The Biology of Homosexuality

The Biology of Homosexuality
Oxford Series in Behavioral Neuroendocrinology
Editors
Gregory F. Ball, Johns Hopkins University
Jacques Balthazart, University of Liège
Randy J. Nelson, Ohio State University
Luis Miguel Garcia-Segura, Hormones and Brain Plasticity

Jacques Balthazart, The Biology of Homosexuality
J ACQUES BALTHAZART, P HD
GIGA Neurosciences
University of Liège, Belgium
1
1
Published in the United States of America by Oxford University Press, Inc.,
198 Madison Avenue, New York, NY, 10016
United States of America
Oxford University Press, Inc. publishes works that further Oxford University’s objective
of excellence in research, scholarship, and education
Oxford is a registered trade mark of Oxford University Press in the UK and in certain other countries
Copyright © Jacques Balthazart 2012
All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or
transmitted, in any form or by any means, without the prior permission in writing of Oxford University Press,
Inc., or as expressly permitted by law, by licence, or under terms agreed with the appropriate reproduction
rights organization. Inquiries concerning reproduction outside the scope of the above should be sent to the
Rights Department, Oxford University Press, Inc., at the address above
You must not circulate this work in any other form and you must impose this same condition on any acquirer
Library of Congress Cataloging-in-Publication Data
____________________________________________________
Library of Congress Cataloging-in-Publication Data
Balthazart, J. (Jacques), 1949-

The biology of homosexuality / Jacques Balthazart.
p. cm.
Includes bibliographical references and index.
ISBN 978-0-19-983882-0 (hardback)
1. Homosexuality. 2. Sexual orientation. I. Title.
HQ76.25.B35 2011
306.76'6—dc23 2011036460
____________________________________________________
1 3 5 7 9 10 8 6 4 2
Typeset in Minion Pro

Printed on acid-free paper
Printed in the United States of America
This material is not intended to be, and should not be considered, a substitute for medical or other
professional advice. Treatment for the conditions described in this material is highly dependent on the
individual circumstances. And, while this material is designed to offer accurate information with respect to
the subject matter covered and to be current as of the time it was written, research and knowledge about
medical and health issues is constantly evolving and dose schedules for medications are being revised
continually, with new side effects recognized and accounted for regularly. Readers must therefore always
check the product information and clinical procedures with the most up-to-date published product
information and data sheets provided by the manufacturers and the most recent codes of conduct and safety
regulation. The publisher and the authors make no representations or warranties to readers, express or
implied, as to the accuracy or completeness of this material. Without limiting the foregoing, the publisher and
the authors make no representations or warranties as to the accuracy or efficacy of the drug dosages
mentioned in the material. The authors and the publisher do not accept, and expressly disclaim, any
responsibility for any liability, loss or risk that may be claimed or incurred as a consequence of the use and/or
application of any of the contents of this material.
ACKNOWLEDGMENTS
I would like to thank my wife, Claire, and my daughter-in-law, Nathalie, not

only for their constant support but also for the constructive criticism that they
provided at multiple stages during the writing of this book. This book would
not have been written without their help.
At Oxford University Press, Catharine Carlin initiated this project and Joan
Bossert brought it to completion. They both played a key role in publication.
Steve Holtje greatly improved my original translation of the text from French to
English, and I would like to thank him for his essential contribution.
Finally several members in my laboratory, in particular, Dr. Charlotte A.
Cornil, provided useful critiques and suggestions at various stages of writing.
Thank you very much.
This page intentionally left blank
CONTENTS
Introduction: Why This Book? ix
1. Sexuality 101: The Basics 3
2. Sexuality and Sexual Orientation: Learning or Biology? 12
3. The Hormonal Control of Sexual Behavior 27
4. Biological Determinism of Sexual Orientation in Animals 47
5. Gender Differences in Humans 61
6. The Effects of Sex Steroids in Humans: Organizing Effects 69
7. The Effects of Sex Steroids in Humans: Activating Effects 91
8. Sex Differences Suggest Homosexuality Is an

Endocrine Phenomenon 101
9. Sexual Orientation in Clinical Cases 130
10. A Genetic or Immunological Mechanism

Underlying Homosexuality? 141
11. General Conclusions 154
References 163
Index 179
INTRODUCTION: WHY THIS BOOK?
The attitude of Western societies toward homosexuals has changed consider-

ably over the centuries. The very tolerant attitude that prevailed for much of
Greco-Roman antiquity and the Middle Ages was followed by a time of greater
intolerance largely inspired by changes in Judeo-Christian thinking. Homo-
sexuality was then seen as a stigma or a perversion that should be fought and
treated like a disease. Later, various schools of thought inspired by Freudian or
post-Freudian theories attributed a critical role in the development of homo-
sexuality to the attitude of parents toward their young children.
Based on scientific results accumulated over the past few decades, I believe, in
contrast, that the origin of homosexuality must be sought in the biology of indi-
viduals who express this particular sexual orientation rather than in the behavior
of their parents or in the life choices made by individuals during their develop-
ment. It is the purpose of this book to summarize the arguments that support
this conclusion. Neither the education that parents provided nor their own
behavior (e.g., absent father or dominant mother) induced the homosexuality of
their child. In this conceptual framework, to blame parents is irrelevant. (Please
understand that the use of the word “blame” here and elsewhere in the book is
not a statement that homosexuality is blameworthy, but an observation and
comment on behavior in societies where stigma is attached to homosexuality.)
It is crucial to understand that for the vast majority of subjects, homosexual
behavior develops independently of any kind of choice. It is even frequently in
conflict with the choices that are consciously made by the individual. The dis-
covery of their homosexual orientation is, for many adolescents, painful and
only accepted gradually in their life, if ever. This undoubtedly helps to explain
the so-called cases of change in sexual orientation–homosexuals who con-
tracted a heterosexual marriage and then apparently changed their orientation.
The first orientation was often the result of social pressures that the individual
took decades to overcome before accepting his or her true nature. This difficulty
in accepting an unexpected sexual orientation is also at the root of the increased
suicide rate in the homosexual population as compared with matched hetero-
sexual populations (Kourany, 1987; Lebson, 2002).
This reflection on the social consequences of one or the other interpretation
of homosexuality does not affect our analysis of its causes. The conclusions
based on evidence must of course take precedence. That said, in light of this
x Introduction: Why This Book?
evidence, I hope this book will help make society aware that sexual orientation
is probably under the control of embryonic endocrine/genetic phenomena in
which there is little room for individual choice. Therefore society needs to
rethink its attitudes toward homosexuals and their parents.
Animal studies are of great support in understanding the controlling forces
of sexuality. These studies demonstrate the existence in animals of neuro-
biological mechanisms determining the behavior and sexual orientation that
seem also to be present, mutatis mutandis, in human studies. This experimental
work, carried out in strictly controlled conditions, provides a strong support
to interpret correlative studies conducted in humans, even if the details of the
mechanisms involved are currently not fully understood. They probably never
can be, given that it is impossible to perform most types of experiments in
humans (for obvious ethical reasons) and that there is a very long latency
(greater than 20 years) between the events that probably induce homo- or hetero-
sexuality (endocrine environment of the embryo and its genetic heritage) and
the identification of their effects (sexual orientation of the adult). Prospective
studies are thus extremely difficult and expensive if you consider, in addition,
the limited number of homosexuals who appear in the population studied.
Retrospective studies are possible, but less convincing because the selection of
subjects and their controls, even if they are well matched, can be skewed by
unidentified factors.
For more than 35 years, I have been studying the hormonal and neural
mechanisms that control the so-called instinctive behaviors (spontaneously
expressed and species-specific) in animals. Much of this work has been devoted
to the study of male sexual behavior, and led me to turn my attention to sex dif-
ferences that are associated with this behavior. In most species, males and
females indeed show different behaviors, and furthermore they mostly direct
these behaviors toward the opposite sex (i.e., they show a marked sex difference
in sexual orientation). These differences in behavior and its orientation are
clearly controlled in animals by steroid hormones produced by the gonads.
The action of these steroids on behavior takes place mostly in the brain, par-
ticularly in some of its oldest parts from an evolutionary point of view, the
hypothalamus and preoptic area. The primary goal of research in neuroendo-
crinology is to advance, at the basic level, our knowledge of how the brain func-
tions. Clearly, however, the potential application of this knowledge to the
human species is an issue that cannot be escaped. Is human sexual behavior
under the control of testicular and ovarian hormones? And do these hormones
affect the orientation of this behavior?
It is interesting that research on the neuroendocrine basis of sexual behavior
in animals and humans is a discipline that has developed more in the Anglo-
Saxon countries, mainly the United States and to a lesser extent Great Britain.
Introduction: Why This Book? xi
The bulk of the scientific literature in this area is therefore published in English.
For the same reason, there are also a few English-language books devoted to the
biological mechanisms that control animal or human sexuality in general,
including sometimes a discussion of sexual orientation (LeVay, 1993; 1996;
Strong & DeVault, 1997; Rosenzweig et al., 2004; LeVay & Valente, 2006; Agmo,
2007).
However, there is not, to my knowledge, a simple description of the biologi-
cal mechanisms that are involved in the determination of sexual orientation in
animals and also presumably in humans. By showing herein that the sexual
orientation of a subject is controlled by a set of biological and environmental
factors, I hope to provide that book. This text was originally written and pub-
lished in French in early February 2010. While its English version was being
prepared, another book written by Simon LeVay was published that also focuses
on the biological bases of human homosexuality (Gay, Straight, and the Reason
Why, Oxford University Press, 2010). Although much of the experimental evi-
dence presented in LeVay’s book is also discussed here, these two books differ
in their overall approach. LeVay starts from the notion of sexual orientation in
humans and discusses this orientation based on human studies largely, using
animal studies mainly to support conclusions derived from observations in
humans. In contrast, the present text starts from the animal work that is the
focus of my scientific research and in a second step analyzes how this animal
work potentially applies to humans. It’s quite interesting that while these two
books start from different premises, they reach the exact same conclusion, that
is, that homosexuality in humans is to a very large extent, if not exclusively,
determined by biological factors acting prenatally or soon after birth and that
the social or educational environment plays at best a subsidiary role in this
determinism. The conjunction of these two books thus strongly reinforces the
general conclusion presented in the following pages.
It must also be noted that while describing the experimental support of this
thesis, the book will focus on more than just homosexuality, as the context in
which biological factors act produces multiple results. Ignoring the nonhomo-
sexual results of the process would present an incomplete, context-free picture
of the factors at work. Thus, even readers not particularly interested in homo-
sexuality will find this book a useful guide to the biological bases of sexual ori-
entation. Understanding the roots of homosexuality requires also investigating
the bases of heterosexuality, since both forms of sexual orientation only repre-
sent the two extremes of a continuum called sexual orientation. This question
is then of interest to everyone. I will also note in advance that because male
homosexuality is more clear-cut than female homosexuality and its biological
origins are better documented, at times it will dominate the discussion. This is
merely an unfortunate result of where our current scientific understanding lies
xii Introduction: Why This Book?
and nothing more. This book presents as balanced a portrait as possible, and no
slight or judgment is intended.
Scientists reading this book will think, “We have heard all this before.”
Certainly the content of this book is not scientifically revolutionary; it has been
known for at least a decade, though over that time more evidence has continued
to pile up. But somehow that information has either not made its way into the
world outside the laboratory or has not been presented in a sufficiently defini-
tive manner to affect the general population’s views on the matter (with the
exception of the book of Simon LeVay mentioned in the preceding paragraph
[LeVay, 2010]). Scientists are thus not the only audience for which this book is
intended. This is a book also for laymen, but it contains enough science that
rather than just being told “X is true,” they are shown why it’s true in a way that
I hope will give them a genuine understanding of the subject, not just an opin-
ion. In this respect, this book can thus also be used by scientists and teachers for
their seminars or advanced courses on sexuality. However, if they wish, readers
can skip over the more hardcore science in Chapters 3–4 and still emerge with
plenty of useful knowledge. There’s a glossary at the end that will help those
readers navigate the later chapters.
Of course, presenting a mass of facts without a guiding organization would
result in a mere collection of lab reports, not a book. Some will say this book
advocates a specific position rather than letting readers draw their own conclu-
sions. I arrived at this position because the scientific facts led there. They also
led Simon LeVay to this same conclusion, and that’s why this book leads readers
to the same place.
1
Sexuality 101
The Basics
THE FOUR DIMENSIONS OF HUMAN SEXUALITY

Human sexuality is a complex and multidimensional phenomenon that covers
various aspects of behavior and personality. Four dimensions that are partly but
not completely independent can be distinguished : (1) the type of specific action
patterns that are produced by the individual (performance) and the motivation
underlying the expression of these behaviors; (2) the orientation of the behav-
ior and of sexual fantasies associated with it (sexual orientation); (3) the sexual
identity that the individual believes he or she has; and (4) the sexual role that
the individual plays in the society. These four dimensions are often correlated
within an individual and linked to his or her physical and genetic sex, but major
discrepancies may occur. In order to clarify important points of terminology,
sexologists speak of homosexuality when the sexual orientation of an individ-
ual does not correspond to his or her physical sex (the form of his or her geni-
tals), of transsexuality when the sexual identity is at odds with the physical sex,
and of transvestism when the gender role played by an individual does not cor-
respond to his or her physical sex. Generally speaking, the action of sex steroids
and biological variables on these various components of sexuality is extremely
variable.
4 T H E B I O L O G Y O F H O M O S E X UA L I T Y
The first aspect of human sexuality is linked to the performance of sexual

acts and the motivation behind them. It has been shown in animals that this
feature is sexually differentiated and controlled by steroid hormones during
both development (organization) and adulthood (activation) (Chapter 3).
Human sexual behavior is not limited to sexually differentiated motor actions.
Instead of having a stereotyped mating, as is the case in rats (mounting and
penetrating by the male, lordosis posture [downward arching of the spine and
raising of the hips] in the female), the human species mates in a variety of
positions. A position is often preferred in a given culture (the woman lying on
her back, face to face with the man—the so-called “missionary position”—
in Western civilization), but other positions are commonly used, and favorite
positions vary from one culture to another (Nelson, 2005). Men and women
do not play stereotypical roles in the sexual act, nor is there a major differ-
ence between the motor acts performed by both sexes, and consequently, the
animal literature on mechanisms that control dimorphism affecting this
aspect of behavior has little contribution toward understanding the human
species. Furthermore, there is little difference between men and women in
intensity of sexual motivation. Differences in modality have been reported, but
they are minor and often related to culture. This motivation, even if it is not
very sexually differentiated, remains partly under the control of sex steroids in
humans and animals, and I shall review the arguments that support this thesis
(Chapter 7).
If the motor aspects of sexual behavior are quite similar among men and
women, the three other dimensions of sexuality are highly differentiated.
Sexual orientation, a term that I prefer to sexual preference because of its
neutrality in relation to the will of the individual, identifies the sex of persons to
which an individual directs not only his or her behavior but also sexual fanta-
sies. Most men and women are sexually attracted and excited by individuals
of the opposite sex. They are heterosexual. However, there are individuals—
occurring at quite a regular percentage—attracted to persons of their own
sex; they are homosexual. This distinction is not absolute; following qualitative
studies from Kinsey and colleagues in the late 1940s, it was recognized that all
intermediaries could exist. It is considered that 3 to 10% of men are homosexu-
als in all cultures irrespective of their culture’s attitude vis-à-vis homosexuality.
These numbers vary slightly depending on the study methods (see Chapter 2).
The figures for women are less accurate, but should be of the same order of
magnitude with the addition of a significant population of individuals regard-
ing themselves as bisexual (attraction to women as well as men). There thus
exists in all human populations a significant proportion of gays and lesbians.
Although very significant, this ratio never exceeds 10%, which means that the
vast majority of the general population is heterosexual. Sexual orientation is
Sexuality 101 5
one of the behavioral traits most differentiated between men and women that
has been identified.
Regardless of which sex is found interesting or exciting, each human being
is also confident of belonging to one sex, male or female. This conviction is
unchangeable, and it often seems to develop during early childhood. Most
people have a sexual identity that matches their genitalia, although a small
number of people are convinced to the contrary and believe themselves “to be
born in a body that does not correspond to their gender.” They are called trans-
sexuals by sexologists. This gender identity is already apparent in early child-
hood and, contrary to what was thought until recently, it is usually difficult or
impossible to change it later (see the story of John/Joan, Chapter 2). Clinical
data suggest that the organization of sexual identity could be controlled by pre-
natal sex steroids. This idea is difficult to assess, in part because there is no
animal model for studying sexual identity. It is impossible to ask an animal,
whatever its species, to what sex it belongs. In addition to communication dif-
ficulties associated with such an undertaking, this would imply that the animal
is aware of its own body and sex, which is far from proved even though recent
research daily shows new sophisticated cognitive skills not only among pri-
mates but also in species more distant from humans, such as dolphins and even
birds such as parrots and corvids (a family including crows and jays).
The fourth and last dimension of human sexuality concerns the role played
by an individual in society. This gender role is deeply influenced by the struc-
ture of the society in which the individual lives, even if prenatal hormonal
influences also seem to play a role (see Chapter 2). Thus, housework tasks were
typically performed by women in Western Europe until the First World War,
but are now widely shared by men and women. Farm work tasks also are typi-
cally male or female depending on the human group. It would probably be futile
to search for biological bases for this type of sexual difference, even if sociobiol-
ogy suggests that selective pressures acting on our ancestors may have selec-
tively adjusted men and women for various jobs (this idea is broadly contested).
Other aspects of the gender role (play behavior during childhood, hobbies,
etc.), however, may be influenced by prenatal factors.
The four dimensions of human sexuality that I have distinguished are cor-
related and in agreement with each other in the majority of individuals. This
correlation can be expected both on the basis of coordinated learning that
affects these different features and on the basis of a prenatal hormonal envi-
ronment that plays a role in their organization. However, it is important to
distinguish them clearly, because these dimensions may be exceptionally dis-
cordant. In particular, it is critical to distinguish between orientation and gender
identity. A transsexual man who thinks he in fact is a woman may be sexually
attracted by men or by women; sexual identity does not determine sexual
attraction, or vice versa. This orientation becomes difficult to classify because it

