared to when they heard juvenile male song (2.09±0.36
vs. 1.26±0.24: Wilcoxon T=51, n=10, P<0.02). Adult fe-
males also approached adult females more when they
heard adult male song, but this finding was not significant
(1.35±0.12 vs. 0.86±0.22: Wilcoxon T=92, n=15, P>0.07).
Individual behavior
Although changes in female NN association patterns
occurred in response to male song, observations of indi-
vidual female behavior suggest that these changes
emerged over the course of the two song blocks. The
videotaped responses of individual females revealed
that they typically did not change their behavior during
or immediately following male song playback (n=518
of 592 observations of individual females). For example,
if a female was hopping down a perch prior to the
playback, she continued this behavior during the play-
back. In the 98 behavioral responses that occurred, we
recorded 11 different behaviors, 9 of which occurred in-
frequently (less than 3 times each) and thus will not be
discussed.
The majority of the females responded to the play-
backs of juvenile and adult male song. We were sur-
prised to record wing stroking responses by females, be-
cause we had observed this behavior previously only in
early spring in response to live males singing at close
range (King and West 1988). In response to the playback
of male song, we observed wing stroking by 7 of 12 dif-
ferent juvenile females and 10 of 16 adult females (n=33
occurrences). Although a larger proportion of adult fe-
males wing stroked to adult male song than did juvenile
females (AF=0.31; JF=0.08), this result was not signifi-
cant (Fisher’s exact test, n=28, P>0.2). A roughly equal
proportion of adult and juvenile females wing stroked to
juvenile male song (AF=0.56; JF=0.50). In addition, of
the 17 females who responded with wing strokes, there
were four adult females who were more responsive than
the other 13 females. The mean number of wing strokes
performed by these four females was significantly
greater than the mean number of wing strokes by the
other 13 females (mean number of wing strokes=4.3±0.6
vs. 2±0.3; Mann-Whitney U=62, n1=4, n2=13, P<0.002,
corrected for ties).
We also documented females moving either laterally
along a perch or hopping to a different perch in response
to male song (n=48 occurrences). Seven of 12 juvenile
female and 9 of 16 adult females moved in response to
song. Similar to the results obtained for the wing strok-
ing data, a non-significantly larger proportion of adult
females moved in response to adult male song than did
juvenile females (AF=0.25; JF=0.08; Fisher’s exact test,
n=28, P>0.4) whereas a roughly equal proportion of
adult and juvenile females moved when they heard juve-
nile male song (AF=0.44; JF=0.50).
Discussion
We documented different patterns of association and in-
dividual behavior in response to different male charac-
teristics in two groups of juvenile and adult female cow-
birds. The results from male introduction and playback
experiments indicate that female social organization
changed specifically in response to adult male song;
females showed no changes in association patterns in
response to the presence of silent males of either age
class. When we introduced singing adult males into the
aviaries, juvenile and adult females associated more with
170

Page 9

females of their own age class. By contrast, females did
not preferentially associate with females of the same age
class when we placed singing juvenile males in the
aviary. Playback experiments confirmed the results from
introduction experiments; adult song caused female age
class assortment.
Why would females assort by age class exclusively in
response to adult male song? Because near neighbor
associations are an outcome measure of female behavior,
we cannot determine what drove age class assortment.
Changes in social organization could result from quanti-
tative or qualitative differences in responses between age
classes. For example, one age class may respond more
than the other class. Or, it may be that one age class re-
sponds in a distinct way from the other age class, such as
moving toward or away from a singing male (or in this
case, the speaker).
Although we cannot identify the mechanism for age
class assortment, there are several possible functional
causes of assortment. In several species, females aggre-
gate to avoid harassment from males (e.g. McComb and
Clutton-Brock 1994; Cassini 2000). Adult females, who
have had experience with adult males in previous years,
may cluster together upon hearing adult male song, which
would leave juvenile females primarily with juvenile fe-
males. Age class assortment may also arise from female-
female mate competition (e.g. see Dugatkin 1992; Gibson
and Hoglund 1992; Jennions and Petrie 1997). Lastly,
age-specific responses by females to particular song char-
acteristics that are independent of other females’ respons-
es could lead to age class assortment (e.g. Grant and
Green 1996; Gibson et al. 1991; Slagsvold and Viljugrein
1999).
