Neuroendocrine contributions to sexual partner preference in birds.

Abstract

A majority of birds are socially monogamous, providing exceptional opportunities to discover neuroendocrine mechanisms underlying preferences for opposite-sex partners where the sexes form extended affiliative relationships. Zebra finches have been the focus of the most systematic program of research to date in any socially monogamous animal. In this species, sexual partner preference can be partially or largely sex reversed with hormone manipulations during early development, suggesting a role for organizational hormone actions. This same conclusion emerges from research with Japanese quail, which do not form long-term pairs. In zebra finches, social experience manipulations during juvenile development also can sex reverse partner preference, either alone or in combination with an early hormone manipulation. Although there are several candidate brain regions where neural mechanisms could underlie these effects of hormones or social experience, the necessary research has not yet been done to determine their involvement. The neuroendocrinology of avian sexual partner preference is still frontier territory.

Full text

Page 1

Review
Neuroendocrine contributions to sexual partner preference in birds
Elizabeth Adkins-Regan⇑
Department of Psychology and Department of Neurobiology and Behavior, Cornell University, Ithaca, NY 14853-7601, USA
a r t i c l e i n f o
Article history:
Available online 28 January 2011
Keywords:
Sexual partner preference
Zebra finch
Japanese quail
Estradiol
Aromatase inhibitor
Androgen receptor antagonist
Sexual imprinting
Social experience
Sexual differentiation
Homosexual behavior
a b s t r a c t
A majority of birds are socially monogamous, providing exceptional opportunities to discover neuroen-
docrine mechanisms underlying preferences for opposite-sex partners where the sexes form extended
affiliative relationships. Zebra finches have been the focus of the most systematic program of research
to date in any socially monogamous animal. In this species, sexual partner preference can be partially
or largely sex reversed with hormone manipulations during early development, suggesting a role for
organizational hormone actions. This same conclusion emerges from research with Japanese quail, which
do not form long-term pairs. In zebra finches, social experience manipulations during juvenile develop-
ment also can sex reverse partner preference, either alone or in combination with an early hormone
manipulation. Although there are several candidate brain regions where neural mechanisms could under-
lie these effects of hormones or social experience, the necessary research has not yet been done to deter-
mine their involvement. The neuroendocrinology of avian sexual partner preference is still frontier
territory.
? 2011 Elsevier Inc. All rights reserved.
1. Introduction
Sex differences in behavior have been a major focus of interest in
the study of the neuroendocrine mechanisms of social and repro-
ductive behavior. Research with a wide array of species, including
several birds, has contributed to an understanding of how such
sex differences develop, and how they are expressed in reproduc-
tively active adults [24]. Across species, one of the most widespread
and pronounced sex differences is sexual partner preference. Most
mating occurs between animals of different sexes, and in socially
monogamous species, most pairs consist of a female and a male.
In tests designed to measure preferences or choices for mating or
pairing, preferences for the other sex predominate. The sexes differ
dramatically in their preferred partner sex, with females preferring
males and males preferring females. Paradoxically, this strikingly
obvious and taxonomically broad sex difference is still not well
understood with respect to its neuroendocrine mechanisms, apart
from some progress with rats, mice, ferrets and sheep, none of
which form pair relationships [26,27,82,87]. This is a problem both
for a comprehensive view of the reproductive lives of animals and
for formulating biologically well-grounded hypotheses for human
sexual orientation. Neuroendocrine research with birds has been
limited to two species, the zebra finch (Taeniopygia guttata) and
the Japanese quail (Coturnix japonica). This body of work will be
the primary focus here.
Although heterosexual behavior and preferences for the other
sex predominate in the lives of birds and other animals, many
exceptions have been reported in both free-living and captive or
domesticated populations [16,18,69,93]. These exceptions have
the potential to illuminate the processes responsible for the pre-
dominant heterosexual preferences, as shown in research with
sheep [82,87].
Sexual partner preference is one of three types of mating prefer-
ences that are important for reproductive success. The other two,
preference for conspecifics rather than heterospecifics, and prefer-
ences for particular traits in opposite-sex conspecifics such as orna-
ments, displays,or songqualities,
investigated in birds. Both develop in part through learning (for
example,sexualimprinting)orexperience(forexample,priorexpo-
sure to high quality songs) [39,60,92]. This raises the possibility
that similar processes might also contribute to sexual partner pref-
erence, together with any neuroendocrine mechanisms. This is
especially likely if the same cues that are used to recognize species
and mate quality also identify sex. For example, the song of a male
zebra finch could identify to a female that he is her species, a male,
and a good quality male. The importance of attractive cues or sig-
nals in mate choice means that the relevant neuroendocrine mech-
anisms include not only those responsible for the preferences of the
choosers, but also those underlying the traits of the chosen that at-
tract the other sex.
This review will refer to ‘‘sexual partner preference’’ or ‘‘pairing
partner preference’’ rather than ‘‘sexual orientation.’’ ‘‘Orientation’’
tends to imply a long-lasting, even permanent, stable, exclusive
preference for one sex even when other possibilities are available.
havebeenintensively
0091-3022/$ - see front matter ? 2011 Elsevier Inc. All rights reserved.
doi:10.1016/j.yfrne.2011.01.003
⇑Address: 218 Uris Hall, Cornell University, Ithaca, NY 14853-7601, USA. Fax: +1
607 255 8433.
E-mail address: er12@cornell.edu
Frontiers in Neuroendocrinology 32 (2011) 155–163
Contents lists available at ScienceDirect
Frontiers in Neuroendocrinology
journal homepage: www.elsevier.com/locate/yfrne