will depend on the sex of reference considered (physical/genetic and genital
sex, or sex corresponding to sexual identity). For example, a transsexual man-
to-woman attracted to men is gay if one refers to his physical sex but hetero-
sexual if we consider his sexual identity.
For various technical and ethical reasons, the experimental analysis of the
neurobiological bases of sexuality is difficult for humans, and it must often be
limited to the study of clinical cases. The interpretation of these data usually
remains problematic because they are essentially correlative (existence of a rela-
tionship between a behavioral character and hormonal changes, for example).
It is impossible to be certain that the observed correlations are due to the effects
of one variable or the other. The confirmation of these interpretations can
therefore only come indirectly from comparison with true causal studies per-
formed on animals.
Most animal studies devoted to sex differences affecting sexual behavior
have unfortunately focused on the differences in the type of behavior patterns
that are expressed by the two sexes. These studies are therefore of little interest
for the understanding of human behavior, in which patterns of behavior
expressed by men and women during sexual interactions show little or no dif-
ference. More recently, studies of sexual orientation were also started in ani-
mals, and we will see that these studies provide important insights into the
origins of human homo- or heterosexuality. The other two dimensions of sexu-
ality, sexual identity and role, are specific to humans and cannot be studied in
animal models.
It is important to distinguish three aspects of homosexuality, namely the
expression of homosexual behavior(s), the homosexual attraction, and finally
the recognition/acceptance of this attraction vis-à-vis society. Expressing occa-
sional or even regular sexual behavior directed toward an individual of the
same sex is found frequently in various animal species (Bagemihl, 1999; Sommer
& Vasey, 2006; Poianni, 2010). This is generally associated with conditions
under which partners of the opposite sex are not (easily) available. One can
point to captive animals (zoo, farm), to animal societies where a few dominant
males monopolize all females and dominated males are prevented from mating
with them, or to groups of animals where the sex ratio (relative number of males
and females) is highly skewed so that many individuals cannot find a partner of
the opposite sex. In all these cases, homosexual behavior will often disappear
quickly when partners of the opposite sex become available (Sommer & Vasey,
2006; Poianni, 2010).
The same is true in humans: there is a fairly regular basis of homosexual
behavior in circumstances where partners of the opposite sex are not available,
such as in prisons or single-sex institutions of learning. I also discuss in this
Sexuality 101 7
book so-called primitive societies where homosexual relations are the norm
among teenagers, even encouraged, because access to girls is forbidden before
marriage (Chapter 2). In all these circumstances, it was observed that homo-
sexual behaviors regress or disappear when opposite-sex partners become
available, so there is not preferential homosexual attraction in these cases. The
homosexual activity is simply an outlet for sexual motivation without necessar-
ily being preferred to heterosexual activity.
The preferential or exclusive attraction to partners of the same sex is in con-
trast a largely human characteristic that does not occur frequently in animals
but can be induced by changes in the embryonic hormonal environment or by
injury to a specific area of the brain. Such spontaneous exclusive homosexual
attraction has been identified in certain populations of sheep (Chapter 4). This
is the preferential or exclusive attraction to individuals of the same sex that I am
talking about mainly in this book. I will show that this attraction is dependent,
at least in part, on prenatal biological phenomena that are largely beyond the
will of the concerned individual.
The open display in society of this homosexual attraction is the third and
last aspect to be considered independently. It is clear that this public revelation
can only follow prior personal acceptance of homosexual orientation. Its public
disclosure (“coming out”) is, however, a conscious act that largely depends on
the willingness of the person. This willingness interacts with the greater or
lesser tolerance of those around him or her, which will facilitate or complicate
the process. Many homosexuals, including some very famous ones (to name
a few, writer Gertrude Stein, blues singer Alberta Hunter, and tennis player
Billie Jean King among women; writer Marcel Proust, composer Peter Ilyich
Tchaikovsky, and actor Rock Hudson among men), have historically lived with
their sexual orientation in a more or less hidden condition.
The public disclosure and recognition of homosexuality are largely influ-
enced by social context and the willingness of the individual. However, homo-
sexual orientation develops in a distinctively independent manner from the
will of the individual, under the influence of biological factors that act mostly
during embryonic life. That is the point of this book.
Note, though, that I am well aware that homosexuality (and sexual orienta-
tion in general) is a complex phenomenon that is probably not the result of
a single cause. In addition, male and female homosexuality may have different
explanations, at least in part. Finally, within the same sex, it is likely that differ-
ent reasons or causes, or a combination thereof, can induce and thus explain
the homosexual orientation. I put forward in this book all the work showing
that at least a part of male and female homosexuality has a biological origin
related to endocrine or genetic factors. This does not preclude a potential
contribution of social influences and education, or conscious choices made
in adulthood. But though many popular books claim to explain homosexuality

on the basis of educational or social factors, the currently available scientific
studies show little or no influence of education on the development of sexual
orientation.
I shall adopt in this book a fully deterministic attitude, assuming that a same
cause always produces the same effects. However, in biology and even in psy-
chology, causes are often multiple and frequently combined to create an effect.
It is therefore, in some cases, difficult or sometimes impossible to formally
identify the underlying causes of a given phenomenon. This does not, however,
bring into question the deterministic principle. Accepting that events or behav-
iors are not determined would lead a scientist to admit that the causal study
of behavior is impossible. This can be justified in some areas of mental activity,
such as religious belief, but not in science. This being said, religion and science
are not contradictory; they simply represent different approaches to reality that
are not comparable and cannot be reduced to one another. I believe that animal
and human behavior can in most of its aspects be analyzed by the objective
methods of Cartesian deductive logic.
Multiple genetic, hormonal and environmental factors interact to modulate
and control individual human behavior. They include the structure of the
genome and its individual variations; and the social, hormonal, and environ-
mental stimuli to which the individual is exposed throughout his or her life-
time. In most cases it is thus impossible to attribute a behavior to a single specific
cause. A given cause only contributes to the explanation of a behavior for
a certain percentage of the observed variance. We must use statistics and prob-
abilities to draw firm conclusions. This does not, however, contradict the con-
cept of determinism.
MALE AND FEMALE HOMOSEXUALITY: HOMOSEXUALITY

AND BISEXUALITY
Because more or less negative or even hostile reactions to homosexuality and
bisexuality are present in most societies, the incidence of homosexuality
and bisexuality has always been difficult to assess. The first published attempt
at objective quantification was that of sexologist Alfred Kinsey and his collabo-
rators in the mid-twentieth century (Kinsey et al., 1948). Kinsey concluded that
about 10% of men are exclusively or nearly exclusively homosexual, while the
corresponding proportion was about 1.5% among women (Kinsey et al.,
1953).
Studies based on approaches ensuring absolute anonymity of respondents
have led to slightly different estimates. For example, in 2005, the National
Health Statistics Center of the Centers for Disease Control and Prevention
Sexuality 101 9
(NHSC) used a self-administered survey and reported that 7.1% of men and
13.6% of women feel sexual attraction for people of their own sex (Mosher
et al., 2005). But only 1.5% of men and 0.7% of women reported exclusively
homosexual attraction (see Figure 1.1). These data could be an underestimate,
but it is likely that they are closer to reality than the estimates of Kinsey and
colleagues, given the greater precautions taken to ensure the objectivity of the
responses of subjects and the representativeness of the analyzed samples.
Despite the biases that are potentially present in all such studies, reported per-
centages of homosexuality in all societies studied mostly vary between 2 and
10% (Chapter 2). While slight variations are observed, these percentages are
relatively similar in two studies conducted more than 50 years apart. The atti-
tude of society toward homosexuals has changed profoundly in this period;
based on that, we would expect an increase in their relative numbers, but the
opposite is observed in all cases for men. This stability obviously suggests ques-
tions about the mechanisms that lead to such a consistency.
The work of Kinsey and colleagues used a seven-point scale to characterize
the sexual orientation of individuals ranging from zero (totally heterosexual) to
six (exclusively homosexual). The latest NHSC research is based on a five-point
scale that seems to provide more reproducible results. As shown in Figure 1.1,
100 Men Women

90
80
10
1. Opposite sex
% of population
2. Mostly opposite sex

8 3. Both sexes
4. Mostly same sex
6 5. Same sex
0
1 2 3 4 5 1 2 3 4 5
Categories
Figure 1.1 Distribution of sexual attraction in the U.S. population of men and
women aged 18 to 44 years (according to data in Mosher et al., 2005). Note that
the vertical axis is interrupted between 10 and 80% (angled lines) in order to permit
visualization of detail in the low range of this axis.
the vast majority of individuals, for both sexes, has a preferential or exclusive
attraction to individuals of the opposite sex.
It is interesting that the curve describing the distribution of sexual orienta-
tions is different for men and women. For men, there is clearly a bimodal distri-
bution that has two peaks at the extremes (strictly heterosexual and strictly
homosexual), while among women there is instead a gradual decrease of per-
centages when moving from an exclusively heterosexual to exclusively homo-
sexual orientation. Thus, the categories homosexual and heterosexual describe
relatively separated populations in men, while this distinction is more subtle in
women.
Correspondingly, the stability of sexual orientation would be greater in men
than in women. Longitudinal studies show that bisexual women or women with
a nonexclusive preference for one sex more frequently change their sexual ori-
entation than men. Sexual orientation is, however, generally stable over time.
Only a small percentage of subjects change during their lifetime.
Finally, bisexuality, defined as the existence of any degree of attraction for
both sexes, seems to be more common than strict homosexuality. This broad
definition of bisexuality, however, includes a large number of individuals
who are attracted much more to one sex than the other. If we define a more
restrictive bisexuality as an attraction more or less equal to subjects of both
sexes, it is then observed much more rarely in men than in women. Some sex-
ologists even contest that such a bisexuality is present in men. Measures of
genital arousal (tumescence and erection) in men who were presented erotic
pictures of men and women have shown that men who call themselves bisexual
show genital arousal much more for one sex (often men) than the other (Rieger
et al., 2005) and often show excitement exclusively for photos of male subjects
(Freund, 1974) (Rieger et al., 2005). LeVay also notes that the term bisexual is
frequently used by young men who are in the process of becoming aware of and
publicly revealing their homosexual orientation (LeVay & Valente, 2006). Some
studies show that a percentage of homosexuals that can be as high as 40% iden-
tified themselves at an early stage of their young adult life as bisexual before
describing themselves later as homosexual. The frequency of true bisexuality in
men could be substantially overestimated.
Female bisexuality is, in contrast, more widely recognized. At the genital
level, most women are in fact bisexual; when shown erotic videos, their reac-
tions show genital arousal in response to men and women regardless of their
attraction to one sex or the other and their identification as homo-, bi-, or hetero-
sexual (Chivers et al., 2004). These data do not, however, exclude the existence
of exclusive homo- or heterosexuality among women. They only show that the
physiological genital arousal can be (partly) independent of verbal statements
of sexual attraction.
Sexuality 101 11
Finally, one must clearly distinguish between sexual physical attraction and
the desire for emotional intimacy. These two aspects of the human personality
are not always correlated, and persons (usually women) physically attracted by
one sex may fall in love with people of the opposite sex or of both sexes. This
dichotomy obviously complicates the definition of bisexuality. It particularly
affects women in Western cultures, but in other cultures there are also passion-
ate friendships between heterosexual men that do not include any physical
sexual element (Nardi, 1992, in LeVay & Valente, 2006, p. 228).
2
Sexuality and Sexual Orientation

Learning or Biology?
EDUCATION AND ENVIRONMENT CANNOT SUFFICE

TO EXPLAIN HOMOSEXUALITY
In this chapter I briefly review the various theories that have been advanced
to explain homosexuality from an environmental perspective and show why
these theories, although they are still widespread, are not satisfactory and are
not supported by experimental evidence. [Interested readers may usefully refer
to more comprehensive books on the subject (e.g., LeVay, 1996).] Remember
that the arguments presented here concern exclusively homosexual orientation
as defined in Chapter 1: exclusive or preferential sexual attraction for persons
of the same sex. The expression of homosexual behavior when heterosexual
partners are not available and public disclosure of homosexual orientation
(coming out) are, in contrast, deeply influenced by an individual’s environment,
previous history, learning, and education.
EFFECT OF EARLY SEXUAL EXPERIENCES

Some have argued that sexual orientation is largely determined by early life
experiences. This early experience could be traumatic and aversive (e.g., rape
Sexuality and Sexual Orientation 13
of a girl by a man) and induce a repulsion for members of the same sex as
the perpetrator of the assault: the girl would become disgusted with men and
sexually attracted by women, therefore being a lesbian. Alternatively, the early
experience could be attractive and, in this case, lead to a sexual orientation
toward members of the sex having played an active role in this experiment. One
could imagine a young boy, involved in a pedophile relationship with an adult
male, experiencing pleasure and consequently developing as homosexual and
finding sexual pleasure in the company of men (Churchill, 1967; Cameron
& Cameron, 1995). This theory is similar to the concept of impregnation
(“imprinting”) developed by ethologists, which proposes that the first experi-
ences of a young animal induce lifelong preferences (Lorenz, 1937; Lorenz,
1950).
This interpretation does not, however, take into account a number of obser-
vations. There are many societies where homosexual relations are the rule
among adolescent boys [e.g., some cultures in New Guinea, see (Diamond,
1993)]. This social structure favoring homosexuality in adolescents is imple-
mented in a more or less organized way in order to preserve the virginity of
girls before marriage. The homosexual relationship of young boys with an adult
male would also play an initiatory role. Various anthropological studies have
clearly established that in adulthood, the percentage of men in these societies
who persist in a homosexual orientation is not above the average found in other
societies that forbid juvenile homosexuality. The first sexual experiences of
these young boys do not seem to significantly affect their future sexual orienta-
tion (see Table 2.1) (Diamond, 1993).
Table 2-1. Percentages of Homosexual Men

in Different Societies
Great Britain 5.0–9.0

Japan 5.8
The Netherlands 7.8
USA 4.8
Phillipines* 2.0
Pilau (Malaysia)* 4.7
Thailand* 3.6
Mean 4.8–5.4
Table 2.1: The table lists average percentage of male homosexuality in several
European, North American and Asian countries that differ widely in their atti-
tude toward homosexual orientation. The 3 countries whose name is followed
by an asterisk are very tolerant and even favor adolescent male homosexuality.
This does not seem to affect the occurrence of homosexual orientation in adult-
hood. (Diamond, 1993).
Furthermore, there is no difference in Western societies between the frequency
of homosexuality among adults who spent their youth in unisex schools (boys
and girls being educated in totally separate locations) or in mixed-sex schools,
although it is well known that voluntary homosexual relations occur more fre-
quently in a unisex education environment. If the early life experiences theory
had some support, we should expect a higher incidence of homosexuality among
adults raised in single-sex schools, which is absolutely not the case (Wellings
et al., 1994).
It should be noted that the percentage of adult gay men is remarkably con-
stant in all human societies despite significant differences in the ease with which
this orientation is openly accepted. An analysis published by Milton Diamond
in 1993 indicated percentages fairly stable between 2 and 8% in societies as
diverse as Great Britain, the Netherlands, and the United States. In addition,
Diamond did not find any obvious relationship between the attitude of the soci-
ety vis-à-vis homosexuality and its incidence (see Table 2.1) (Diamond, 1993).
The percentage of homosexual men is no higher in very tolerant societies such
as the Philippines and Thailand than in societies where it is more difficult to live
with a homosexual orientation (Europe, United States). These comparisons
across different cultures provide percentages of homosexuality comparable to
those reported in the first systematic study of its kind, published by Kinsey (less
than 5–10% of men strictly or almost exclusively homosexual in the United
States at the time) (Kinsey et al., 1948). These estimates are also confirmed by
more recent studies: Mosher and colleagues reported in 2005 that 2 to 3% of
men would be preferentially or exclusively homosexual, while the attitude of
society vis-à-vis homosexuality has become considerably more lenient (Mosher
et al., 2005). This may be anecdotal, but it is intriguing to note that this percent-
age is in the same order of magnitude as what has been reported for homosex-
ual rams (see Chapter 4). One might therefore wonder whether this percentage
might reflect a fairly deterministic mechanism that acts in both animals and
humans and is thus more likely biological in nature than environmental.
PSYCHOANALYTIC THEORIES
On the basis of his psychoanalytic practice, Sigmund Freud developed a theory
of homosexuality stating that the dynamics of child–parent relationships ulti-
mately determine the sexual orientation of adults. For example, a dominant and
possessive mother and/or a distant or absent father might foster a boy develop-
ing a homosexual orientation by disrupting the resolution of the Oedipal phase
of psychosexual development.
These Freudian and post-Freudian theories were popular for about a half
a century, but in recent decades many studies have challenged their scientific
validity and even suggested various forms of scientific fraud (from pure lie to
partial concealment of results) from Freud himself, and the Oedipal theory has
been thoroughly discredited for over two decades now on this basis alone, not
to mention that psychoanalysis in general has been described as a nonscientific
theory because it is not falsifiable or refutable (Karl Popper sense). It has a rea-
sonable internal consistency but has systematically refused to consider obser-
vations going against the theory (Van Rillaer, 1980; Bénesteau, 2002; Dufresne,
2007; Onfray, 2010). In American and British scientific circles, it is now largely
ignored, but it must be dealt with here because of its persistence in the culture,
since the beliefs of the general public understandably lag behind science’s
advance.
While some observations seem to support psychoanalytic theory, keep in
mind that “correlation is not causation.” Thus, while retrospective studies have
shown that homosexual men tend to describe their relationship with their
mother as closer than the average and with their father as distant and even hos-
tile (Bell et al., 1981; Freund & Blanchard, 1983), it may be that parental atti-
tudes of indifference or hostility toward a child who is going to become
homosexual are induced by the character traits of the child that do not meet
the expectations of parents. It is in this context important to note that one of
the best predictors of adult homosexuality is the presence during childhood of
gender-nonconformism. Children that will become gay adults conform less
strictly to gender norms. Boys who will become gay engage in less aggressive
play, less typical boys’ activities, and enjoy interacting with girls and engage in
typically female activities. The reverse is true for girls who will later become
lesbians. It is thus quite conceivable that the feminized attitude of a pre-gay boy
is by itself responsible for the more distant relationship with the father. If the
remote or absent attitude of the father were responsible for homosexuality in
boys, one would expect an increased incidence of male homosexuality in chil-
dren raised by single mothers, a situation encountered quite frequently in
Western Europe and the United States during the last 20 to 30 years. However,
not a single study has been able to identify such an effect.
THEORIES BASED ON LEARNING DERIVED FROM BEHAVIORISM