Our observations of the behaviors of females during
the playback phase suggest to us that the changes in so-
cial organization could function in the shaping of female
preferences (Freeberg 1998; Freeberg et al. 1999). We
base this conclusion on the fact that females approached
and observed females who wing stroked (Fig. 2). West
and King (1988a) showed that females wing stroked to
high quality songs, i.e., those songs most likely to stimu-
late copulatory responding later in the breeding season.
Wing strokes and song occur simultaneously, as do the
reactions of observers to them (Fig. 2). As a result, song
quality can potentially be communicated between non
singing females. What might be the consequences of this
communication? Juvenile females show significant vari-
ability in their copulatory responsiveness to adult songs
whereas adult females show strong agreement as to the
relative quality of the breeding season songs (King et al.,
submitted). Female wing stroking and approach could
provide the information necessary to fine tune individual
preferences, thereby reducing variability. Thus, juvenile
females may become more discriminating in their prefer-
ences through a longitudinal process of observing re-
sponses by adult females (e.g., see also Dugatkin and
Godin 1993; Stohr 1998).
It is not necessary to assume that the primary function
of wing strokes is to communicate about preferences;
rather, visual signals such as wing stroke responses may
function as an intersexual communicative signal that
can be co-opted by other females. Thus, wing strokes
and changes in social organization serve as a form of
‘public information’ that is readily available to others (cf.
Doligez et al. 2002; Nordell and Valone 1998; Templeton
and Giraldeau 1996; Valone 1989, 1996). Individuals in
a number of species are known to ‘eavesdrop’ on song
contests and fights between males. Information obtained
in these contexts appears to be used to assess mates
(Doutrelant and McGregor 2000; Mennill et al. 2002;
Otter et al. 1999) and competitors (Earley and Dugatkin
2002; Naguib et al. 1999; Oliveira et al. 1998; Peake et
al. 2001). Recent studies demonstrate that female song-
birds appear to assess males based on their song, possibly
because it indicates something about their song learning
ability (cf. Searcy et al. 2002; Nowicki et al. 2002).
Therefore, female cowbirds may obtain information
through ‘eavesdropping’ on female responses to male
song which may, in fact, be directed to males given that
wing strokes influence the development of male song
(West and King 1988a). Wing strokes during the fall
months may provide feedback to males at a time when
their songs are developing but, at the same time, wing
strokes also may have the potential to inform females
about other females’ preferences.
171
Fig. 2 Photographs depicting a videotaped sequence of a female’s
wing stroke response, followed by a neighboring female observing
her. The first frame shows the females at song onset. The second
frame shows one female wing stroking 100 ms after song onset.
The last frame shows a second female observing the female
659 ms after she wing stroked. Photographs were made using Stra-
ta Videoshop 4.5.1 to capture the frames of the video recorded
during the playback experiment

Page 10

Acknowledgements NSF for provided funding for M.J.W. and
A.P.K. NICHD provided J.G.L. with a post-doctoral fellowship
through a Developmental Training Grant awarded to The Depart-
ment of Psychology, Indiana University. D.J.W. received a post-
doctoral fellowship from NSERC. Maral Papahian assisted in data
collection. The protocols and research presented here were ap-
proved by The Institutional Care and Use Committee of Indiana
University (01–085). We would like to thank Dr. William Searcy
and three anonymous reviewers for their insightful comments and
suggestions.
References
Agrawal AF (2001) The evolutionary consequences of mate copy-
ing on male traits. Behav Ecol Sociobiol 51:33–40
Altmann J (1974) Observational study of behavior: sampling
methods. Behaviour 49:227–267
Baptista LF, Trail PW, DeWolfe BB, Morton ML (1993) Singing
and its functions in female white-crowned sparrows. Anim
Behav 46:511–524
Cassini MH (2000) A model on female breeding dispersion and
the reproductive systems of Pinnipeds. Behav Proc 51:93–99
Doligez B, Danchin E, Clobert J (2002) Public information and
breeding habitat selection in a wild bird population. Science
297:1168–1170
Doutrelant C, McGregor PK (2000) Eavesdropping and mate
choice in female fighting fish. Behaviour 137:1655–1669.