Page 2

Most animal research has not been designed to determine how
long-lasting, stable or exclusive the preferences are. Concluding
that some free-living birds are exclusively homosexual for life
would require marking individuals, knowing what sex they are
(not always easy, because some birds are sexually indistinguish-
able to humans), following them throughout most of their lives,
and knowing that individuals of the other sex were available and
willing to be chosen. In addition, in humans the phrase ‘‘sexual ori-
entation’’ often implies a particular internal psychological identity,
which does not translate very well to the non-human world [18].
Why study birds to understand sexual partner preference? The
two species to be discussed here offer some practical and scientific
advantages, such as ease of husbandry and a sequenced genome
(zebra finch) or a taxonomic relative with a sequenced genome
(Japanese quail, in the same family as the chicken) [67,75,107]. Fur-
thermore, birds have great potential for a more integrative view of
sexual partner preference, one that combines insights into mecha-
nisms with those of evolution and ecology. Among free-living ani-
mals, birds are the best-studied group of vertebrates, and quite a
bit is known about their lives and mating systems in the wild.
These mating systems are an essential context for conceptualiz-
ing sexual partner preference and studying its neuroendocrine and
other influences. This point can be appreciated by considering
three very different mating systems from among the many that
have been discovered in birds. In one, a lek mating system, seen,
for example, in North American sage grouse (Centrocercus urophasi-
anus) and neotropical manakins (e.g., Chiroxiphia lanceolata), males
display in a group arena [45,46]. A female that is ready to mate ob-
serves the males displaying, and then approaches one and copu-
lates. She then leaves the lek to lay and incubate her eggs. Here
sexual partner preference is limited to attraction to male cues
and copulation; there is no further relationship between the male
and the female. It would not be surprising if the neuroendocrine
mechanisms for sexual partner preference turned out to overlap
to some extent with those for copulation itself. Sexual partner pref-
erence would result from a connection between male vs. female
cues (or high quality vs. lower quality male cues) and sexual moti-
vation and performance systems. In the second mating system,
seen, for example, in large bodied birds such as geese, swans,
and albatrosses, males and females form an exclusive pair relation-
ship that lasts for many years or even for the long life of the birds
[30,53]. The birds raise young together (are biparental), seldom
copulate with other individuals (i.e., are largely genetically as well
as socially monogamous), and are in close physical proximity all
year, even when the gonads are regressed and no reproduction is
occurring. There is an affiliative social relationship as well as a
mating relationship, the social relationship consists of much more
than just copulation, and pairs can form outside of the breeding
season. Here there is greater potential for mechanisms for pairing
partner preference to differ from those for sexual motivation and
copulation. Some might be more closely linked to parental behav-
ior or parent-offspring relationships [38]. In a third mating system,
seen in superb fairy-wrens (Malurus cyaneus), females and males
form socially monogamous biparental pairs, but over half (on aver-
age) of the chicks in the nests have resulted from extra-pair copu-
lations between the female and males other than her social partner
[44]. Here there is a striking difference between a female’s pairing
and co-parenting partner preference and her copulation prefer-
ence. Both are preferences for males, but they focus on different
traits of males for different purposes (social vs. genetic paternity).
Where do zebra finches and Japanese quail fit into this picture?
Zebra finches exhibit a small-bodied short-lived version of the sec-
ond mating system described above (monogamous and biparental)
[111]. They are gregarious and breed colonially. They form socially