Following the discovery of numerous rules governing the acquisition of
behaviors by classical conditioning (Pavlovian) and especially by operant or
Skinnerian conditioning, Behaviorism attached great importance to the influ-
ence of learning in the development of animal and human behavior (Skinner,
1965; Richelle, 1966; Skinner, 1971). No behaviorist or psychologist would now
defend the idea that behavior can be entirely innate or entirely acquired. All
behaviors are partly innate and partly acquired by learning; the respective con-
tribution of these two causes varies according to the behavior in question.
Various learning theorists have suggested that sexual orientation could be
the result of conscious and unconscious learning imposed by parents, educa-
tors, and even society in general. However, homosexuality does not appear to
be a desired character trait in any Western society, so heterosexual parents and
educators would not wish to instill homosexuality in their children or students,
and probably are doing all they can to prevent its development. This learning
could only be transmitted by parents/teachers against their own will. Fur-
thermore, we now have enough data to analyze the fate of children raised by
homosexual parents who, we might imagine, might be at least more tolerant
even if they do not encourage the development of a sexual orientation identi-
cal to their own. However, the data currently available suggest that children of
gay parents are generally heterosexual (Stacey & Biblarz, 2001). Similarly, chil-
dren born to lesbian mothers who are raised by a couple of lesbians are in gen-
eral heterosexual. Thus, this behaviorist theory has not received support from
studies in the last 30 years. Conscious attempts to change the gender identity
and sexual orientation of young children also often end in failure (witness the
case of John/Joan also called Bruce/Brenda later in this chapter), and numerous
attempts to change sexual orientation of gay adults have all completely failed.
These attempts included various forms of behavioral therapy and aversive ther-
apy involving the presentation of erotic or pornographic images to homosexu-
als in association with the swallowing of an emetic (a compound inducing
vomiting) or other aversive stimulation (see LeVay, 1996 for more detail).
Medical approaches, as well, attempted to “cure” homosexuality–when it was
still considered a disease–by radical approaches such as castration, injection
of various hormones, electroconvulsive shocks, and even lobotomy (surgical
removal of the frontal lobe). American historian Jonathan Katz counted no less
than 36 methods used during a century in North America to deal with homo-
sexuality (Katz, 1976). No reproducible results could ever be achieved by all
these mutilations or behavioral therapies.
THEORIES BASED ON “SOCIAL CONSTRUCTIVISM”

Social constructivism refers to a current of contemporary sociology that envis-
ages social reality and social phenomena as being “built,” i.e., created, institu-
tionalized, and later turned into traditions. Social constructivism focuses on
the description of institutions and actions by asking how they “construct” real-
ity. The scientific facts would themselves be products of the dynamics of insti-
tutional structures. According to this perspective, sometimes called relativism,
truth itself in scientific knowledge would be related to a particular historical

and institutional context, and would have no absolute value.
The constructivist school of thought suggests that the development of sexual
orientation as hetero-, homo-, or bisexual would simply be a concept imposed
by society and internalized by the individual (Halperin, 1990). Some thinkers
have even suggested that gender identity would be built and maintained by
social interactions (Kessler & McKenna, 1978). Some aspects of sexual identity
would be easily learned and not necessarily directly related to reproduction.
Constructivism has contributed to our understanding of human sexuality in
its cultural context, but the relevance of this interpretation of sexual orientation
appears to be relatively limited. Although gender role is largely socially deter-
mined, as noted at the beginning of the book, sex steroids also clearly have
a role in some aspects of gender role, as indicated by studies of girls exposed
prenatally to abnormally high concentrations of androgens (the congenital
adrenal hyperplasy or CAH syndrome) (see Chapters 6 and 8). The determina-
tion of gender role is therefore partly in agreement with an explanation based
on constructivism, but this is not the case for two other dimensions of sexuality:
orientation and gender identity. When they deviate from the “norm” in cases of
homosexuality or transsexuality, these two aspects of personality almost always
develop in opposition to social demands, the opposite of what we would expect
if characteristics developed in response to the needs of the family. This interpre-
tation is also contradicted by the comparisons between societies that show that
the incidence of homosexuality is relatively constant in different human societ-
ies, whatever their attitude toward it. If homosexuality is a cultural construct, its
distribution should vary depending on the attitude of society.
No theories that attribute the development of homosexuality to nonbiological
causes have produced convincing data to support their interpretations.
If any role of social and educational factors exists, that it so far has escaped
a rigorous demonstration strongly suggests that these roles are severely limited.
It should be noted that, conversely, biological mechanisms that can repro-
ducibly induce homosexuality in various animal species have been identified,
especially by research in genetics and behavioral endocrinology, research that
we will spend the bulk of this book examining. All these studies suggest that
human or animal homosexuality is the result of an interaction between genetic
and hormonal embryonic factors with probably a fairly minor effect of postna-
tal social and sexual experiences.
BIOLOGICAL CONSTRAINTS ON LEARNING

Nevertheless, many authors, often under the influence of sociology, psychology,
or psychoanalysis, believe that human behavior and gender differences that
affect this behavior are largely, if not exclusively, the result of learning and social
influences. These statements sometimes claim to be based on findings of
modern neurobiology that have demonstrated a remarkable plasticity in the
structure and function of the human brain during its development and even in
adulthood. However, this is an incomplete understanding of recent results in
neurobiology. There is indeed a great plasticity of the brain during its develop-
ment. The human brain is not fully formed at birth, and only a very small per-
centage of its neurons are mature and connected. During the first few years
of life, the vast majority of the neurons that were present at birth will die; only
a small number of them (one in ten by some estimates) will survive and be
incorporated into functional circuits. The human brain is thus extremely plas-
tic, and this plasticity is guided in part by learning. The adult brain is thus
largely the result of experience gained during development. This does not mean,
however, that behavioral predispositions do not exist at birth that will resist the
potential influence of learning.
Now that the human genome has been sequenced and to some extent inter-
preted, we know that it contains only 25,000 to 30,000 genes [sequences of
deoxyribonucleic acid (DNA) directing the synthesis of one or several proteins;
see Chapter 3]. The amount of information contained in these 25,000 genes is far
from sufficient to account for the complexity of the tasks that are performed by
a human subject and thus by his or her brain. For example, it is impossible to
imagine that the different concepts that we master in adulthood–multiplication
tables, to take an extreme example–can be encoded in our genetic material. They
must be learned. But learning is limited to some extent by the underlying neural
structures, and these clearly depend on our genetic heritage. Learning is limited
or even impossible in other species that have different genetic makeup (rats;
even chimpanzees) or in human beings suffering from a genetic disease that has
changed the development of the brain (e.g., some cases of Down syndrome).
Learning takes place in the presence of very specific biological constraints
whose complexity is currently being unraveled. For example, modern linguis-
tics suggests that all human languages tend to use the same sounds to describe
the mother and father (mama, papa, baba), the first two words that are often
made by young children. Some see this as a relic of linguistic structures that
were present in the original language spoken by the first humans. Alternatively,
given that one limiting factor for language development in young children is
the poor motor coordination of the muscles of the larynx, one can also think
that this community of early vocabulary stems, at least in part, from the relative
ease with which these sounds are pronounced. Language in humans is entirely
cultural and learned, but it could be limited, at least partially, by anatomical and
physiological constraints. Alternatively, one might even think that some pho-
nemes of human languages are genetically controlled, even if such a proposal
will probably always remain impossible to prove. Learning therefore enhances

intrinsic genetic properties and, vice versa, genes constrain and limit the poten-
tial for learning.
Depending on the physical, functional, or behavioral trait under study, the
relative importance of genes or of environment/education varies in a major
way, but this control is almost always based on an interaction between these
factors. For some physical traits, the innate is obviously much more important,
but the genetic message cannot develop without environmental conditions that
permit it. For example, the size of an individual has a strong genetic component
(parents of tall children are taller than average), but the genetic potential can
only be expressed in the presence of adequate supplies of food and good sani-
tary conditions. This explains the rapid growth of the average stature of the
population in Western Europe during the last century. The average size of males
was 5’3” in 1900 and reached 5’9” at the end of the 20th century. Such a differ-
ence cannot be the result of genetic evolution in such a short period (100 years,
about five generations). It is the result of improved living conditions and espe-
cially the health care and diet of young children.
I shall not deny here the importance of cultural, educational, and environ-
mental factors. I accept without any problem the idea expressed by Serge Hefez
in his latest book: “Sexism in children is in full swing between two and six years
when they face all the stereotypes” (Hefez, 2007). Without a doubt, this differ-
ential treatment plays a key role in the genesis of many behavioral differences
between men and women. Western society’s recent evolution to a state of less
clearly defined gender roles is accordingly associated with a partial disappear-
ance of these differences (Guiso et al., 2008; Hyde et al., 2008).
However, I believe that these differences do not develop on a tabula rasa.
Boys and girls are not identical at birth. The presence of a penis in one sex
and a vagina in the other are the signs of a deep hormonal imprinting that
determined genital form and also most likely modulated the development of
certain aspects of the brain’s structure and function. Research on animal models
has clearly demonstrated the profound and irreversible role of hormones on
embryonic brain development and behavior. The action of sex steroids in
humans is clearly highlighted by the presence of large sex differences in genital
form and reproductive function (see Chapter 6). The study of the human brain
also shows that sex steroid receptors are present at the same locations as in
animals. Finally, many clinical observations and correlative studies show
the existence in humans of indisputable effects of sex hormones on behavior
(see Chapter 7).
There are therefore at least traces of behavioral effects of hormones in
humans, and a good part of this book will describe the respective contributions
of biological (genetic or hormonal) and environmental factors (education,
learning) to the control of human behavior. I mainly focus in this context on

differences between men and women. I demonstrate that while society indeed
plays an important role in the genesis of sex differences by dramatically expand-
ing preexisting differences, the influence of biological mechanisms can clearly
be detected in humans. This is especially the case if we look at two specific
aspects of reproductive behavior that display sex differences at their largest
magnitude: sexual orientation and gender identity.
As we study the influences of hormones on brain development and other
biological factors in this book, there are a couple of key concepts to keep in
mind: the evolutionary history of life on earth means that we can find evidence
of the causes of human behavior through the study of similar behaviors in other
animals, and the role of neurons in initiating and controlling our actions.
THE UNITY OF LIFE

Because single-cell organisms evolved into multicellular animals, vertebrates,
mammals, apes, and man, all animals and man have parts of their heritage in
common. For example, vertebrates, the large animal group to which we belong—
which includes fishes, amphibians, reptiles, birds, and mammals—all have in
common the same general organization of the body. All are organized around a
dorsal nervous system under which there is a spine and then a ventral digestive
system.
Similarly, the brains of all vertebrates are formed the same way and are
made of similar parts in humans, monkeys, rats, amphibians, and fish. They
also deal with information coming from the outside world and control the
expression of emotions and behavior through the same cellular mechanisms.
This finding is extremely important for the thesis developed in this book. The
study of animal behavior and its mechanisms, therefore, forms a logical basis
on which to elaborate explanations of human behavior and thus, probably, of
homosexuality.
All living creatures, including humans, have a functional organization based
on cascades of biochemical reactions that are identical or very similar. For
example, almost all these forms of life (with of course a few exceptions due to
adaptations) produce their energy by similar chains of biochemical reactions
that “burn” sugars into a compound called ATP (an acronym for adenosine tri-
phosphate) capable of providing energy to all parts of the cell where it is
required.
All living organisms also share a common system for encoding the genetic
information required to build a complete individual from a single cell. In the case
of sexual reproduction, adopted by most sophisticated multicellular animals,
including humans, the adult will result from an egg fertilized by a spermatozoa.
Genetic information is encoded in all living creatures on earth by special mole-

cules called nucleic acids that exist in two forms: deoxyribonucleic acid (DNA)
and ribonucleic acid (RNA). The structure of DNA and RNA is nearly identical
(except for a few details) in all living beings; encoding of the information is based
on the same four components (adenine, thymine, guanine, and cytosine) orga-
nized in two associated strands: the famous double helix of DNA.
This unity of life enables researchers to extrapolate conclusions from studies
on animals and thus to draw conclusions for humans. This is one of the impor-
tant foundations of biomedical research.
NEURONS GOVERN ALL OUR ACTIONS

Before entering the maze of animal research, which strongly suggests that
homosexuality in humans or animals has deep biological bases, I would like to
review a few philosophical and scientific concepts that are needed to better
understand the implications of the experimental data I discuss.
With the development of the neurosciences, the idea that brain activity is the
basis for any mental activity took hold. The mainstream of scientific thought
now considers the very complex but highly material brain (rather than some
nonphysical cause) to be the initiator and controller of behavior, emotions, and
thoughts.
Anatomical analysis of the brains of patients who in their lifetimes had expe-
rienced brain injuries (mechanical accident; injury by object penetration or
stroke) demonstrated already during the 19th century that deficits not only of
the motor systems but also of behavioral and cognitive capacities of the patient
could often be correlated with the exact location of the lesion as detected post-
mortem. This approach enabled, for example, the identification of brain areas
that control the production of language in humans, areas that were labeled after
the names of their discoverers, Pierre Paul Broca (1824–1880) and Carl Wernicke
(1848–1905).
Furthermore, during the 20th century, advances in pharmacology made it
possible to treat neurological disorders by administration of drugs that affect
specific chemical systems of communication between neural cells. Neurons
communicate with each other by chemical messengers called neurotransmit-
ters. Any alteration in the activity of these neurotransmitters has deep repercus-
sions, often very specific, on our behavior. Thus, Parkinson’s disease, which is
characterized mainly by a loss of motor control, results from the degeneration
of a particular type of neurons that communicate through a neurotransmitter
called dopamine. The decrease of the communication supported by dopamine
is responsible for the motor problems observed in Parkinson’s patients. If the
patient is administered a drug that corrects dopaminergic signaling, his or her
motor disorders are corrected at the same time, clear evidence of their specific
chemical origin.
At a more complex behavioral level, it is well established that drugs that aff-
ect the transmission of information between neurons by another neurotrans-
mitter, serotonin, have reproducible effects on the mood and general state of
well-being of the patient. Several drugs are able to change the transmission
by serotonin and accordingly have major effects on mood. Also note that addic-
tive drugs (opium, cocaine, heroin, alcohol, etc.) are actually chemical com-
pounds extracted from plants or prepared by man that modify the activity of
specific neurotransmitter systems and consequently the perception of the world
and the behavior of consumers (Cooper et al., 1996).
Finally, the most accurate data that now support the notion that the brain
controls our behavior as well as our mental activity comes from sophisticated
studies performed during the last 10 to 20 years thanks to the development of
medical imaging techniques such as positron emission tomography (PET) and
functional magnetic resonance imaging (fMRI). These techniques, based on
principles of physics too complex to be explained here, allow the researcher or
the physician to identify brain areas that display, at some point in time, a high
electrical or metabolic activity (Miller, 2008; Op de Beeck et al., 2008). This
information is collected in the form of anatomical images on a computer screen.
It can be obtained from awake patients who must simply remain motionless
with their head placed in the imaging device. It is then possible ask the subject
to perform a specific task or produce a specific mental activity (e.g., imagine
a sequence of words beginning with the same letter, make mental calculations,
pretend to play a specific social role, etc.) and to identify which areas of the
brain are active at that time. By combining the progress of these imaging tech-
niques with sophisticated experimental protocols, researchers have been able to
locate nerve sites controlling subtle mental activities such as handling specific
concepts (numbers, words, etc.) or the expression of emotions (anxiety, disap-
pointment, etc.).
We will see that the brain areas that control sexual behavior of mammals
are identified and their functioning is well understood. These brain areas are
present in humans and are anatomically and functionally very similar to their
counterparts in animals. The understanding of human behavior can thus be
based, to some extent, on the results of experiments conducted on various spe-
cies of mammals and other higher vertebrates.
It should nevertheless be noted that if the brain controls behavior, behavior
can conversely have a massive impact on brain function and even brain structure.
The human brain remains plastic throughout adulthood and is influenced by the
lifestyle of an individual and the behavior he or she expresses. For example, it has
been shown that London taxi drivers have a hippocampus (a brain structure
specialized in the treatment of spatial memory) that is larger than in control

individuals. They apparently “develop” this part of their brain by intensive use.
This effect of behavior on brain structure obviously complicates the interpreta-
tion of many data. The interpretation of correlations between structure and
function in the brain must therefore always be done with caution: the direction
of causal relationship, if any, is not necessarily obvious.
EDUCATION DOES NOT DETERMINE SEXUAL