Dufty AM Jr (1986). Singing and the establishment and mainte-
nance of dominance hierarchies in captive brown-headed cow-
birds. Behav Ecol Sociobiol 19:49–55.
Dugatkin LA (1992) Sexual selection and imitation: Females copy
the mate choice of others: Am Nat 139:1384–1389
Dugatkin LA, Godin JGJ (1993) Female mate copying in the
guppy (Poecilia reticulata): age-dependent effects. Behav Ecol
4:289–292
Earley RL, Dugatkin LA (2002) Eavesdropping on visual cues
in green swordtail (Xiphophorus helleri) fights: a case for
networking. Proc R Soc Lond Ser B 269:943–952
Freeberg TM (1998) The cultural transmisssion of courtship pat-
terns in cowbirds, Molothrus ater. Anim Behav 56:1063–1073.
Freeberg TM (1999) Spatial associations provide a context for
social learning of courtship patterns in brown-headed cow-
birds (Molothrus ater). J Comp Psychol 113:327–332
Freeburg TM, Duncan SD, Kast TL, Endstrom DA (1999) Cultur-
al influences on female mate choices: an experimental test in
cowbirds, Molotrus ater. Anim Behav 57:421-427
Gibson RM, Höglund J (1992) Copying and sexual selection.
Trends Ecol Evol 7:229–231
Gibson RM, Bradbury JW, Vehrencamp SL (1991) Mate choice
in lekking sage grouse revisited: The roles of vocal display,
female site fidelity, and copying. Behav Ecol 2:165–180
Grant JWA, Green LD (1996) Mate copying versus preference for
actively courting males by female Japanese medaka. Behav
Ecol 7:165–167
Hoglund J, Alatalo RV, Gibson RM, Lundberg A (1995) Mate-
choice copying in black grouse. Anim Behav 49:1627–1633
Jennions MD, Petrie M (1997) Variation in mate choice and mat-
ing preferences: a review of causes and consequences. Am Nat
72:283–327
King AP, West MJ (1983) Female perception of cowbird song: a
closed developmental program. Dev Psychobiol 16:335–342
King, AP, West, MJ (1988) Searching for the functional origins of
song in eastern brown-headed cowbirds, Molothrus ater ater.
Anim Behav 36:1575–1588
King AP, West MJ (1989) Presence of female cowbirds (Molo-
thrus ater ater) affects vocal improvisation in males. J Comp
Psychol 103:39–44
King AP, West MJ, White DJ (2002) The presumption of sociality:
Social learning in diverse contexts in brown-headed cowbirds
(Molothrus ater). J Comp Psychol 116:173–181
Losey GS, Stanton FG, Telecky TM, Tyler WA III, the Zoology
691 graduate seminar class (1986) Copying others, an evolu-
tionarily stable strategy for mate choice: a model. Am Nat
128:653–664
McComb K, Clutton-Brock T (1994) Is mate choice copying or
aggregation responsible for skewed distributions of females on
leks. Proc R Soc LondSer B 255:13–19
Mennill DJ, Ratcliffe LM, Boag PT (2002) Female eavesdropping
on male song contests in songbirds. Science 2996:873
Naguib M, Fichtel C, Todt D (1999) Nightingales respond more
strongly to vocal leaders of simulated dyadic interactions. Proc
R Soc Lond Ser B 266:537–542
Nordell SE, Valone TJ (1998) Mate choice copying as public
information. Ecol Lett 1:74–76
Nowicki S, Searcy BA, Peters S (2002). Quality of song learning
affects female response to male bird song. Proc R Soc Lond
Ser B 269:1949–1954
Oliveira RF, McGregor PK, Latruffe C (1998) Know thine enemy:
fighting fish gather information from observing conspecific
interactions. Proc R Soc Lond Ser B 265:1045–1049
Otter K, McGregor PK, Terry AMR, Burford FRL, Peake, TM,
Dabelsteen T (1999) Do female great tits (Parus major) assess
males by eavesdropping? A field study using interactive song
playback. Proc R Soc Lond Ser B 266:1305–1309
Peake TM, Terry AMR, McGregor PK, Dabelsteen T (2001) Male
great tits eavesdrop on simulated male-to-male vocal interac-
tions. Proc R Soc Lond Ser B 268:1183–1187