monogamous pairs at a young age (at the onset of reproductive
maturity) that last until one member dies. The pair relationship
is marked by a distinctive behavioral repertoire involving close
physical contact: clumping (perching with bodies in contact),
allopreening (preening each other), and spending time in a nest
box together. Birds are highly motivated to pair, and unpaired
adult birds have not been reported [111]. Same-sex pairs have also
not been reported in the wild, but do occur sporadically in captive
colonies. Copulation occurs infrequently during systematic behav-
ioral observations. Extra-pair fertilization rates in wild populations
are low, but are slightly higher in captive colonies [28,33,54]. Pair-
ing preferences are more salient and easier to measure than copu-
lation preferences. Most of the little that is known about the
neuroendocrinology of opposite-sex pairing preferences in birds
comes from work with this species. In fact, this is the only pair-
forming species of any kind that has been studied systematically
for the purpose of determining neuroendocrine mechanisms of
pairing preference.
The mating system of wild Japanese quail is poorly understood,
and thus difficult to place into any of the categories described
above, but is clearly very different from that of the zebra finch.
Observations of genetically near-wild birds in outdoor enclosures
suggest short-term pair relationships with occasional mate switch-
ing and extra-pair copulations [80]. Field observations of the clo-
sely related European quail (Coturnix coturnix) paint a similar
picture [86]. Parental care (egg incubation and brooding of chicks)
is mainly or entirely by females alone in both species [70].
2. Experience, learning and sexual partner preference in
Japanese quail and zebra finches
As mentioned above, we can expect some influences in addition
to neuroendocrinology on the development and expression of sex-
ual partner preference in birds, such as experience and learning,
including sexual imprinting. Inexperienced male Japanese quail
are initially fairly indiscriminate with respect to copulation partner
preference, approaching and attempting to mate with either sex.
With more mating experience, they come to approach females
more than males in a choice apparatus (Y-maze) [78]. A systematic
program of well-designed research has established that this
change in behavior over mating trials is due to Pavlovian (classical)
conditioning [42]. Female cues are the conditioned stimuli and suc-
cessful mating (which is more likely to occur with females) is the
reinforcement. Simply having visual access to females without get-
ting to copulate improves sexual performance in subsequent mat-
ing trials, an effect that does not occur to the same degree if the
visual access is to males instead of females [37]. This effect of expe-
rience occurs even if the males are castrated and receiving testos-
terone replacement; therefore, changes in circulating testosterone
are not the mechanism.
A role for prior social experience in zebra finch sexual partner
preference was initially shown in a study in which all adult males
were removed from breeding aviaries when chicks were still in the
nests [7]. When the offspring were adults, they paired with same-
sex birds at a higher rate than controls did, and paired with oppo-
site-sex birds at a lower rate (Fig. 1). It is not known whether this
effect was due to the absence of fathers, the absence of adult males
as models, or not seeing male–female pairs. In a more recent study,
it was the mothers that were removed, and experimental groups
were in aviaries divided in half with control groups in the other
half, so that mother-deprived subjects could see and hear adult fe-
males and opposite-sex pairs in the other half [25]. As adults the
female offspring paired with males similarly to control females,
but the male offspring nearly all paired with other males. Zebra
finches are known to sexually imprint on traits of the opposite-
sex parent, and so the experimental males may have imprinted
on their fathers in the absence of the mothers [31,103,109].
156
E. Adkins-Regan/Frontiers in Neuroendocrinology 32 (2011) 155–163