ORIENTATION AND SEXUAL IDENTITY
Now we return to the topic examined at the start of this chapter: education.
It was long believed, and some still believe, that the sexual identity and sexual
orientation of an individual are determined by the education received during
early childhood. The newborn baby would be completely neutral; during the
early years of life after birth the little boy or girl would become aware of his or
her sex. The developing baby would build his sexual identity and, correspond-
ingly, establish the foundation for what would later be his sexual orientation
based only on the social interactions he would experience.
Sexologists themselves defended until very recently, and some still defend,
the idea that sexual identity is a cognitive concept that develops in young chil-
dren during the early years of life and is completely dictated by the type of
education received. The newborn child is from this point of view seen as
a tabula rasa, and it would be possible to impose the desired identity by appro-
priate education (Money & Ehrhardt, 1972) and, conversely, inappropriate edu-
cation could impose an undesired identity. Based on this belief, it was customary
in the medical community to assign a gender more or less arbitrarily to children
born with intersex genital structures. Since it is easier to surgically reconstruct
female (vaginoplasty) than male (penile reconstruction) genital structures in
children born with a penis poorly or not developed, female sex was more fre-
quently assigned to children and it was thought that subjects could deal with
this assignment very well.
However, this attitude ignores the results of more than 50 years of biological
research that has, in recent decades, accumulated data clearly showing that the
young baby is not neutral from the sexual point of view. The genetic material of
the embryo has led during the early phases of embryonic life to the development
of testes or ovaries. The embryo has consequently been exposed to sexually dif-
ferentiated concentrations of sex steroids (in particular, high levels of testosterone
in male embryos) that have a profound influence on later psychosexual develop-
ment. Much of the present book will be devoted to discussion of these data.
The notion of a hormonal embryonic imprinting connected in normal phys-
iological conditions to genetic sex that has irreversible consequences on further
development and on the sexual identity and sexual orientation is slowly finding
its place in general public awareness. Medical doctors, and even sexologists,
often largely underestimate its importance, yet many clinical cases clearly high-
light the permanent effects of the embryonic hormonal milieu. A typical example
of this influence hit the headlines in the late 1990s and made a major contribu-
tion to the initiation of a change in interpretation of sexual development that
I hope will become widespread. Because of its exemplary character and the his-
toric role it played, this clinical case deserves to be exposed in detail here.
Two Canadian monozygotic twins named Bruce and Brian Reimer, born in
the early 1960s, were affected at the age of seven months by phimosis, a benign
problem in which the glans penis is stuck in the foreskin. Doctors advised the
parents to circumcise the children, but during the operation, the penis of Bruce,
one of two boys, was completely destroyed by electrocautery. Local doctors
indicated to the parents that the mutilated child would be unable in adulthood
to have a normal sex life and would suffer from social maladjustment. The par-
ents, who were obviously devastated by this prediction, consulted Professor
John Money at Johns Hopkins University to help them decide how to manage
this problem.
Because it was impossible at that time to reconstruct by surgery a penis for
the boy, Money advised that Bruce be transformed into a girl and raised as such.
Indeed, it was thought at the time that if a child were raised in an unambiguous
way as a girl, he would develop into a normal heterosexual woman in adult-
hood (Money & Ehrhardt, 1972). From that time Bruce was thus called Brenda
and was dressed and treated like a girl. At the age of two years, the sex change
was completed by a medical intervention. The child was surgically castrated
and a rudimentary vagina was constructed from the skin of the scrotum.
Brenda/Bruce and her/his brother Brian were raised as a brother and sister and
monitored periodically by Professor Money, who for years advised parents on
what to do to optimize the feminization of Brenda/Bruce.
In the years that followed, scientific reports and presentations at conferences
by Professor Money showed that the sex change was a complete success and
that Brenda (Bruce) was developing as a normal girl, apart from the fact that
she/he was slightly masculine in terms of playing activity (a feature of gender
role). According to Money, Brenda/Bruce was copying her/his activities, includ-
ing domestic activities, from her/his mother, while her/his brother Brian was
imitating his father’s activities. Brenda/Bruce chose dolls as a gift and Brian
chose cars (LeVay, 1996). This case thus supported the contention that culture
has a decisive influence on sexual identity and gender role and denied the
importance of genetics and biology on human behavior. Sexologists were con-
vinced and claimed that babies are born neutral from the sexual point of view
and if you take them early enough and you’ll make girls or boys.
Money then lost contact with the family of Brian and Brenda/Bruce. Several
decades later (in the 1990s), biologist and sexologist Milton Diamond (University
of Hawaii) and his colleague Keith Sigmundson were able to discover what had
happened to Brenda/Bruce during late childhood and puberty (Diamond &
Sigmundson, 1997). The child was in fact never socialized as a normal girl and
he rebelled early. Brenda/Bruce was refusing to wear feminine clothing, uri-
nated standing up, and felt permanently that there was something wrong.
A “female puberty” was induced by treatment with estrogen, but Brenda/
Bruce hated her/his breasts and refused the hormone. At the age of fourteen, he/
she insisted on knowing the truth, which his/her parents reluctantly revealed.
He/she was relieved to finally understand the root of the conflicting feelings he/
she experienced. At the age of 15 years, Brenda/Bruce asked people to call him/
her David and he went through additional surgery to recover a masculine iden-
tity. He underwent a double mastectomy (surgical removal of breasts) and a
phalloplasty (surgical reconstruction of a penis). He was also treated with
testosterone. He had always been exclusively attracted to women (sexual orien-
tation in agreement with genetic and hormonal sex during embryonic develop-
ment, not with sex of education during childhood) and, in adulthood, he
married a woman with whom he had male-typical sex with his prosthesis. He
also adopted children and from that time on lived a relatively normal life as
husband and father. Unfortunately he committed suicide at the age of 38 years,
for reasons that remain unknown but probably include the breakdown of his
marriage, financial difficulties, and the premature death of his brother Brian.
His traumatic childhood experiences may have also contributed to this event.
Bruce/Brenda/David Reimer was given the pseudonym of John/Joan in the
work of John Money. Before his death in 2004, however, David wanted to make
his life experience public so as to prevent such errors from recurring in the
future. His story is told in a book written by John Colapinto, As Nature Made
Him: The Boy That Was Raised as a Girl (New York, Harper Collins, 2000). This
story illustrates the dramatic importance of how our society has dealt and still
deals with issues related to sexual identity and sexual orientation while ignor-
ing the power of hormonal influences.
This case suggests a conclusion very different from what was widely accepted
by Money and his contemporaries in the 1960s. It provides evidence that human
beings are not psycho-socially neutral at birth. In keeping with their legacy as
mammals, they have trends and predispositions to react according to a male or
female pattern. According to Diamond, sexual identity is influenced by social
factors but also by the embryonic hormonal environment. John/Joan had obvi-
ously been exposed in utero to androgen levels typical of a young boy (his penis
was normally formed at birth). In this case, the hormonal factors would have
outweighed the social influences in determining identity and gender role.
A similar case of penile ablation has been reported (Bradley et al., 1998). This
genetically male (XY) subject accidentally lost his penis at the age of two months
and was raised as female from the age of seven months on. In adulthood, con-
trary to John/Joan, he adopted a female sex identity despite the fact that during
childhood he enjoyed stereotypically masculine toys and games (partially mas-
culinized sex role). His sexual orientation was essentially bisexual. He was pre-
dominantly but not exclusively attracted by women. The role of education as
compared to embryonic hormones was thus apparently more prominent in this
case than in the patient studied by John Money, possibly because the sex of
rearing (as female) was adopted earlier.
It should also be noted that these cases of penile ablation are not isolated and
that the conclusions they lead to are reinforced by the analysis of many other
clinical conditions. This is particularly true of males suffering from cloacal
exstrophy (Chapter 9), a malformation in which external genitalia do not grow
during embryonic development. Although “corrected” surgically and raised as
girls, many of these subjects have male characteristics and nearly half of them
identified as boys and later men.
3
The Hormonal Control of Sexual Behavior
BEHAVIORAL ENDOCRINOLOGY
In this chapter I present a brief review of a mass of data widely accepted in the
scientific world concerning the neuroendocrinology of behavior [for a more
detailed discussion, interested readers may refer to Becker and colleagues (2002;)
and Nelson (2005)]. Although I present a dense array of technical points, I encour-
age nonscientists to concentrate on general principles rather than trying to under-
stand every single term. In this context it is less important to know exactly what
17β-estradiol and 5α-dihydrotestosterone are and more important to know what
they do.
In this section I describe what is known concerning the endocrine control of
copulatory behavior in males and females. The elements of hormonal control of
these behaviors need not be the same as those of sexual orientation, but recent
research has shown that they frequently do overlap. Since much more research
has been done concerning copulatory behavior itself, it is clearly useful to fi rst
review this huge corpus of data. I then consider what is known specifically
about the control of sexual orientation in animals.
From a reductionist point of view, most behaviors can be considered as a series
of muscle contractions triggered by a series of organized nerve impulses. To acti-
vate a behavioral pattern, a hormone should thus alter the electrical activity in
specific neural circuits. This effect can be achieved in particular by changing

the concentration and/or activity of neurotransmitters and/or their receptors.
At this level, testosterone, estradiol, and progesterone play decisive roles.
By studying the binding sites (receptors) and the action of these hormones in
the brain, we have made progress in understanding the neural mechanisms
underlying the expression of sexual behavior. The brain is the primary site of
action of hormones in relation to the expression of sexual behavior. The periph-
eral action of hormones may also influence behavior by other means. First, hor-
mones can change the sensory inputs to the brain. For example, androgens
affect ejaculation in male rats by increasing the sensitivity of the penis. Estrogens
modulate the olfactory sensitivity of females in many species of mammals,
including humans. Sexual behavior in mammals is largely induced by olfactory
stimuli (pheromones), and thus this variation in olfactory performance could
deeply affect behavior. We will see that these olfactory responses are profoundly
altered in rodents whose sexual orientation has been altered by hormonal treat-
ments and that remnants of these mechanisms seem to be present in humans.
Second, hormones may influence the development of effector organs such as
the muscles that control the syrinx, which is the specialized structure that pro-
duces vocalizations in birds. A similar effect exists in humans: testosterone
reduces the frequency of the voice in man (makes it deeper) by acting on the
structure and muscles of the larynx, and the tone of the voice can be a signifi-
cant variable in sexual attractiveness.
Finally, hormones can change the social signals provided by external super-
ficial structures. For example, testosterone induces the growth of secondary
sexual structures such as the crest of roosters and the antlers of male deer, and
these structures in turn modify the interaction of the male with the females of
his species. Similar phenomena exist in humans: the secondary sexual struc-
tures (beard, body hair, and musculature in men; breasts and wide hips in
women) are under the control of sex steroids, and their value as sexual signals
is evident. Therefore, hormone-induced behavior modification does not result
only from the action of hormones in the brain.
WHICH HORMONES ARE INVOLVED IN THE CONTROL

OF SEXUAL BEHAVIOR?
In 1849, Berthold showed that castration of young chicks produces subjects
devoid of adult male sexual characteristics (comb and wattles) and lacking
sexual behavior. The reimplantation of the testes in these castrated males, how-
ever, restored the appearance and behavior of the male (Berthold, 1849). The
testicular secretion responsible for these physical and behavioral effects was
identified in the early 20th century as testosterone.
The Hormonal Control of Sexual Behavior 29
As already mentioned, testosterone, or similar androgenic compounds, rep-

resents the hormonal stimulus that controls male sexual activity in virtually all
vertebrates. In females, estrogens and progestagens play a similar role. It is said
that steroids activate sexual behavior. They are not by themselves capable of
triggering these behaviors, but they greatly increase the probability of their
appearance in the presence of adequate stimuli. In adults of most species, these
effects are reversible. They appear after a few days of treatment with sex steroids
and disappear at the same speed after castration or ovariectomy.
Other hormones secreted by the adrenal glands (corticosteroids), the pitu-
itary gland (oxytocin, vasopressin), or the hypothalamus also play a role clearly
identified in the activation of sexual behavior of males and females, but their
analysis is not necessary for the understanding of human behavior and sexual
orientation, which are the subject of this book. They are not considered here.
MECHANISMS OF ACTION OF SEX STEROIDS

In almost all vertebrates, the physiological and behavioral aspects of repro-
duction are thus controlled by a class of steroid hormones called sex steroids
[as always, with a few exceptions; see especially the work of David Crews at the
University of Texas, Austin (Crews & Gartska, 1982; Crews et al., 1984)].
Steroids are compounds resembling lipids (fatlike) that are synthesized from
cholesterol. They are produced in specialized endocrine glands, especially the
gonads and adrenal glands. Gonads play a dual role in reproduction. On one
hand, they produce reproductive cells (sperm or ova), which after fusion will
form the young embryo. On the other hand, they also secrete hormones that
control sexual behavior, thus allowing the interactions between the sexes and
therefore the encounter of the egg and sperm.
Three major classes of sex steroids are involved in the control of reproduc-
tion: androgens, estrogens, and progestagens. Among androgens, the best-
known example is testosterone, which is primarily secreted by the testicles.
Other, less active, androgens are also produced by the adrenals and may in
some pathological cases affect sexual differentiation in girls (see Chapter 6).
The ovary is the main source of estrogens, a class of steroids, including estra-
diol. The ovary produces at other times of the ovarian cycle large quantities of
progesterone, the progestagen prototype. These steroids are synthesized by a
chain of enzymatic reactions that convert cholesterol into pregnenolone, the
precursor of all sex steroids, and then sequentially to progesterone, testoster-
one, and estradiol in the end (Figure 3.1).
Although androgens are traditionally considered to be male hormones
and estrogens/progestagens to be female hormones, testosterone can be trans-
formed, in the ovary but also in the brain, into estradiol through the action of
CH3
CH3 CH3 C=O
CH3 CH3
O OH
Cholesterol Pregnenolone
CH3
CH3 C=O
CH3
Progesterone
OH
CH3
CH3
Testosterone
Aromatase 5α-reductase
OH OH
CH3 CH3
CH3
HO O
H
Estradiol 5α-Dihydrotestosterone
Figure 3.1 Simplified illustration of metabolic pathways for the synthesis of sex steroids
from cholesterol.
an enzyme called aromatase. This metabolic conversion plays a crucial role in

the control and sexual differentiation of reproductive behavior in mammals
and possibly also in humans. Testosterone is also converted in certain target
structures into 5α-dihydrotestosterone (by the action of 5α-reductase), and this
transformation plays a key role in the action of androgens, particularly in the

differentiation of external genital structures (see Chapter 6).
Steroid hormones produce an almost infinite number of physical, functional,
and behavioral responses. For example, testosterone is responsible for the
growth of the crest of the rooster and the antlers of the deer. It controls the
growth of the mammalian penis during puberty and activates male sexual
behavior and also aggressiveness. Testosterone or similar androgen synthetic
derivatives are used (illegally) by many athletes to increase their athletic perfor-
mance because androgens also have trophic (nourishing) effects on muscles.
Estradiol controls the development of the oviduct in the chicken, as well as in
the rat and women. It promotes the development of secondary sexual charac-
teristics such as breasts in women, induces the deposition of yolk (yellow) in
the egg of birds and mammals, and activates female sexual receptivity and pro-
ceptivity (search for male). More recently it was reported that estradiol also has
many other effects outside the field of sexuality. Estradiol and estrogens affect a
series of processes of considerable importance in the brain and other organs
such as plasticity and neuronal death, the excitability of neurons, the percep-
tion of pain (nociception), various complex intellectual processes (memory,
attention), and tumor growth in the brain and other organs such as the breast
and uterus.
Testosterone, estradiol, and progesterone act on the brain to activate sexual
behavior by molecular mechanisms similar to those described for the physical
effects at the level of peripheral organs. The sex steroids, as lipophilic compounds
(similar to fat), enter freely into their target cells (through the cell membrane) and
bind to their specific receptors. The complex formed by the hormone and its
receptor acts as transcription factors in the nucleus of these cells to activate or
inhibit at the level of DNA the synthesis of new proteins that ultimately change
the cells’ function. In the brain, these proteins, whose synthesis is modulated by
steroids, include receptors (for neurotransmitters, for steroids themselves) and
enzymes that modulate the synthesis and degradation (catabolism) of neurotrans-
mitters and receptors.
This type of action, called genomic because it implies a change in the expres-
sion of the genome, is by far the best known and probably most important in
controlling behavior. The effects it produces are relatively slow (latency of sev-
eral hours to several days) because the production of new proteins and their
incorporation into specific neural circuits is a relatively long process. In addi-
tion to this genomic action, steroids, particularly estrogens, are also able to
quickly change the physiology of the brain and therefore behavior in a short
time, which is hardly consistent with a change in protein synthesis. These
effects are induced by interaction with receptors located in the membrane of
neurons or by a more or less direct interaction with chains of chemical signaling
within neurons. The mechanisms involved remain poorly understood and are
currently the subject of a lot of research, namely in our laboratory (see, for
example, McEwen & Alves, 1999; Balthazart & Ball, 2006; Cornil et al., 2006;
Taziaux et al., 2007; Balthazart et al., 2009; Vasudevan & Pfaff, 2008; Micevych
& Dominguez, 2009). I shall not discuss them in this volume, but it is neverthe-
less important to bear in mind that steroids are probably able to produce rapid
effects on behavior with latencies of about a minute or even a second.
BINDING SITES AND ACTIVITY OF SEX STEROIDS IN THE BRAIN

The anatomical distribution in the brain of sites of action of sex steroids is remark-
ably consistent in all vertebrates, including humans. The sites are characterized
by neurons expressing specific receptors for these steroids. The receptors are
located primarily in the preoptic area, in some parts of the hypothalamus, and in
the telencephalon/cerebrum in the structures that are part of the limbic system,
such as the amygdala and bed nucleus of the stria terminalis (see Figure 3.2).
Only part of these receptors are involved in the activation of sexual behavior.
They were identified by analyzing the behavior of animals that either had been
subjected to lesions of small discrete regions of the brain or had been castrated
and had received an implant of steroid in a specific brain nucleus. In general the
medial part of the preoptic area and ventro-medial nucleus of the hypothala-
mus are sites of action of steroids necessary and sufficient to activate male and
female sexual behavior, respectively. Implantation of crystals of testosterone in
the medial preoptic area of a castrated male is sufficient to restore sexual activity
in all species of birds and mammals that were studied (Balthazart & Ball, 2007).
Nucleus of the
Cortex stria terminalis
Olfactory
bulbs Optic tectum
Septum Cerebellum
Amygdala Preoptic
area Hypothalamus
Figure 3.2 Schematic representation of the binding sites of sex steroids in the brain of
vertebrates (according to Morrell & Pfaff, 1978).
In females, estradiol and progesterone induce behaviors associated with sexual

receptivity by acting in the ventro-medial hypothalamus and in a more caudal
part in the posterior mesencephalon (the mesencephalic ventral tegmental
area). The action of testosterone in various parts of the limbic system such as the
nucleus of the stria terminalis and the amygdala is also important for the activa-
tion of male sexual behavior. I shall return to this issue in the discussion of the
physical differences of small areas of the brain that are associated with male
homosexuality and transsexuality.
CELLULAR MECHANISMS OF ACTION OF STEROID HORMONES