Pruett-Jones S (1992) Independent versus nonindependent mate
choice: do females copy each other? Am Nat 140:1000–
1009
Searcy WA, Nowicki S, Hughes M, Peters S (2002) Geographic
song discrimination in relation to dispersal distances in song
sparrows. Am Nat 159:221–230
Selander RK, Giller D (1960) First year plumages of the brown-
headed cowbird and red-winged blackbird. Condor 62:202–
214
Shuster SM, Wade MJ (1991) Female copying and sexual selec-
tion in a marine isopod crustacean Paracerceis sculpta: Anim
Behav 41:1071–1078
Sirot E (2001) Mate choice copying by females: the advantages of
a prudent strategy. J Evol Biol 14:418–423
Slagsvold T, Viljugrein H (1999) Mate choice copying versus
preference for actively displaying males by female pied fly-
catchers. Anim Behav 57:679–686
Smith VA, King AP, West MJ (2000) A role of her own: Female
cowbird influences on vocal development. Anim Behav 60:599–
609
Smith VA, King AP, West MJ (2002) The context of social learn-
ing in brown headed cowbirds: association patterns in a cap-
tive flock. Anim Behav 63:23–35
Stohr S (1998) Evolution of mate choice copying: a dynamic
model. Anim Behav 55:893–903
Templeton JJ, Giraldeau LA (1996) Vicarious sampling: the use of
personal and public information by starlings foraging in a
simply patchy environment. Behav Ecol Sociobiol 38:105–114
Valone TJ (1989) Group foraging, public information, and patch
estimation. Oikos 56:357–363
Valone TJ (1996) Food-associated calls as public information
about patch quality. Oikos 77:153–157
Wade MJ, Pruett-Jones SG (1990) Female copying increases the
variance in male mating success: Proc Natl Acad Sci USA
87:5749–5753
West MJ, King AP (1988a) Female visual displays affect the
development of male song in the cowbird. Nature 334:244–
246
West MJ, King AP (1988b) Vocalizations of juvenile cowbirds
(Molothrus ater ater) evoke copulatory responses from fe-
males. Dev Psychobiol 21:543–552
West MJ, King AP, Freeberg TM (1996) Social malleability in
cowbirds: new measures reveal new evidence of plasticity in
the Eastern subspecies (Molothrus ater ater). J Comp Psychol
110:15–26
172

Page 11

West MJ, King, AP, Freeberg, TM (1998) Dual signaling during
mating in brown-headed cowbirds (Molothrus ater, Family
Emberizidae/Icterinae). Ethology 104:250–267
West MJ, White DJ, King AP 2002 Female brown-headed cow-
birds’ (Molothrus ater) organization and behaviour reflects
male social dynamics. Anim Behav (in press)
Westneat DF, Walters A, McCarthy TM, Hatch MI, Hein WK
(2000) Alternative mechanisms of nonindependent mate
choice. Anim Behav 59:467–476
White DJ, Galef BJ, Jr. (1999) Mate choice copying and conspe-
cific cueing in Japanese quail, Coturnix coturnix japonica.
Anim Behav 57:465–473
White DJ, King AP, Duncan SD (2002a) Voice recognition tech-
nology as a tool for behavioral research. Behav Res Meth Inst
Comp 34:1–5
White DJ, King AP, West MJ (2002b) Facultative development of
courtship and communication skills in juvenile male cowbirds,
Molothrus ater. Behav Ecol 13:487–496
White DJ, King AP, Cole A, West MJ (2002c) Opening the social
gateway: early vocal and social sensitivities in brown-headed
cowbirds (Molothrus ater). Ethology 108:23–37
Widowski TM, Ziegler TE, Elowson AM, Snowdon CT (1990)
The role of males in the stimulation of reproductive function
in female cotton-top tamarins, Saguinus o. oedipus. Anim
Behav 40:731–741
Widowski TM, Porter TA, Ziegler TE, Snowdon CT (1992) The
stimulatory effect of males on the initiation but not the mainte-
nance of ovarian cycling in cotton-top tamarins (Saguinus
oedipus). Am J Primatol 26:97–108
173