Page 3

Social context is another potentially important influence on
sexual or pairing partner preference. In several species of birds,
outcomes seem to vary according to the sex ratio in the population
at the time of pairing or mating. Female-biased sex ratios are now
thought to be responsible for the female–female pairs that have
been observed in some populations of gulls, albatrosses, and geese
[85]. Similarly, in species with high male reproductive skew
(monopolization of reproduction by dominant males), lack of
opportunities to mate with females has been hypothesized to pre-
dict greater male–male sexual activity [69]. Strictly speaking, these
are not necessarily altered preferences, but instead could be oppor-
tunistic responses to a lack of availability of the preferred option.
Furthermore, the individuals in female–female pairs in wild bird
populations are thought to be lower quality birds that are making
the best of a bad situation by obtaining fertilizations through ex-
tra-pair copulations and co-parenting with another female, a better
option than remaining unpaired and not reproducing at all
[64,81,110]. Nonetheless, these observations indicate a previously
unsuspected degree of plasticity in mate selection, suggesting that
whatever the neuroendocrine mechanisms of sexual partner pref-
erence, they are also plastic, or at least not always strongly
determinative.
3. Activational neuroendocrine mechanisms
A diverse array of hormones and neuropeptides have been
shown to be involved in stimulating the expression of courtship
and mating behavior in adult birds, including gonadal sex steroids,
steroids produced locally within the brain, and nonapeptides such
as vasotocin [24,49]. The expression of sexual interest and motiva-
tion is similarly dependent on these mechanisms. Thus far, how-
ever, none of these has been shown to account for the fact that
these behaviors and interest are preferentially directed at oppo-
site-sex individuals. Female Japanese quail given testosterone crow
and strut a bit (male-typical mate attraction and courtship behav-
ior), and exhibit rhythmic movements of the foam gland portion of
the cloacal sphincter muscle to the sight of another bird, just as
males do, but are not more interested in looking at other females
than control females are [8,23]. Adult castrated male Japanese
quail are not interested in looking at females, but are not sexually
interested in males either [23]. Adult male zebra finches that were
castrated and given estradiol benzoate (EB) implants still paired
with females, and adult females that were ovariectomized and gi-
ven testosterone propionate (TP) implants still paired with males
[5]. Adult zebra finches that were given two drugs to reduce both
estrogen and androgen action (ATD, to lower estrogen synthesis,
and flutamide, an androgen receptor antagonist) still paired nor-
mally and with opposite-sex partners [99]. The drugs were effec-
tive, in that the females did not lay eggs and the males had
regressed vasa deferentia, but pairing preference and motivation
were unaffected. Thus in zebra finches, unlike quail, even the
expression of sexual partner preference does not seem to require
reproductively active levels of sex steroid action. Juvenile zebra
finches that were given implants of EB, TP, or ATD plus flutamide
showed accelerated (TP in males) or delayed (ATD plus flutamide
in females) timing of the developmental shift in preference from
family members to opposite-sex unpaired birds, but still preferred
opposite-sex to same-sex unpaired birds (Fig. 2) [9]. That is, alter-
ing the sex steroid environment beginning early in juvenile devel-
opment (around 25 days), when the birds were still dependent on
the parents, and continuing until sexual maturity, did not alter
their sexual partner preference when it began to be expressed.
Fig. 1. Zebra finches raised by mothers alone, by removing the fathers at a young
age, do not show the normal strong preference for pairing with the other sex seen in
control birds raised by both parents. M = males raised by both parents (M-control)
or by removing fathers (M-removal); F = females raised by both parents (F-control)
or by removing fathers (F-removal). From [7].
Fig. 2. (a) Juvenile male zebra finches with testosterone implants develop a
premature onset of interest in females compared to controls in tests where they can
choose between unpaired females, unpaired males, or their family members. (b)
Juvenile female zebra finches with implants of ATD (an aromatase inhibitor) and
flutamide (an androgen receptor antagonist) fail to develop any preference for
males in choice tests.⁄p < 0.05 compared to controls in the same test week. From
[9].
E. Adkins-Regan/Frontiers in Neuroendocrinology 32 (2011) 155–163
157