Studies in rats have provided most of the available data on the neurobiological
bases of male sexual behavior, but these data seem broadly generalizable to
most vertebrate species. The action of testosterone in the medial preoptic area
is necessary and sufficient to activate all aspects of sexual behavior in castrated
males. Androgens, by binding to their receptors, alter the transcription of spe-
cific genes that encode proteins such as neurotransmitter receptors (e.g., for
dopamine, norepinephrine) or enzymes that ensure the synthesis or catabolism
of these neurotransmitters. These modifications of genomic transcription result
in changes in the activity of many neurotransmitters such as dopamine, norepi-
nephrine, and serotonin, which play a key role in the activation of male behav-
ior. Steroids also affect the concentration and release of the specific neuropeptides
vasopressin/vasotocin, neuropeptide Y, or gonadoliberin (GnRH) in specific
brain structures (Figure 3.3).
Sexual activity of females in many species is limited to a period of a few days
(estrus) related to a specific endocrine state. Hormonal determinants of female
behavior differ according to species. In rats, for example, the sexual receptivity
associated with estrus is induced by elevated circulating levels of estrogen, for
about two days, followed by a peak of progesterone. Ovariectomy removes all
aspects of female sexual behavior, but behaviors can be reactivated in ovariecto-
mized females by hormonal treatment that mimics the hormonal sequence of
events (sequential treatment with estrogen and progesterone) normally observed
during the ovarian cycle. In other species (e.g., prairie vole, ferret), estrogens
alone are fully able to activate sexual receptivity in ovariectomized females. Just
as testosterone does in the male, estradiol, associated or not to progesterone, acti-
vates behavior by acting in specific regions of the brain to modify the activity of
many neurotransmitters and neuropeptides by changing either their synthesis/
degradation or the concentration of their receptors.
Many details remain to be researched before we fully understand these pro-
cesses. However, we can say that in rats and in most other animal models, the
complex neurochemical mechanisms described above are responsible for the
TESTOSTERONE
Aromatase 5α-Reductase
Estradiol 5α-Dihydrotestosterone
Estrogen Androgen
receptors receptors
Modification of
transcription
New mRNA
New proteins
Changes in
synaptic transmission
MALE SEXUAL
BEHAVIOR
Figure 3.3 Schematic illustration of hormonal mechanisms at the cellular level that
activate the expression of male sexual behavior in vertebrates.
activation of sexual behavior. Based on the continuity of evolutionary and brain

physiology between man and animals, we may think that this conclusion also
applies, at least in part, to humans. Given the moral impossibility of performing
direct experiments in humans, this conclusion is obviously supported by less
direct arguments (e.g., the analysis of clinical cases). Although it is also clear
that the behavioral and cognitive development of humans is far more complex
than that of animals, and neurotransmission mechanisms that underlie behav-
ior are also of great complexity, there is no logical reason to believe that quali-
tatively different phenomena are brought into play at the human level.
INTRACELLULAR METABOLISM OF STEROIDS

After they have entered their target cells in the brain but before binding to
their receptors, some steroids, particularly testosterone, are submitted to impor-
tant metabolic changes. Two enzymes (aromatase and 5α-reductase) catalyze
the conversion of testosterone into behaviorally active metabolites, namely
17β-estradiol and 5α-dihydrotestosterone (DHT: see Figure 3.1), respectively.
Depending on the species, 17β-estradiol and DHT, alone or in synergy, repro-
duce most, if not all behavioral effects of testosterone. The male copulatory
behavior may, for example, be activated by DHT alone in rabbits or in guinea
pigs, and by estradiol-17β alone in rats and Japanese quail. In rats, estradiol-17β,
a hormone classically regarded as “female,” is responsible for the activation of
male copulatory behavior if we look at the level of neurons where this activation
takes place. This conclusion could also be applicable, at least in part, to man
(Bagatell et al., 1994; Carani et al., 1999; Carani et al., 2005). In most species, the
two metabolites of testosterone, however, act synergistically and a more intense
response is observed after simultaneous treatment by DHT and 17β-estradiol.
The importance of the relative role of estrogen (17β-estradiol) and androgen
(DHT) metabolites from testosterone varies greatly from species to species, and
experimental data are partially lacking in humans, thus preventing an entirely
accurate assessment of its relative importance.
The activity of enzymes that metabolize testosterone varies in different
regions of the brain in specific ways depending on factors such as gender, age,
season, or hormonal status of the subjects. This produces a local variation of
the quantity of metabolites’ behavioral assets, which provides a mechanism to
precisely modulate the action of steroids on behavior.
HORMONES AND SEXUAL MOTIVATION

Sexual behavior is divided by ethologists into two distinct components: an
appetitive component that includes all behaviors to allow the encounter of the
two sexes (partner search, approach, sexual displays) and a consummatory
aspect (performance) represented by the act of mating itself (Marler & Hamilton,
1966; Everitt, 1995; Balthazart et al., 2009). The majority of studies on the neu-
robiology of sexual behavior have been devoted to the analysis of the consum-
matory aspects, but a substantial amount of work has also been devoted to the
appetitive component, also called anticipatory phase or sexual motivation. The
male sexual motivation can be measured indirectly in animals by quantifying
the time latency between the introduction of a female in the cage of a male and
his first sexual reaction, the number and persistence of pursuits of the female in
an experimental cage with several compartments, or the willingness to endure
an aversive stimulus (electrified grid) or the activity in an operant conditioning

procedure to access the female.
It is interesting to note that hormonal stimuli that activate sexual motiva-
tion appear identical to those involved in controlling performance. This can be
understood from a evolutionary point of view: for the reproduction of the spe-
cies, it is essential that the two phases of behavior take place in sequence, and it
is therefore logical that the evolution has selected identical or at least very sim-
ilar mechanisms to control the two phases of behavior. Depending on the spe-
cies, the DHT and/or 17β-estradiol formed by aromatization of testosterone in
the preoptic area and hypothalamus stimulate full appetitive sexual behavior in
castrated males. However, distinct regions of the preoptic area seem to control
the appetitive and consummatory aspects of sexual behavior in male quail, and
studies in rats indicate a double dissociation between the neuroanatomical sites
responsible for two aspects of behavior (Everitt, 1990; 1995). Indeed, it appears
that in rats, lesions of the medial preoptic area suppress copulatory behavior
but leave intact sexual motivation as measured by the pressure of a lever finally
giving access to a sexually receptive female in an operant conditioning cage.
Furthermore, lesions of the amygdala drastically reduce sexual motivation
without affecting performance. The interpretation of these data is complex,
however, and there are still many questions concerning the sites involved in
the nervous control of various components of sexual behavior. These sites
appear to be similar, but differences might also exist (Balthazart & Ball, 2007).
In contrast, it seems that the same steroids activate the two phases of sexual
behavior.
SEX DETERMINATION AND SEXUAL DIFFERENTIATION

DNA during cell division is concentrated in the nucleus to form small rods
called chromosomes, which are visible under a microscope. There are 23 pairs
in humans. In 22 of them, the two chromosomes are identical and each contains
a copy of any genetic information inherited in part from the mother and partly
from the father. The 23rd pair is formed in men (males) from two different
chromosomes, one called X and a much shorter one called Y. Women (females)
have two X chromosomes. These chromosomes separate during the cell division
that precedes the formation of sperm and egg cells (meiosis). All eggs obviously
contain an X chromosome, but sperm cells have either an X or a Y. Depending
on whether the egg is fertilized by an X or Y sperm, the embryo formed–and
later the individual–will be respectively a woman (XX) or man (XY). This system
of sex determination operates in an identical manner in all mammals that have
been studied. Only the total number of chromosomes varies from one species
to another.
Non Differentiated
(6 weeks)
Gonad
Wolffian duct
Müllerian duct
Urogenital sinus
Genital tubercle
Genital folds
Testicle Ovary
Deferent Fallopian
duct tube
Uterus
Bladder
Wolffian duct
Müllerian duct (atrophied)
(atrophied) Bladder
Glans penis
Clitoris
Male Female
(8 weeks) (8 weeks)
Masculinization (Demasculinization)
(Defeminization) Feminization
Figure 3.4 Schematic representation of the sexual differentiation process of internal

sexual organs in humans. Müllerian and Wolffian ducts are both present in the
undifferentiated embryo 6 weeks after conception. Under the influence of hormones
produced by the testes, the Müllerian ducts (in gray) regress and Wolffian ducts develop
in vas deferens and seminal vesicles. In the absence of testosterone, Wolffian ducts
(white) regress and Müllerian ducts develop in the fallopian tubes and uterus.
In early embryonic development, there is no visible difference between the

genital structures, internal or external, in the male and female embryos. It is
said that the sexual phenotype is undifferentiated (see Figures 3.4, 3.5). The
embryonic tissue that will form the embryonic gonad, a generic term referring
to the testicles and ovaries, is identical for both sexes, and the genital tract con-
tains both types of ducts: the Müllerian ducts (sketches for future female genital
tract) and the Wolffian ducts (drafts of the male genital tract).
From this undifferentiated stage, sexual differentiation proceeds in two steps.
The DNA of the Y chromosome contains a gene (SRY, for sex-determining
region of the Y chromosome) that induces the synthesis of a protein called TDF
(testis-determining factor). This protein alone determines the formation of a
male embryo.
6 weeks
Genital tubercle
Genital
swellings
Genital folds
Anus
Boys and girls
2nd trimester
Boys Girls
Birth
Clitoris
Glans penis
Uretral opening
Penis Outer labia
Testicles Inner labia
Scrotum Vaginal opening
Anus Boys Girls Anus
Figure 3.5 Embryonic development and sexual differentiation of the external genitalia
in humans. A single genital outline is present at six weeks after conception. Under the
influence of testicular testosterone and its metabolite, 5α-dihydrotestosterone, the genital
folds fuse to form the scrotum and the genital tubercle develops in the penis. Without
these hormones, genital folds form the labia of the vulva and genital tubercle forms the
clitoris.
In a first step during the early embryonic development, the SRY gene is acti-
vated and its transient expression in the gonads results in the synthesis of TDF
protein in specific cells that are the future Sertoli cells. This TDF protein stimu-
lates, directly or indirectly, the expression of many genes (organizers or architect
genes) that lead to the differentiation of the undifferentiated gonad into testis.
There is no SRY gene on chromosome X and, therefore, in the absence of the
protein TDF, the primordial gonadal tissue will not produce a testis. In humans,
differentiation of the female ovaries, however, is a complex process that is still
not fully understood and involves additional control factors.
During the second stage, specialized cells of the differentiated testis will
begin to secrete two chemical messengers, the testicular hormones, which will
spread in the whole body to organize it into a male individual (man). The inter-
stitial cells (Leydig cells) secrete testosterone, which determines the genitals’
form [fusion of the lips (labia) of the vulva to form a scrotum and develop-
ment of the genital tubercle to form a penis (see Fig. 3.5)] and maintenance of
Wolffian ducts (which form the epididymis, the vas deferens, and seminal vesi-
cles). Testosterone is also implicated in the development and organization in a
male direction of all physical and behavioral characters that are sexually dif-
ferentiated. Another type of testicular cell (Sertoli cells) in parallel will produce
anti-Müllerian hormone, which induces the regression of the Müllerian ducts.
The establishment of the male sex phenotype is therefore under the action of
two testicular hormones, testosterone and anti-Müllerian hormone.
In females, the gonads become ovaries. The absence of testicular hormones
is responsible for the persistence of the Müllerian ducts, disappearance of the
Wolffian ducts, and development of external female genitalia (labia majora and
minora, clitoris, feminization of the genitalia). The development of the female
phenotype is for the most part spontaneous. It occurs in the absence of SRY and
hence in the absence of testosterone and anti-Müllerian hormone. The female
sex is considered the “default” sex, since it develops in the absence of a specific
genetic or hormonal influence.
In summary, sexual differentiation takes place in two stages, a first leading
from genetic sex to gonadal sex and a second during which the gonadal sex
induces, by hormonal influences, the phenotypic sex differences. At birth, the
sexual phenotype is visible in many mammals, including humans. The sexual
organs are differentiated, but these organs are not functional. It is only at puberty
that the hormones (testosterone in males and estradiol/progesterone in the
females), secreted by the gonads in increasing quantities, will take control of the
reproductive function of typical adult. These hormones then control the final
maturation of the reproductive tract, the appearance of secondary sexual char-
acteristics, and sexual behavior.
BEHAVIORAL SEX DIFFERENCES AND ORGANIZATIONAL EFFECTS

OF STEROIDS DURING DEVELOPMENT
Many behaviors in animals and humans are sexually differentiated and are pro-
duced preferentially or exclusively by one sex. These behavioral differences partly
result from the presence of various hormones in adult males and females (testos-
terone and estradiol/progesterone, respectively). However, in many species, the
activation of male copulatory behavior is dependent at the cellular level on the
action of estradiol (a hormone characteristic of females), which is produced by
aromatization of testosterone in the brain. It is not the type of hormone that

determines the type of behavior that is produced, but rather the neural substrate
on which this hormone acts, with the nature of the neural substrates depending
on the sex of the subject. Estrogens are often unable to activate behavior typical
of females (e.g., responsiveness) in males, and testosterone does not activate
mounting behavior in females even after its conversion into estradiol.
We now know that these sexually differentiated behavioral responses to
steroids are the result of early actions of these steroids and that during ontogeny
the brain differentiates into a male or female brain. These differentiating effects
occur during the embryonic period or just after birth and are completely irre-
versible. In mammals, the early exposure to testosterone produces a male pheno-
type: the behavioral characteristics of the male are strengthened (masculinization)
and the ability of males to show behavior typical of females is reduced or lost
(defeminization). The female phenotype apparently develops in the absence of
hormone (or in the presence of very low levels of estrogen). In summary, the
female phenotype develops spontaneously in the relative absence of hormones
but testosterone is required to impose masculinity. The differentiation process
that takes place spontaneously during the early development of animals can be
reproduced at will by experimental manipulation of circulating concentrations
of steroids in embryos or newborn animals (Figure 3.6).
EMBRYONIC CX + T or E2
TREATMENTS (-T)
Adult female
Adult male
Pseudo-male Pseudo-female
Female behavior
After treatment with E2/P − − +++ +++
Male behavior
After treatment with T +++ +++ + +
Figure 3.6 Schematic representation of the hormonal processes during ontogenesis
that induce sexual differentiation of brain and behavior. CX: castration; T:
testosterone; E2: estradiol; P: progesterone. + + + Behavior regularly
expressed; + rare behavior; – behavior absent.
Thus, the injection of testosterone in an embryonic female rat produces a pro-

found masculinization of sexual behavior. This genetic female is capable as an
adult of responding to injections of testosterone by producing the full range of
sexual behaviors typical of males. Conversely, if a newborn male is castrated at
birth, his behavior will not be fully masculinized and he will be unable in adult-
hood to achieve the behaviors typical of his sex with high frequency. Correlatively,
his sexual behavior is not defeminized: he is capable of producing the female
posture of sexual receptivity (lordosis) in response to a sequential treatment with
estradiol and progesterone. Thus, the behavioral phenotype of rats can be com-
pletely reversed by early hormonal manipulation. One can produce individuals
who will, in adulthood, exhibit the behavioral characteristics of males or females
regardless of the genetic sex of individuals (Goy & McEwen, 1980).
As shown in Figure 3.6, prenatal injection of testosterone (but also of estra-
diol) masculinizes sexual behavior in female rats. This observation stems from
the fact that testosterone masculinizes behavior during ontogeny largely through
its conversion into estradiol (by aromatization in the brain). This notion is con-
firmed by the finding that if a genetic male, whose embryonic testis produces
large quantities of testosterone, is injected during the last week of uterine life
and the first week of postnatal life with an aromatase inhibitor, that is, a com-
pound that pharmacologically blocks the conversion of testosterone to estra-
diol, sexual behavior will not be fully masculinized. It therefore appears that
the effects of testosterone in rats are produced at the cellular level by estradiol
derived from aromatization, both during ontogeny (masculinization of behav-
ior) and during adulthood (activation of sexual behavior mature). In the next
two paragraph, I discuss two other aspects of the mechanisms of hormonal con-
trol of sexual differentiation of behavior that also have a significant impact on
the interpretation of homosexual behavior and its possible hormonal origin.
It is very important to note that the changes of behavioral sex produced by
testosterone or estradiol occur only during a limited and well-defined period of
development of the animal. This period is called the critical period and corre-
sponds to a stage of brain development during which the brain is still very plas-
tic and can therefore be changed by exposure to sex steroids. In rats, a species
in which the young are born at an early stage of development (they are naked,
blind, do not regulate their body temperature, and are unable to move by them-
selves), this period is essentially the last week of embryonic life and the first
week of postnatal life. In other species, such as the guinea pig, where the young
are more developed at birth, this period of brain development takes place
entirely during embryonic life and, therefore, the critical period of sexual dif-
ferentiation is entirely prenatal.
It is also important to note that all these organizing effects of steroids are
completely irreversible. If testosterone or estradiol are present during the
critical period and have the opportunity to masculinize and/or defeminize

sexual behavior, these effects will be present during the entire life of the animal.
A female who has been masculinized and defeminized by early exposure to tes-
tosterone (or estradiol) is unable during her whole life to present the behavior
of lordosis in response to an appropriate treatment by estradiol and progester-
one in adulthood. Correlatively, this masculinized female responds to treatment
with testosterone in adulthood by showing copulatory behavior typical of the
male. This reaction, normally characteristic of the male, remains masculinized
in the female throughout her life. Conversely, a male castrated immediately after
birth can display lordosis behavior for all his life but never responds in adult-
hood to treatment with testosterone. Contrary to the activating effects of ste-
roids, which are fully reversible and are seen only during the period of exposure
to hormones, the organizing effects are permanent and last for the entire life of
the animal from the time they were induced during early life.
The mechanisms of hormonal control of sexual differentiation of behavior have
now been identified in many species of mammals and birds and thus appear to
be general, although there are fine differences in the mechanisms involved. Thus,
the relative importance of the process of masculinization and defeminization
varies from one species to another. For example, sexual differentiation of male rats
involves an almost complete defeminization linked to a limited behavioral mascu-
linization, whereas in rhesus monkeys there is rather a marked masculinization of
behavior that is not associated with an important process of defeminization.
More subtle differences in exposure to hormones during development can also
influence the differentiation of behavior. A female rat embryo placed between
two males in the uterus will be exposed to concentrations of steroids slightly
higher and, when an adult, will show higher levels of male-typical behavior than
will control females.
Defeminization and masculinization of male mammals are induced by testic-
ular androgens; however, it is not testosterone itself that differentiates the brain
but its metabolite produced by aromatization in the brain, estradiol. The ovaries
of pregnant females produce high amounts of estradiol, which passes the placen-
tal barrier easily. One might therefore wonder why the maternal estrogen does
not also masculinize female embryos. Plasma from rat embryos contains a pro-
tein secreted by the liver, alpha-fetoprotein (AFP), which binds with high affinity
to estrogens and prevents them from entering the brain (Bakker et al., 2006). This
is why all embryos are not masculinized and defeminized by maternal estrogens.
However, AFP does not bind to testosterone, which, after being secreted by the
testes, can reach the male brain where it exerts its effects after being aromatized
to estradiol locally.
Some species, such as primates, do not appear to have discernible levels of
circulating AFP. The relative importance of the aromatization of testosterone in
the brain to sexual differentiation is also variable from one species to another,
and, in monkeys, for example, androgens themselves seem more important
than estrogens. Extrapolation to humans of the mechanisms described on the
basis of studies in animals should thus be made with due caution and still
requires verifications that must be based on clinical or epidemiological studies.
However, the main principles of control of sexual differentiation deducted from
animal studies appear to be applicable to humans (see Chapter 6). Differences
between species lie in the details.
THE SEXUALLY DIMORPHIC NUCLEUS OF THE PREOPTIC AREA,

AND OTHER SEXUALLY DIFFERENTIATED NEURAL STRUCTURES
When the process of sexual differentiation of reproductive behavior was first
identified in the late 1950s, it was thought that the brains of males and females
were physically identical. Researchers naturally assumed that, during perinatal
life, hormones defeminized/masculinized reproductive behavior by changing
specific aspects of the physiology of the brain irreversibly (such as the function-
ing of specific neurotransmitter systems), but these physiological changes were
not reflected in the neuronal structure of affected animals.
In the decade that followed, however, a few researchers analyzed the brains
of males and females in more detail with the hope of finding physical differ-
ences between sexes. In 1969 Günter Dörner demonstrated that a specific part
(the nucleolus) of the nucleus of neurons in an area of the hypothalamus (the
ventro-median hypothalamus, which controls the expression of female sexual
behavior) is larger in female rats than in male rats (Dörner & Staudt, 1969).
A few years later, Raisman and Field demonstrated via electron microscopy that
synapses in the ventro-medial hypothalamus are different in males and females
and, most important, that this sexual difference develops under the action of
testosterone during perinatal life (Raisman & Field, 1971). The treatment of
young female rats with testosterone induced synaptic connections of the male
type. This correlation suggested that the change in synaptic connectivity is
causally involved in the sexual differentiation of reproductive behavior. The
morphological differences known in the early 1970s, however, were details of
organization at the subcellular level, and it was then still believed, on a more
macroscopic level, that the brain of males was identical to that of females.
A discovery made in canaries altered that view. In the canary, as in many other
birds of the group called passerines or songbirds, males are characterized by the
abundant expression of song during the reproduction period, whereas singing is
generally absent or much more discrete in females. In the early 1970s, Fernando
Nottebohm and his colleagues at the Rockefeller University in New York identi-
fied the neural circuits that control song, and two years later Fernando Nottebohm
and Art Arnold discovered that several components of this neural circuit are
affected by a major physical sex difference (Nottebohm & Arnold, 1976). In par-
ticular, three groups of neurons of the telencephalon—HVC (formerly known as
the high vocal center), RA (the robust nucleus of the arcopallium), and area X of
the medial striatum—have a volume two to five times larger in males than in
females, depending on the species considered. The difference in size is so great
that it can be seen with the naked eye in a brain section stained by appropriate
methods.
The discovery of a large neuroanatomical sex difference in songbirds led
various researchers to reconsider the idea that such differences did not exist in
mammals. In particular, the group of Professor Roger Gorski (University of
California at Los Angeles) re-analyzed the structure of the preoptic area in male
and female rats. He discovered the existence of a sexually dimorphic nucleus
(SDN) in the preoptic area of this species (Gorski et al., 1978). This difference is
also very important: the SDN of the male rat is five times larger than the SDN
of females, and this difference may also be seen with the naked eye on a brain
section prepared for microscopic examination (see Figure 3.7).
The location of this nucleus in the middle of the preoptic area—which is,
remember, the area of the brain involved in a privileged way in the activation by
100
Volume of SDN (% of control males)
80
60
40
20
0
Ctrl CX CX Ctrl T E2 DES
(d1) +T
Males Females
Figure 3.7 In rats, the volume of the sexually dimorphic nucleus (SDN) of the preopic
area is about five times larger in males than in females—a difference induced by steroids
during the perinatal life. Castration (CX) during the first day after birth (d1) reduces
nucleus size in males, and this may be offset by treatment with testosterone (T). Perinatal
treatment of females with testosterone, estradiol (E2), or a synthetic estrogen (DES)
increases the volume of this nucleus, but these organizing effects will only be observed
later in adulthood (according to Gorski, 1984).
testosterone of male sexual behavior—strongly suggests that the nucleus plays a

key role in controlling the behavior. Many experiments have been conducted to
test this idea and have so far obtained only a partially positive answer. It was
shown that lesions in the preoptic area covering all or part of the SDN inhibit
the expression of male copulatory behavior but that more discrete lesions local-
ized specifically at the SDN are usually without effect. It seems that the SDN
and the preoptic area that surrounds it are involved in the expression of male
sexual behavior but the specific lesion of SDN is generally insufficient to inhibit
behavior.
The analysis of the hormonal mechanisms that control the volume of the rat
SDN, however, provides additional suggestions of a causal relationship between
the size of the nucleus and behavioral activity. The size of the male rat SDN does
not depend on the hormonal status of animals in adulthood but rather is deter-
mined by the endocrine environment experienced by the animal during the last
week of embryonic life and the first week of postnatal life. Thus the injection of
testosterone in young females during this critical period produces adult females
displaying a size of the SDN identical to that of a male (which is five times
greater than in untreated females). Conversely, if a small male rat is castrated at
birth, the size of his SDN in adulthood is significantly lower than that of a typi-
cal male (about half). However, the castrated rat’s SDN is larger than in a female
because, as explained above, the differentiation of the SDN is induced by testos-
terone in the last week of prenatal life and in the first postnatal weeks. Castration
at birth interrupts a process that is already under way, and the size of the SDN
is already greater than in a female who has never been exposed to testosterone
(see Figure 3.7). Once the nucleus size is determined at the end of the first week
of postnatal life, it is fixed for life; and like the type of sexual behavior (male or
female), this feature cannot be changed by hormone treatments in adulthood.
It should be noted that the “masculinization” of the volume of SDN by peri-
natal actions of testosterone is induced at the cellular level by estrogens pro-
duced in the brain by aromatization of testosterone. This conclusion is amply
demonstrated by the finding that injection of estrogen increases the SDN volume
in the same way as testosterone (if the injected dose is sufficient to saturate the
binding capacity of alpha-fetoprotein, or if a specific estrogen such as diethylstil-
bestrol, DES, which does not bind to alpha-fetoprotein, is used). Furthermore,
injection of an aromatase inhibitor blocks the effects of testosterone on mascu-
linization of the volume of SDN. These experimental arguments, and others not
described here, therefore demonstrate with confidence, in rats at least, that the
masculinization of the SDN induced by testosterone is produced in the brain by
estrogens that are derived by aromatization.
Following this discovery of the SDN in rats by Roger Gorski and his staff,
many similar sexually dimorphic structures have been discovered in the preoptic
area of many other species of mammals, including humans (see Chapter 4),
birds, reptiles, and amphibians. These dimorphic structures of the preoptic area
are apparently not homologous in different species and are not necessarily con-
trolled by the same hormonal mechanisms. The larger size in males of the preop-
tic SDN is indeed, in some species such as adult Japanese quail, the result of the
presence of a higher concentration of testosterone in males than in females. In
technical terms, this difference is linked to a different activation in adulthood by
steroids rather than to a differential organization during early life. It seems that
in most mammals studied, the sexual dimorphism in the size of the SDN of the
preoptic area is the result of a differentiation by sex hormones during early life,
either embryonic or immediately postnatal.
Finally, following this identification of an SDN in the preoptic area in many
species, neuroanatomists have analyzed the possible existence of physical dif-
ferences between other parts of the brains of males and females. They have
identified a variety of structures that are either larger in males than in females
(nucleus of the stria terminalis in rats) or in females than in males (ventro-
medial nucleus of the hypothalamus). I return to this topic in more detail when
analyzing the physical differences between the brains of men and women
(Chapter 6).
4
Biological Determinism of Sexual

Orientation in Animals
All studies of hormonal control of behavior and sex differences described in

the previous chapter deal exclusively with the mechanisms that determine the
type of behavior performed by the animal at a given time. This research, which
is focused on determining the origins of sex differences affecting behavioral
activity in the stricter sense, is of limited usefulness to explain human behavior.
The behavioral differences between men and women are indeed numerous and
include sexual orientation, which is the subject of this book. However, it is clear
that these differences do not concern the kind of sexual behaviors produced by
men and women. There is no stereotypical position adopted by men and women
in sexual relationships. Sexual relations between men and women occur in
a variety of positions and without systematic differences behavior patterns.
These positions vary from one moment to another and from one culture to
another. They are largely determined by cultural influences, but are neither ste-
reotypical nor probably controlled by specific biological or hormonal mecha-
nisms. Animal research on the analysis of mechanisms of hormonal controls
of motor patterns differentially expressed by males and females therefore pro-
poses at best analogous models for differences that affect the sexual behavior
and sexual orientation of humans.
Unlike other aspects of human sexuality that have no equivalent in the animal
(e.g., gender role or sexual identity, see Chapter 5), sexual orientation can be
easily studied in animals. The test animal can be offered a choice between
a male or female sexual partner, and the observer can record toward which
of these partners the test animal orients its sexual behavior. This type of research
is not as developed as the research on sexual behavior in the stricter sense, and
it was started more recently. However, two important principles are already
firmly established: (1) that the sexual orientation of reproductive behavior is
controlled, both in adulthood and during its development from conception
through maturation, by the same hormones that control sexual behavior, and
(2) that these hormones act in the same brain regions to activate sexual behav-
ior and determine its orientation. I summarize these data in the following
pages as they establish the theoretical context from which I propose thereafter
an explanatory model of human homosexuality based on early (embryonic)
and irreversible effects of sex steroids.
SEXUAL ORIENTATION IN ANIMALS IS CONTROLLED

BY THE HYPOTHALAMUS
In animals as in humans, sexual orientation is a characteristic of reproductive
behavior that is strongly differentiated between males and females. Most sexual
activity of males is oriented toward females and vice versa. This orientation can
be quantified in standardized experimental conditions in cages with three com-
partments (see Figure 4.1).
The test animal is placed in the central compartment of the cage and a stimu-
lus male and stimulus female are randomly placed in the left and right compart-
ments. The stimulus subjects are maintained by a harness or another similar
experimental device that prevents their passage into the central cage, while
allowing the test animal to freely visit the three compartments. This experimen-
tal procedure was used for a variety of animal species, but the most detailed
data are available for the laboratory rat (Rattus norvegicus) and the ferret
(Mustela putorius). It was observed, as might be expected, that in such a device
males spend most of their time (often more than 80–90%) in the compartment
containing a sexually receptive female, with whom the male will mate. He will
visit the compartment containing a male for much less time, and in his pre-
sence he will exhibit only investigative or aggressive behaviors. A sexually
mature and receptive female will spend most of her time in the compartment
containing a male with whom she will mate even if he is held captive in his
compartment.
The neuroendocrinologists studying behavior showed that the preference
of males or females for the compartment containing a subject of the opposite
sex is controlled by steroid hormones that, in parallel, modulate the expression
of male copulatory behavior and female responsiveness. The castration of the
Biological Determinism of Sexual Orientation in Animals 49
A
?
Stimulus male Stimulus female
B Towards
100 male
75
Percentage of tests
Ctrl Bilat Unilat Other Ctrl

50
75
MALES FEMALES
100 Towards
female
Figure 4.1 Experimental three-compartment cages used to measure sexual orientation

in small rodents and in ferrets (A), and effect of lesions of the medial preoptic area
on these preferences in the ferret (B). Male control ferrets (Ctrl) visit the compartments
containing females for most of the test time and control females preferentially visit
the male compartments. Male preferences are reversed following a bilateral lesion
(Bilat) of the preoptic area but not by unilateral lesions (unilateral) or lesions
of a different site of the brain (Other) (according to Paredes & Baum, 1995).
male or the female ovariectomy eliminates or greatly reduces the preference of

the experimental subject for the compartment containing a subject of the oppo-
site sex.
A limited number of studies have also researched which nerve sites under-
lie the expression of these preferences. Overall these studies show that sex-
ual preferences are controlled in males by the medial preoptic area. The first
experiments supporting this conclusion, and probably the most convincing,
were obtained in the ferret by Michael Baum and his colleagues at Boston
University.
Like the rat, the ferret has a sexually dimorphic nucleus at the level of the
medial preoptic area. Its volume is significantly larger in males than in females.
If we carry out a bilateral electrolytic lesion of the preoptic area at the level
of this dimorphic nucleus, males who previously spent most of their time in
the room of the three-compartment unit containing a female have a nearly
complete reversal of this choice and during the postlesional tests spend most
of their time in the chamber containing a stimulus male (Paredes & Baum,
1995). They therefore show a preference that in humans might be described as
homosexual.
This effect of lesions of the preoptic area is anatomically specific, as demon-
strated by the experimental subjects in which the lesion more or less missed
the target. Indeed, carriers of a unilateral lesion of the preoptic area (the con-
tralateral lesion missed the target) continue to show a preference for the com-
partment containing a sexually receptive female. The same is true for males in
whom the lesion has completely missed the preoptic area on a bilateral basis.
Only bilateral lesions placed exactly in the preoptic area are able to reverse the
sexual orientation of males. Similar results were observed in male rats (Paredes
et al., 1998).
Fewer data are available for female sexual preferences although one study
showed that lesion of the ventromedial nucleus of the hypothalamus (a struc-
ture that controls female receptivity) decreases the interest of female ferrets for
male olfactory signals (Robarts & Baum, 2007).
. . . AND DETERMINED BY THE EMBRYONIC HORMONAL MILIEU

The same type of experiment was used to investigate how these sexual prefer-
ences develop during ontogenesis. More specifically, the researchers wondered
whether hormonal mechanisms identical or at least similar to those that induce
the differentiation of sexual behavior (see Chapter 3) control the organization
of its orientation. A positive response was given to this question.
We have seen previously that embryonic testosterone is responsible for the
masculinization of reproductive behavior in male rats. This effect is produced
in neurons by the action of estradiol produced by aromatization of testosterone.
Indeed, injection of an aromatase inhibitor during the end of embryonic life
or during the first week of postnatal life inhibits masculinization of behavior.
It was also demonstrated that the same treatment by an aromatase inhibitor
blocks, to a large extent, the development of heterosexual preference in male
rats (Bakker et al., 1993a; Bakker et al., 1996b).
When tested in adulthood in a cage with three compartments (similar to that
described in the previous section), rats in which aromatase was blocked during
the perinatal period spend, unlike normal rats, the majority of their time in the
compartment containing another male and largely ignore the compartment
containing a sexually receptive female. Moreover, these rats rarely attempt to
mate with sexually receptive females that are present and, even more tellingly,
allow the male stimulus to mount them. They are thus showing a form of sexual
preference that, in humans, would be qualified as homosexuality or at least
bisexuality. In addition, these rats have, as adults, a small SDN in the preoptic
area (see previous chapter), which is characteristic of the female (Houtsmuller
et al., 1994). It therefore appears that absence of estrogen exposure in male rats
during the perinatal life has a lasting effect on the sexual preferences of the male
in adulthood. These rats are not masculinized and tend to prefer other males to
females.
Recent studies confirm this interpretation by showing that the treatment
(during the first three weeks of life) of young female rats with estrogen (estra-
diol benzoate) has the opposite effect. This treatment increases their preference
for females, a preference that would be classified as homosexual in women
(Figure 4.2) (Henley et al., 2009) (see Henley et al., 2011 for review). These pref-
erence tests were conducted at the adult age in ovariectomized females sub-
jected to various hormone treatments (estrogen alone or estrogen plus
progesterone) to activate different aspects of sexual behavior. The preference
reversal was observed in both hormonal conditions, thus suggesting that
“homosexual” preferences induced by perinatal treatment cannot be modified
by hormones in adulthood.
Towards female
500 1st test
400 2nd test