Page 4

All told, there is no evidence that the heterosexual direction of
the sexual partner preference of birds results from adult sex ste-
roid levels (activational hormone actions), either in the circulation
or in the brain. In quail, as in mammals, the expression of the bird’s
preference requires activational hormones, but the fact that the
preference is for the other sex instead of the same sex is not deter-
mined by activational hormones. If sex steroids shape the direction
of sexual partner preference, they must be doing that earlier in
development.
4. Organizational hormone effects
Among birds, the role of sex steroid hormones produced early in
development in sexual differentiation of behavior is best under-
stood in Japanese quail [24]. The copulatory behavior pattern
shown by adult males (head-grabbing, mounting, and cloacal con-
tact movements) requires testosterone or products of testosterone
aromatization for expression (activation). The fact that males but
not females show this behavior is not a product of hormonal acti-
vation, however, because adult females given testosterone still do
not head-grab or mount. Instead, females will show these behav-
iors if they hatched from eggs treated with estrogen synthesis
inhibitors or estrogen receptor antagonists and are then given tes-
tosterone as adults. Furthermore, males hatched from eggs treated
with estradiol or aromatizable androgens will be demasculinized,
and as adults they will not head-grab and mount.
Male but not female quail will learn to spend time looking at fe-
males through a high window [40]. If interest in females (looking
time) is measured instead of mating behavior, the same result,
embryonic demasculinization by estrogens, is obtained as for cop-
ulatory behavior [23]. Females hatched from eggs injected with the
aromatase inhibitor R76713 and given testosterone as adults (to
activate any male-typical tendencies) spend time looking at other
females through the window, whereas control females given tes-
tosterone do not. In these tests, the female stimulus is visual and
conceivably olfactory as well [20]. It will be interesting to find
out what the steroids are doing organizationally, and where, that
makes the same stimulus attractive to males but not to females.
Strictly speaking, these tests are not measuring sexual partner
preference, because the bird is not choosing between female and
male stimuli. Nonetheless, male-specific interest in female stimuli
seems like a closely related phenomenon or even essential compo-
nent of sexual partner preference.
In zebra finches, sexual differentiation of behavior is less well
understoodgenerally.Althoughthere is markedsexualdimorphism
in singing (only males sing) and in the telencephalic song system
(nuclei are much larger in males), it is still not clear how those
sex differences develop, or what the role of sex steroids is
[24,104]. Onthe onehand,treatmentoffemalenestlingswithestra-
diol masculinizes singing (normal females never sing) and the neu-
ral song system, and there is good evidence from an in vitro study
that the development of the projection from song system nucleus
HVC (located in the nidopallium) to RA (robust nucleus of the arco-
pallium) requires estradiol [58]. On the other hand, estradiol treat-
ment produces only partial, not complete, masculinization of the
songsystem[104].Evenmoreproblematicis thattreatmentofmale
nestlings with estrogen synthesis inhibitors does not prevent nor-
mal masculine singing and song system development, and birds