Preference score
300
(female-male)
200
100
Control
0
EB EB
−100 high low
−200
Towards male
Figure 4.2 Effect of treatment with an estrogen, estradiol benzoate (EB), at high
or low dose, during the first three weeks of life on the sexual preferences of female
rats. The preference score represents the time spent by the animal in the test chamber
containing a female minus the time spent in the chamber containing a male. A negative
score indicates a preference for males (usually observed in female controls). A positive
score indicates a reversal of this choice (preference for female) following treatment
with EB at both doses used (according to Henley et al., 2009).
The brains of “homosexual” male rats described above also respond in an

atypical manner to the presentation of olfactory stimuli of a sexual nature.
In control animals, the nuclei that process olfactory information in relation to
sexual behavior are indeed activated in males by presenting odor typical of
females (i.e., the cage bedding soiled by a female’s urine), while the female brain
is activated by the odor of the male. In contrast, the brain of a control male is
not activated by the presentation of a litter soiled by a male, and vice versa. This
is very different in male rats treated during the perinatal period with an aro-
matase inhibitor: their brain is instead highly activated by the odor of other
males (Bakker et al., 1996a). This change in early hormonal environment has
apparently produced a profound change in what the animal sees as sexually
attractive or exciting. We shall see that similar changes in brain activity induced
by olfactory stimuli with sexual connotations have been detected in homosex-
ual men and women (Chapter 8).
These observations, performed initially in rats, were confirmed in mice,
although in this species androgens themselves seem to play a more important
role in the differentiation of sexual preferences. Christian Bodo, working in the
laboratory of Emilie Rissman (University of Virginia), has shown that sexual
differentiation is deeply disturbed in mice suffering from a mutation of the
androgen receptor that prevents the action of testosterone or DHT (Bodo &
Rissman, 2007). This mutation, called TFM for “testicular feminizing,” has its
equivalent in humans, as we shall see in Chapter 6. In behavioral tests per-
formed on adult gonadectomized animals treated with estrogen to activate the
behavior and sexual motivation, TFM male mice had responses similar to those
of females for all characteristics that are sexually dimorphic and in which
females typically differ from males. For example, in simultaneous choice tests,
males spend more time investigating bedding soiled by the urine of females,
while females and TFM males preferred bedding soiled by the urine of males.
More females and TFM males showed no preference for one sex partner or
another in choice tests while the control males showed a strong preference for
a female partner. Finally, exposure to bedding soiled by males, but not to clean
bedding, induced neuronal activation (measured by an induction of c-fos gene
expression) in the preoptic area and the nucleus of the stria terminalis in females
and TFM males but not in control males. Recent results from the same research-
ers also demonstrated that treatment of neonatal females with an androgen
(dihydrotestosterone) masculinizes for life all these behavioral features and the
response of the nervous system to male odors (Bodo & Rissman, 2008).
Studies with genetically engineered (knock-out, KO) mice that either lack
the enzyme aromatase [and thus cannot synthesize estrogens, ArKO (Honda
et al., 1998; Bakker et al., 2002b)] or do not express the alpha-fetoprotein [and
are thus not protected from the masculinizing and defeminizing actions of
maternal estrogens during embryonic life, AFPKO (Bakker et al., 2006)] simi-
larly reveal a major impact of embryonic hormonal alterations on the sexual
preferences of adult subjects. Because these altered preferences cannot usually
be modified by adult endocrine treatments, it is assumed that they derive from
organizational effects of steroids that took place in early life, even if this conclu-
sion is often indirect and has not been formally demonstrated.
In a long series of experiments, Julie Bakker and collaborators analyzed the
preferences displayed by male and female ArKO or AFPKO mice when pre-
sented with the opportunity to investigate body odors or cage bedding soiled
by male or female stimuli. Studies with ArKO mice indicated that the develop-
ment of sexual preferences in influenced by estrogens. Gonadally intact male
ArKO mice (contrary to wild-type males) failed to show a preference for an
estrous female versus a sexually active male when asked to discriminate between
these stimuli based on volatile body odors (Bakker et al., 2002a). Accordingly,
AFPKO females that are not protected from maternal estrogens during embry-
onic life display defeminized sexual behavior and partner preferences. They are
unable to display lordosis behavior in adulthood (Bakker et al., 2006) and they
show, contrary to wild-type females, a robust preference for an estrous female
over an intact male when choosing between these stimuli based on volatile
body odors (Bakker & Brock, 2010; Brock & Bakker, 2011). Together these data
thus point to an important role of estrogens in the development of partner pref-
erence in mice. The relative role of estrogens (this section) vs. androgen (previ-
ous section) in the organization of these sexually differentiated characteristics
remains to be established.
Together these findings thus support the idea that sexual preferences in
mammals are controlled during ontogeny by the action of sex steroids, mainly
estrogenic metabolites of testosterone in rats, possibly androgens and estrogens
in mice. They are in agreement with clinical and correlative data suggesting the
same conclusion in humans (see Chapters 8 and 9).
Finally, it is important to emphasize that these early hormonal manipulations
seem to have absolutely irreversible effects on both the type of sexual behavior
to be made in adulthood (male or female typical) and its orientation (homo- vs.
heterosexual). So far, we have failed to find an experimental manipulation that
would reverse in adulthood these behavioral characteristics induced by the hor-
monal milieu during embryonic or immediately postnatal life. In particular, the
hormones secreted by adults are totally unable to change the type of sexual
behavior expressed by an individual (male or female) and its orientation (homo-
sexual or heterosexual). These animal studies play a critical role when we con-
sider the potential hormonal basis of human homosexuality.
Most mammalian species, and in particular all those species that have been
investigated in order to understand the endocrine controls of sexual preferences
(rats, mice, ferrets, sheep), do not form stable pair bonds. This lack of stable
bonds obviously prevents us from investigating this important aspect of sexual
preference. In contrast, the majority of avian species form intersexual bonds
that last either for one reproductive season or sometimes for the entire life.
A highly relevant body of research has been carried out on the development
of these sexual bonds in one songbird species, the zebra finch (Taeniopygia
guttata) (summarized in Adkins-Regan, 2011). Zebra finches breed colonially
but form socially monogamous heterosexual pairs at a young age that usually
last until one member of the pair dies. Multiple behaviors that can be easily
recorded attest the presence of this pair bond, such as clumping (perching in
close contact), allopreening (preening each other), and spending time in a nest
box together. These behaviors are observed almost exclusively within hetero-
sexual pairs. Homosexual pairs do not occur in the wild and do so only rarely
in captivity.
Although social context during rearing does affect partner preferences, these
behaviors also seem to be controlled to a large extent by the action of steroids
during development (organizational effect) but not in adulthood. Adult castra-
tion or treatment with exogenous androgens or estrogens will eventually mod-
ulate the expression of the behaviors indicative of pair bonding (decrease and
activation of the behaviors, respectively) but will never modify their direction
(toward a male or a female). The heterosexual direction of the sexual partner
preference in zebra finches does not depend on the action of sex steroids in
adulthood.
In contrast, if young females are treated with estrogens during the first two
weeks posthatch, they will prefer other females over males in two-choice prox-
imity tests and will pair with females in mixed-sex aviaries where they have
ample choice of partners (Adkins-Regan & Ascenzi, 1987; Mansukhani et al.,
1996; Adkins-Regan, 1999). It is quite interesting that this “homosexual” part-
ner preference only develops if the estrogen-treated females are housed in all-
female aviaries during their development. Neither all-female housing alone nor
estrogen treatment alone produce a significant modification of the sexual part-
ner preference. This finding obviously raises the question of what is actually
modified by the treatment with estrogens. It has been speculated that the steroid
masculinizes the way in which females learn (or become imprinted on) the sex
of their partner, but additional work would be need to resolve this question.
These experiments therefore suggest that, in birds also, the action of steroids
during development modify sexual preferences in a long-lasting manner.
Interpretation of the results is complicated, however, by a number problems,
including the facts (1) that the estrogen treatment that was used has been shown
to display some toxicity and (2) that different treatments with aromatase inhibi-
tor, which should decrease estrogen production and thus have an effect opposite
to that of treatments increasing estrogen concentrations, also resulted in a shift
from opposite-sex to same-sex preference in females and did not affect partner
preferences in males (Adkins-Regan et al., 1996; Adkins-Regan & Wade, 2001).
Similar paradoxical results have been previously reported during the experi-
mental analysis of the sexual differentiation of singing activity and of the song
control circuits in the zebra finch brain. Mechanisms mediating the sexual
differentiation of brain and behavior in this species thus remain quite mysteri-
ous and clearly involve genetic effects that are not mediated by changes in
gonadal steroid production (the so called “direct genetic effects,” (Arnold et al.,
2004; Arnold & Chen, 2009). More work will be needed on this topic but for the
purpose of the present discussion, it should be noted that steroid action during
ontogeny is able to affect the direction of partner preference in a long-lasting
manner and the type of social bond that will be displayed by adult birds (see
Adkins-Regan, 2011 for more details).
In conclusion, in animals that were studied, sexual orientation differentiates
during ontogenesis under the influence of the same hormonal stimuli that dif-
ferentiate the expression of motor behavior patterns. The embryonic or neona-
tal hormones determine not only the type of behavior that will be present in
adulthood but also the subjects’ homosexual or heterosexual orientation.
Although social factors such as education and social environment must also
potentially be taken into account, it seems that the biological (hormonal) fac-
tors described in animals also contribute to the determinism of sexual orienta-
tion in humans (see Chapters 8 and 9).
SEXUAL DIFFERENTIATION OF THE PREOPTIC AREA

AND SEXUAL ORIENTATION IN THE RAT
It is also important to emphasize that the hormonal mechanisms that control
the development of sexually differentiated brain structures such as the SDN
of the preoptic area (see Chapter 3; Figure 3.7) are, in rats, similar to those that
control the organization of orientation of sexual behavior and sexual perfor-
mance. A newborn female rat treated early by testosterone possesses, in adult-
hood, a large volume of SDN similar to that of a control male and produces
sexual behavior patterns typical of a male toward a female. Conversely, castra-
tion of a newborn male just after birth blocks the development of the SDN and
inhibits masculinization of copulatory behavior (see Figure 3.7) (Jacobson
et al., 1981; Arnold & Gorski, 1984). Finally, if male rats are injected during
the end of their embryonic and early postnatal life with an aromatase inhibitor
(which suppresses production of estrogen by aromatization of testosterone), in
adulthood, these rats will be unable to show active copulatory behavior typical
of the male and will present a partially reversed sexual orientation. They also
possess at adult age a small SDN, characteristic of the female (Houtsmuller
et al., 1994).
Recall that these effects of the hormonal environment on the organization

of sexual behavior and its orientation, and on the size of the preoptic SDN, are
observed only if the treatments occur during the critical period of sexual dif-
ferentiation, which in rats covers the two weeks surrounding birth. The hor-
mone treatments have important and irreversible effects if they are performed
during that period, but no effect will be observed with later treatments.
That lesions of the preoptic area block the expression of male sexual behavior
and reverse its orientation (see the beginning of this chapter) led to questions
regarding the nature of links between the behavioral effects of early hormone
treatments and their effects on the dimorphic nucleus of the preoptic area. Are
the masculinization of behavior and changes in the size of the SDN caused by
testosterone independent phenomena? Alternatively, does the effect of hor-
mones on behavior lead to a change in the structure of the brain or, conversely,
is the change in the brain at the basis of behavioral change?
It is well known that the expression of a behavior can induce changes in brain
structure. For example, violinists have a particularly wide area in their motor
cortex corresponding to the finger of the left hand that they use to block the
strings on a violin (Elbert et al., 1995). Furthermore, rats reared in an enriched
environment show a large number of physical changes in the brain (including
an increased number of synapses, a more complex dendritic tree of neurons,
and possibly a larger number of neurons). These environmental effects are
probably more important for the development of the human brain than for
the most primitive vertebrates, but it is very likely that the early hormonal
effects on the size of the SDN induced by hormonal conditions in rats are
directly responsible for the changes in the expression of behavior, rather than
the reverse. It should be noted that the effects of perinatal treatment with tes-
tosterone can be observed in the young rat’s brain development long before it
reaches sexual maturity and begins to express copulatory behavior. It is there-
fore likely that changes in sexual behavior cannot be the basis of the effect
on brain structure. This leaves only two possibilities: either the change of the
brain governs the expression of behavior and orientation or both types of effects
are induced in a more or less independent fashion by the early action of testos-
terone. Even in the most minimalist interpretation (the second one), the change
in the volume of the SDN is at least a permanent physical signature of early
androgen action that differentiated behavior. This signature has major docu-
mentary value, as the volume of the SDN is apparently not affected by the hor-
monal milieu of the adult animal. In rats, the volume of the SDN enables us to
obtain retrospective information on the hormonal conditions under which an
individual developed. We shall see later that this signature in the brain may play
an important role in our interpretation of the mechanisms that potentially
induce homosexuality in humans.
HOMOSEXUAL SHEEP
All the studies mentioned above describe animals that have certain aspects
of what in humans is called a homosexual orientation. In all cases, however,
note that it is almost never an exclusively homosexual orientation, only a homo-
sexual preference in animals that are essentially bisexual. In addition, all animal
models described so far relate to homosexual behavior induced experimentally,
as opposed to behaviors that occur spontaneously. Over the past ten years,
a spontaneous model of exclusive homosexuality has been described in sheep
and is considered in detail here.
In studies of sexual behavior of male sheep, ethologists had early realized
that the sexual capacity of rams is very variable. When presented with sexually
receptive females, some males are capable of achieving as many as five or six
ejaculations during a period of 30 minutes, while others are much less active or
completely inactive in the presence of females (Resko et al., 1996; Pinckard
et al., 2000). This sexual ability is obviously the object of special attention from
farmers, as it makes some males much more productive breeding stock than
others. Hundreds of males are studied each year in behavioral tests and, consis-
tently, approximately 20% of rams do not qualify to be kept as breeding subjects
in animal husbandry.
During more precise testing, it was noted that a significant fraction of these
males with little or no reaction in the presence of females are in fact not com-
pletely asexual but have active sexual behavior if they are given another ram
as sexual stimulus. During a quantitative study carried out on more than 700
rams, 51% of subjects offered a choice between a male or female partner
oriented their behavior exclusively toward a female (male-oriented toward
female; MOF), 31% were bisexual, 10% were asexual (did not present any
behavior), and 8% oriented their sexual behavior exclusively toward another
male (male-oriented male; MOM). The intensity of motivation and sexual per-
formance among male-oriented males is not in question. It is specifically the
focus of this behavior that is atypical.
Extensive studies were then carried out to identify the potential causes of this
reversed sexual orientation (see Roselli et al., 2011 for review). It was first sug-
gested that same-sex rearing, which is common in sheep, might promote the
development of these MOM sheep. Comparisons between rams raised in mixed-
sex groups and rams raised in all-male groups revealed small differences in
young adults in the rate of mounting and ejaculations with estrous ewes, but
most rams in both groups later developed a heterosexual mate preference. Also,
a search by several independent groups for genetic determinations of sexual ori-
entation did not provide any conclusive results and selection for reproduction in
ewes did not affect sexual behavior or sexual orientation in male offspring.
Another idea put forward was that a differential sexual attraction might be
related to the ability to process sensory, in particular olfactory, cues coming
from the partner, and some limited support was obtained for this notion.
Differential endocrine responses (increases in testosterone and/or luteinizing
hormone concentrations) were detected following exposure to ewes in MOF
and MOM (Perkins & Fitzgerald, 1992; Perkins et al., 1992) but some aspects of
these studies make them difficult to interpret [see Roselli et al., (2011) for dis-
cussion]. Furthermore, brain activation, as measured by the induction of the
protein product of the immediate early gene fos, was more intense in the medial
preoptic area of MOF than of MOM, but this was observed after exposure to
stimulus ewes as well as rams (Alexander et al., 1999). The difference observed
might therefore relate to sexual orientation, as well as to an overall difference in
brain activity. Together, available results indicate that none of these mechanisms
are mutually exclusive, but none of them have been demonstrated conclusively
to play a significant role. Attention then focused on potential differences in the
preoptic area that could relate to sexual orientation.
Anatomical studies identified the existence of a sexually dimorphic nucleus
in the preoptic area of sheep similar to that observed in rats. The sexually
dimorphic nucleus of the sheep preoptic area (ovine SDN or oSDN) is approxi-
mately three times larger in males than in females (Figure 4.3) (Roselli et al.,
2004b). The male oSDN has also about four times more neurons than that of
females. It is interesting that the oSDN of the male-oriented male is signifi-
cantly smaller than in the males oriented to females and contains significantly
fewer neurons (Roselli et al., 2004a). It has a structure similar to that observed
in females, with which male-oriented males share the same sexual orientation.
The sheep SDN is also characterized by a dense expression of aromatase.
Quantitative studies of this enzyme in the preoptic area indicate that aromatase
activity is higher in rams than in ewes, just as observed in rodents. In addition,
aromatase activity in the preoptic area was significantly lower in male-oriented
males than in males oriented to females, thus making them more similar to
females (Roselli et al., 2004b). The volume of oSDN can also be measured by
quantifying, in successive sections, the surface of dense expression of aromatase
mRNA. This confirmed measurements of the oSDN obtained by conventional
histological stains indicating that oSDN is larger in MOF than in females and
MOM have an oSDN smaller than MOF.
These studies indicate that several characteristics of the preoptic area of rams
(volume and number of neurons in the SDN, aromatase activity in the preoptic
area) are correlated with sexual orientation (Perkins & Roselli, 2007). This cor-
relation raises again the problem of the nature and direction of causal links
between the volume and structure of oSDN and sexual behavior of rams. Does
the oSDN control sexual orientation or does the orientation determine the size
0.5
A
0.4
0.3
0.2
ovine SDN volume (mm3)
0.1
0
MOF Fem. MOM
1.00
B
0.75
0.50
0.25
0
Male Fem. Male Fem.
+T +T
Figure 4.3 Volume of the sexually dimorphic nucleus of the preoptic area (oSDN) in
sheep. A. oSDN volume is larger in males directed toward females (MOF) than among
females (Fem.), but the male-oriented males (MOM) have a volume similar to that of
females. B. Effect of treatment with testosterone between days 30 and 90 of gestation on
the volume of oSDN measured at 135 days of gestation (birth at approximately 150 days).
A significant increase in the volume of oSDN is observed among females but the same
treatment has no effect in males [according to Roselli et al., (2004b) and Roselli et al.
(2007)].
of the nucleus? The analysis of the hormonal mechanisms that control the
development of the nucleus provides critical information on this topic.
If rams and ewes are gonadectomized in adulthood (castration of males and
ovariectomy of females) and, after a month without hormone, all subjects are
treated for three weeks with testosterone, at autopsy it can be shown that the
size of the oSDN has not been changed by the manipulations. (The MOF oSDN
remains about twice as large as in females and MOM SDN remains significantly
smaller than that of MOF and thus similar to that of females.) These hormonal
manipulations performed in adulthood suggest that the differentiation affect-
ing the volume of oSDN precedes the establishment of sexual behavior and
potentially its orientation. This idea has been confirmed by the analysis of the
size of oSDN in embryos of sheep (Roselli et al., 2007). During the embryonic
life of the sheep, which lasts about 150 days, between the 50th and 100th days
of gestation there is a peak in the concentration of circulating testosterone in

males that is not present in female embryos. In late gestation (days 135 to 140)
the sheep SDN is already clearly discernible and its volume is significantly
greater (approximately 60%) in males than in females. This difference is obvi-
ously caused by the early peak of testosterone, which is consistent with the fact
that if female embryos are treated with testosterone (100 mg twice per week
between days 30 and 90 of gestation), at 135 days this results in females in which
the volume of oSDN is as large as that of males (for review see Roselli et al.,
2011).
All available data indicate that the volume of sheep SDN is determined, as in
rats, by embryonic exposure to testosterone and cannot be changed in adult-
hood even by major hormonal manipulations. This invariance in adulthood
and early determination indicate that the size of oSDN is determined well
before the beginning of the expression of sexual behavior and orientation. It is
thus impossible that the small size of oSDN in male-oriented rams is the conse-
quence of their behavior. Either it is the cause of the homosexual orientation or
it is the signature of an embryonic endocrine event that would have determined
independently the homosexuality of the males and the size of their oSDN.
IN CONCLUSION (IN PART . . . )