with sex-reversed gonads have song systems reflecting their genet-
ic, not gonadal, sex [104]. Regardless of whether estrogens are play-
ing a major or only minor role, however, it is clear that androgens,
including testosterone, are not key mechanisms, because early
treatment with androgens or an androgen receptor antagonist has
little or no effect on singing or other sexually differentiated behav-
ior [6,55,56]. The fact that a bird’s genetic sex can be a better match
than its gonadal sex with its song system suggests the possibility of
what have been termed ‘‘direct genetic effects’’ on sexual differen-
tiation, effects of genes through non-gonadal products that are ex-
pressed locally in the brain or its song system [15]. The initial
gene(s) of such cascades would presumably be on the sex chromo-
somes (Z or W, male birds are ZZ and females are ZW) or regulated
by one or more sex chromosome genes. Research thus far has
hypothesized sex chromosomal genes such as CHD1 (Chromodo-
main-Helicase-DNA bindingprotein), trkB(tyrosine kinase receptor
B, a receptor for two neurotrophic factors), NGF (nerve growth fac-
tor), orRPL17and RPL37(two
[12,35,62,95]. Furthermore, given that estrogen does appear to play
at least a minor role, other research has targeted genes for estrogen
related mechanisms such as 17b-hydroxysteroid dehydrogenase
type 4, aromatase, the two estrogen receptors, and estrogen recep-
tor coactivators [43,68,83,100].
The same early estradiol treatment that masculinizes singing,
daily injection of EB for the first two post-hatching weeks, also pro-
duces females with masculinized sexual partner preference. Trea-
ted females prefer other females both in two-choice proximity
tests (choice of female vs. male stimuli) and in mixed-sex aviaries
where birds can pair freely [5,73]. These results suggest hormonal
organization of sexual partner preference in this species. It makes
no difference whether the females are given testosterone as adults,
again indicating little or no activational steroid influence [3].
This estrogen-induced masculinized sexual partner preference
only occurs if the females are also housed in all-female aviaries dur-
ing juvenile development, prior to preference testing (Fig. 3) [73].
Neither the estradiol treatment alone, nor all-female housing alone,
produces a significant shift in sexual partner preference. A combi-
nation of altered early steroid milieu and a particular kind of social
environment or experience is required, suggesting that hormones
and social experience might work together during normal develop-
ment to produce the typical adult opposite-sex preferences.
What is it about the all-female environment that is important
for estradiol to be effective? One possibility is that estradiol mas-
culinizes how the females learn or imprint. Sexual imprinting in
this species occurs in two stages, an early stage, while the birds
are still dependent on the parents, in which the birds learn charac-
teristics of the opposite-sex parent, and a second stage, as they be-
gin to mature reproductively, in which preferences for those
characteristicsare consolidated
ribosomalprotein genes)
by interactionswithpeers
Fig. 3. Female zebra finches injected with estradiol benzoate (EB) as nestlings
attempt to pair with other females, unlike controls (Cont), but only if they lived in
all-female (unisex, Uni) aviaries as juveniles and young adults instead of mixed-sex
(Mix) aviaries.⁄p < 0.05 compared to Cont–Mix group. From [73].
158
E. Adkins-Regan/Frontiers in Neuroendocrinology 32 (2011) 155–163

Page 5

[29,61,65]. If estradiol caused the females to imprint in a male-typ-
ical manner (to imprint on female traits), then those preferences
may have been consolidated by being housed with no males pres-
ent, forcing all interactions to be with females. In an effort to
understand what kind of interaction with other females was
responsible, females were treated with EB as nestlings and then
housed with other females in aviaries that either did or did not al-
low clumping, allopreening, and being in a nest box together [4].
Only those females that could have this kind of contact with other
females later showed a shift away from preferences for males. This
is consistent with the hypothesis that physically close interactions
with other females may have consolidated tendencies learned ear-
lier through sexual imprinting that were masculinized by estrogen.
These experiments with nestling estradiol treatment were the
first evidence for an organizational hormone effect on sexual part-
ner preference in any socially monogamous or pair-forming spe-
cies. Their interpretation is not entirely straightforward, however.
The early estradiol treatment is slightly toxic, a common side effect
of systemic estrogen treatment in birds (see [71] for a report of
similar toxicity in fetal mice). Treated chicks suffer greater mortal-
ity than vehicle injected controls. Furthermore, reducing estrogen
production in nestling males with aromatase inhibitors does not
feminize their preferences, the same paradoxical outcome seen in
studies of sexual differentiation of the song system in this species
[11]. This is a serious impediment to concluding that estrogens are
the normal developmental ‘‘signal’’ acting in males to produce
their later preference for females.
A subsequent experiment with zebra finches used a much more
benign treatment: a single injection of 20 lg fadrozole into eggs on
post-oviposition day 5 of embryonic development [10]. Hatching
rates and post-hatching survival were the same for treated and
control (vehicle injected) eggs. Males hatched from these eggs
were no different behaviorally from controls (Fig. 4). Females from
fadrozole injected eggs had testes or ovotestes as adults, consistent
with what had been reported previously [105]. Furthermore, these
females mainly paired with other females in free-choice aviaries,
unlike control females (Fig. 4). This occurred even though the fe-
males spent their juvenile lives in mixed-sex aviaries; i.e., the sex
reversal effect did not require all-female social experience, unlike
the sex reversal produced by early estradiol. Did fadrozole reverse
sexual partner preference because it produced testes, which then
created a different circulating hormonal milieu during early devel-
opment? Or did it occur because of a direct effect of the fadrozole
on the embryonic brain, depriving it of estrogen? Further research
would be necessary to decide between these alternatives. Regard-
less of which pathway was responsible, these results are a strong
hint that some sex steroid mechanism contributes to the develop-
ment of sexual partner preference in zebra finches.
It is also notable that the sex reversal of preference following
egg injection with fadrozole was independent of any effect on sing-
ing, in contrast to nestling treatment with estradiol, which mascul-
inizes both singing and sexual partner preference. Whatever the
mechanisms are for sexual differentiation of singing and sexual
partner preference, they are dissociable. Whatever the genes are
for direct genetic effects on song system development, they must
not overlap entirely with those for sexual partner preference. Some
could be hypothesized to be epigenetically programmed by juve-
nile social experience. Sexual partner preference is also dissociable
from copulatory behavior, because early estradiol masculinizes
partner preference in females without masculinizing their mount-
ing behavior [5]. This contrasts with the situation in Japanese quail,
where interest in females and copulatory mounting are altered in
parallel by embryonic hormone treatment, as if they are based
on similar developmental mechanisms. This difference between
the two species could be because successful copulation with fe-
males but not with males is how male quail learn to prefer females
[40].
5. Potential neural substrates
At this time, very little is known about the brain regions that
might be involved in sexual partner preference in birds, where
the effects of early hormones and social experience might be
occurring. Based on research on other aspects of sexual behavior,
some candidates can be proposed for consideration. One source
of candidates, especially for species that are not socially monoga-
mous, is brain regions involved in copulation. Although sexual
partner preference (choice of partners for copulation) is different
from copulation (the behavior engaged in with the chosen partner),
the motivation for the choice is at least partly sexual. In the Japa-
nese quail, several studies have shown that the neuroanatomical
regions involved in appetitive components of male sexual behavior
(courtship, approach to the female) differ somewhat from those in-
volved in the consummatory components (actual copulation itself)
[19]. Thus within the preoptic area, lesions of the rostral preoptic
affect appetitive more than consummatory behavior, whereas le-
sions of the caudal preoptic affect consummatory more than appe-
titive behavior. A similar dissociation is seen when neural activity
during appetitive vs. consummatory behavior is measured using
immediate early gene expression as the marker of neural activity.
Sexual partner preference is more an appetitive than a consumma-
tory behavior, and so any involvement of the preoptic area would
more likely lie in the rostral portion. In ferrets, lesions of the pre-
optic/anterior hypothalamic region caused males to choose males
instead of estrous females in T-tests [63]. Whether this effect
would occur in male birds, which rely less on female odors for sex-
ual partner preference, is not known.
Another possibility from the sexual behavior literature is the
mesolimbic dopaminergic circuit that originates in the ventral teg-
mental area (VTA), projects to the telencephalic nucleus accum-
bens, and is activated during both appetitive and consummatory
sexual behavior in Japanese quail [34,98]. Two studies have looked
at activation of VTA dopaminergic neurons in male zebra finches in
response to female stimuli. Both used immunocytochemistry, dou-
ble labeling for tyrosine hydroxylase (TH) and c-fos, an immediate
early gene product, to identify recently active dopaminergic neu-
rons. In one study, males that courted females had more TH-immu-
Fig. 4. Female (F) zebra finches hatched from eggs injected with fadrozole on day 5
of incubation show sex-reversed partner preference as adults, preferring to pair
with other females, whereas males (M) are unaffected in their preference.⁄p < 0.05
compared with F – Control. From [10].
E. Adkins-Regan/Frontiers in Neuroendocrinology 32 (2011) 155–163
159