Recent research has identified neuroendocrine processes that control the dif-
ferentiation and activation of many aspects of behavior in animals. Many ques-
tions remain open, particularly about how the hormones act at the cellular level
in the brain. Although so far the bulk of research has been devoted to analyzing
the mechanisms that control the sexual differentiation and activation of motor
behavior patterns, interest has focused recently on the analysis, in animals, of
other aspects of sexuality such as sexual orientation. This work has shown that
this aspect of sexual behavior is largely controlled by the same hormones acting
to a large extent on the same nervous sites as sexual performance.
It should be noted that, apart from the study of Henley and collaborators
demonstrating an increase in preference for females in female rats injected with
estradiol benzoate during the first three weeks of postnatal life (Henley et al.,
2009), there have not been any studies devoted to the analysis of an animal
model of female homosexuality. It is impossible to know whether this relative
lack of data results from an experimental or technical difficulty or is simply the
result of a sexist inclination on the part of researchers. It would be very interest-
ing to test the sexual orientation of female sheep and rats androgenized during
their prenatal or immediately postnatal life.
5
Gender Differences in Humans
MEN AND WOMEN ARE NOT IDENTICAL

In humans, as in most animal species, there are many physical, functional, and
behavioral differences between the sexes. This means that in a quantitative study
on a sufficient number of individuals, a statistically significant difference will be
observed in the results obtained for men and women. This difference does not
necessarily have a large amplitude or biological significance, but it is statistically
significant—it is not simply the result of random fluctuations in the sample.
Physical sex differences, for example, concern not only the sexual organs
(ovaries vs. testes; oviduct, uterus, vagina vs. seminiferous ducts, seminal vesi-
cles, prostate; vulva and clitoris vs. scrotum and penis) but also various charac-
ters commonly classified under the term “secondary sexual characteristics” that
include the presence of facial hair more developed in men (beard, mustache) or
breasts in women. Neuroanatomical differences include a larger dimorphic
nucleus of the preoptic area in males than in females, larger bed nucleus of the
stria terminalis in males than in females, and greater thickness of the anterior
commissure and of parts of the corpus callosum in females than in males.
Note also differences in circulating levels of various hormones (steroid sex hor-
mones, obviously, but also hormones not directly related to reproduction), differ-
ences in various aspects of metabolism (pound for pound, men metabolize
alcohol more quickly than women), and differences in susceptibility to various

diseases, including diseases of the nervous system. Men are more frequently
affected by autism (male:female incidence of 3:1) and schizophrenia (ratio of 2:1),
while women are more often affected by depression (female:male = 2:1) and
Alzheimer’s disease (2:1) (Becker et al., 2008). Dyslexia is also twice as common
among boys as girls in both Europe and the United States. At the behavioral level,
differences between men and women are equally numerous. For example, men
are on average significantly more aggressive than women. There are also repro-
ducible cognitive differences between the sexes, but these are often of a lesser
magnitude. For example, women have higher verbal skills (use of language and
ability to spell words) than men, while men are better in certain aspects of math-
ematical reasoning and analysis of spatial relationships between objects.
The differences between men and women in physiology are, in fact, far too
numerous to be summarized here. A very detailed book of 900 pages has been
published recently on this issue, with references to more than 22,000 scientific
studies on the subject (Ellis et al., 2008). The existence of such differences is not
surprising when you consider that in animals (and probably in humans) most
genes are differentially expressed between males and females. A recent study
showed that in mice, 72% of genes expressed in the liver and 14% of genes
expressed in the brain are different in males and females (Yang et al., 2006). The
existence of differences between sexes may be the rule rather than the excep-
tion, even if the magnitude of these differences can vary greatly depending on
the species and the character in question. Epidemiological or pharmacological
studies would do well to take such differences into account more systematically,
because many experimental results demonstrated in a population of one sex
may well not apply directly to the other sex. A dose of medication that is effec-
tive and safe in one sex may be less effective or have side effects in the other sex
[see the Web site of the Organization for the Study of Sex Differences (OSSD)
at www://http.ossdweb.org].
BIOLOGY OR EDUCATION?
It is clear that the primary physical sex differences are due almost exclusively
to early hormonal influences. As noted in the previous section, the fusion of
genital folds into a scrotum and the development of the genital tubercle in a
penis are induced in the animal embryo (rats, mice) by the action of testoster-
one, which must be transformed into dihydrotestosterone (DHT) by the action
of 5α-reductase to exert these physical effects. The same is true in humans
(see Chapter 6). Well-known disturbances of these processes, observed either in
individuals with androgen insensitivity (due to a mutation in the receptor to
testosterone and DHT) or in subjects deficient in 5α-reductase, demonstrate the
Gender Differences in Humans 63
critical influence of hormones in determining these sex differences, even in

humans. Affected individuals are born with partially or completely female geni-
tals depending on the magnitude of the clinical problem. It is most unlikely that
any aspect of the environment could be influencing such hormonal actions.
Some physical characteristics are more labile (that is, they are variable, unlike
the external genital structures that are determined in their more or less final
form before birth), and although they are markedly controlled by the sex
steroids, they are also affected by the environment. For example, breasts are
normally a female secondary sexual characteristic, but accidental exposure to
overly high quantities of estrogen (by, for example, consumption of poultry
meat contaminated with these hormones) can induce their growth in boys.
Other differences may have physiological determinism that more or less
equally includes biological and environmental factors. It is clear that in a
Western population, the increased capacity that men have to metabolize alco-
hol compared to women may result partly from a differential expression of
genes in the liver (74% of genes expressed by the liver are expressed differently
in male and female mice) (Yang et al., 2006), but also from the fact that men
generally drink more alcohol than women and more frequent consumption of
alcohol stimulates greater production of the enzymes that metabolize alcohol.
Finally, in many other cases, gender differences are mainly a reflection of
past experiences of the individual. Education, activities, and the expectations of
society and parents indeed vary depending on the sex of a child. Given the
importance of learning phenomena in humans, it is not surprising that all these
factors have a profound effect on the development of differences between sexes.
This is particularly the case for many differences in behavior and cognitive
skills, but that does not mean that there is not in parallel some biological
contribution to these differences.
THE DIFFERENCES OF SEXUAL BEHAVIOR: VARIABLE AMPLITUDE

AND THE DIFFICULTY OF IDENTIFYING ORIGIN
The interaction between biological and environmental determinism is particu-
larly complex when it comes to control mechanisms of behavior in humans.
Many differing aspects of behavior are noted along gender lines. For example,
it is widely accepted (at least in Western societies) that women have more devel-
oped skills in verbal communication whereas men perform better in tasks involv-
ing mathematical reasoning or mental visualization of the rotation of an object.
These differences have been systematically studied in controlled populations and
confirmed statistically (e.g., Benbow, 1993).
However, these differences are very much rooted in the different educations
received by boys and girls. This idea is particularly supported by recent data
indicating that gender differences in math skills have greatly reduced or can-
celed, or even reversed, during the last 20 to 30 years, which have also seen an
increase in girls’ access to higher education in Western countries (Guiso et al.,
2008; Hyde et al., 2008). The intercultural comparison also shows that the more
societies treat boys and girls equally, particularly in education, the more the
difference in math ability is reduced (Guiso et al., 2008). Quite unexpectedly,
however, this more equal treatment for boys and girls increases the difference
in reading ability that already favored girls. It seems that girls have an inher-
ently higher ability to read than boys do, but that difference is partly hidden in
societies where girls are less educated than boys and is fully revealed when we
offer girls an equal opportunity to develop their capacities (Guiso et al., 2008).
It must be pointed out that the magnitude of these differences is generally
very low and there is an overlap between the performances of individuals of
both sexes. Statisticians have developed a tool to quantify and compare differ-
ences between populations. This measure is the effect size. Consider the distri-
bution of scores (see Figure 5.1) obtained by a group of men (light gray) and
women (dark gray) on a psychological test measuring a sexually differentiated
behavior or any physical variable in both sexes. The magnitude of sex differ-
ences involved is very different in the two cases that are illustrated. The size of
the effect is simply the ratio of the difference between mean values observed in
both groups (DM, DM’) to the common standard deviation of samples (SD,
SD’), a measure of the average variance within these groups.
The difference between averages (DM) is relatively important in the case
illustrated on the left, while this difference is small on the right (DM’). The vari-
ability between subjects within the same group remains more or less the same
(SD = ± SD’). The size of an effect is especially important when the ratio of DM
to SD or DM’ to SD’ is large. The size of the effect is considered large if greater
than 0.8, moderate for values between 0.8 and 0.2, and small or negligible for
lower values (< 0.2). When the size of an effect is large (left figure), the score of
DM DM’
% individuals
% individuals
Jack Jack
Lucy
Lucy Mary
John Mary John
SD SD’
Dependent variable Dependent variable

Figure 5.1 Graphic illustration of the concept of effect size (according to Nelson, 2005).
an individual can predict his or her sex with a good probability (Lucy and Mary
have scores higher than Jack and John), but this is no longer the case for small
effect sizes. On the right side of the figure, John has a higher score than Lucy for
the variable concerned, while men have an average score lower than women.
The effect size of “sex differences” for many physical variables is large (> 0.8).
This is true for measures such as the height of an individual but also for neuro-
anatomical differences such as that affecting the volume of the dimorphic
nucleus of the preoptic area. This nucleus is indeed about two times larger in
men than in women (Swaab & Fliers, 1985). In contrast, the differences in cog-
nitive abilities (e.g., verbal overall capacity) are often more limited and effect
sizes very often described as low to negligible (< 0.2). There are exceptions. For
example, mental rotation skill of three-dimensional objects is associated with a
sex difference that is considered large (> 0.8, see Figure 5.2).
Many differences in cognitive skills (e.g., mathematical reasoning, verbal
ability, spatial perception) have a low amplitude, even if they are reproducible.
They can be identified by statistics in a relatively large population, but the over-
lap of distribution curves is large, so it is not possible to determine the sex of an
individual from his score on one of these tests. In everyday life, there is no
reason to tolerate any discrimination between the sexes for such tasks. Even if
a job requires a skill known to be sexually differentiated, and for which men are
Men>Women
2.0
1. Height
2. Physical aggression
3. Tridimensional rotations
1.5 4. Math problems
5. Toys choice
6. Aggressive children play
1.0 7. Verbal fluency
Effect size
0.5
1 2 3 4 5 6
0
7
−0.5
Women>Men
Figure 5.2 Magnitude of sex differences affecting various human characteristics.
The amplitude is measured by the effect size (difference between means divided by the
variance of the population). Positive values indicate differences in favor of men, negative
values represent differences in favor of women. The effect size is large (> 0.8) for physical
differences but generally low for behavioral or cognitive differences, except for the
difference in the ability to rotate three-dimensional objects (according to Hines, 2004).
on average more efficient, it is very likely to be found in women candidates for

the position who are better than male candidates (and vice versa). Any dis-
crimination based on sex is not justified by the existence of statistical differ-
ences of the small magnitude that we are talking about here.
Some authors have sought to demonstrate, either for ideological reasons or to
counter existing discrimination, that the behavioral differences that I have men-
tioned do not exist (e.g., Vidal, 2000; Vidal, 2007). I believe that this is a false
battle. The fight against discrimination is laudable, but it will be more effective
if it is based on a correct analysis of the real situation. There are obvious behav-
ioral differences between men and women. The real question is this: Is their
origin biological (genetic, hormonal) or environmental (education, social)? It is
obvious that these differences are induced to a large extent by differently focused
educations given to boys and girls. However, this does not necessarily mean that
certain biological influences are not also involved in their determinism.
The equivalent of a verbal or mathematical reasoning can hardly be studied
in animals, but there are very similar cognitive behaviors present in animals
and humans for which the control mechanisms can be more easily analyzed.
For example, it is generally observed in spatial orientation tasks that male
rodents have superior performance to those of females. This difference recalls
sex differences affecting orientation in space that have been identified in humans
in some studies, although the size of this effect in humans is small. It was shown
that these sex differences in aptitude for spatial orientation tasks are controlled
in rodents in part by circulating levels of sex steroids, which act during develop-
ment (organizing irreversible effects) and until adulthood (activation transitory
effects) (Imwalle et al., 2006).
Sex differences affecting complex behaviors can thus be affected in parallel
by hormones and the environment in which the individual has developed. We
will see that some aspects of cognitive behavior in men and women are changed
in individuals suffering from disorders of the endocrine system—for example,
among girls who were subjected during their embryonic life to abnormally high
levels of androgens (due to a congenital hyperplasia of adrenal glands, see
Chapter 6). Also, homosexuality is associated with significant changes in cer-
tain cognitive abilities known to be sexually differentiated. This association was
used to suggest that homosexuality may represent a case of partial modification
of the sexual differentiation process. I shall return to this idea in Chapter 8.
SEXUAL ORIENTATION AND IDENTITY ARE THE MOST

DIFFERENTIATED BEHAVIORAL CHARACTERISTICS
In this chapter, I have briefly reviewed the different physical, functional, and
behavioral characteristics that have been shown to be sexually differentiated
in humans. The magnitude of these differences is highly variable. At the behav-

ioral level in particular, the size of differences in statistical terms is small. As one
might expect, it is in the sexual area that the largest differences between genders
are observed. Sexual orientation and sexual identity (two different aspects of
human sexuality that should not be confounded; see Chapter 1 for detailed dis-
cussion) are without a doubt the most sexually differentiated characteristics in
humans. In the majority of the population (over 90%) there is indeed usually no
overlap between the sexual orientation of men and women. Homosexuality in
this context represents the absence of difference between sexes for “sexual ori-
entation.” A gay man displays sexual attraction for individuals of the same sex
as a heterosexual woman does. Conversely, homosexual women show sexual
preferences for women just as heterosexual males do. The specific individual(s)
that will be selected as partner(s) might admittedly be different in these catego-
ries (men found attractive by a gay men might not be attractive to a women),
but it remains that in homosexual subjects there is a reversal of the sex that is
considered sexually attractive.
This sex difference has a remarkable consistency throughout the world and
among different cultures, independently of whether they are very or somewhat
tolerant vis-à-vis homosexuality (see Chapter 1). It is only in societies where
homosexuality is severely punished that the official statistics indicate a com-
plete absence of homosexuals. However, once the political regime changes in
these countries, in less than a generation the gay population corresponds to
what is observed in the rest of the world.
The difference between sexes is even more marked at the level of sexual
identity. The number of transsexuals, that is, individuals of one sex who think
themselves to be of the opposite sex, is even lower (well below 1%). U.S. esti-
mates are that 1 in 30,000 men and 1 in 1,00,000 women seek medical treatment
and surgery to change sex. Less important disorders of sexual identity, includ-
ing the desire to be of the opposite sex but falling short of demanding surgery,
are probably more frequent, but accurate statistics are difficult to obtain.
Statistics from the Netherlands, where medical and psychological help is widely
available for patients suffering from sexual identity problems, indicate the pres-
ence of these disorders in about 1 man out of 20,000 and 1 woman in 50,000
(Hines, 2004). These disorders remain rare exceptions, and the vast majority of
men and women self-identify as such.
We are entitled to ask what the mechanisms are that lead to the establish-
ment of a difference so stable and reproducible between men and women, even
if there are “rare” exceptions. It is clear from an evolutionary point of view that
these characteristics are extremely important, as they are closely related to
reproductive success and the sustainability of the species and therefore more
directly to the transmission to the next generation of combinations of genes
that support them (if they exist). Sexual orientation itself is highly differenti-
ated between males and females in most species that reproduce sexually (gender
identity does not exist or is impossible to study in animals). One could imagine
that the human species has invented a new way, based on learning, to ensure
that men and women have an orientation and therefore a sexual attraction
toward the opposite sex, thus ensuring successful reproduction. However, it is
worth considering in such circumstances the famous logical rule of thumb
called Occam’s razor (also known as “law of parsimony”): “Entities must not be
multiplied beyond necessity”—in other words, the simplest explanation is likely
to be the best one. The simplest explanation is that the mechanisms existing in
animals have been transmitted to humans, even if cultural controls, unidenti-
fied so far, have been added.
6
The Effects of Sex Steroids in Humans

Organizing Effects
THE NATURE OF THE EXPERIMENTAL EVIDENCE:

CORRELATION IS NOT CAUSATION
In Chapters 3 and 4 I reviewed many of the principles highlighted in various
animal models on the effects of sex steroids on the expression of reproductive
behavior, on sexual differentiation during development, and on the sexual ori-
entation. I now examine the available data suggesting that these mechanisms
are still active in the human species.
The nature of the data that suggest that similar mechanisms are found in
animals and in humans is quite different. In animals, conclusions are based
primarily on experimental studies. In general, a population of experimental
subjects is randomly divided into two or more subgroups that are then sub-
jected to two or more experimental treatments. For example, 20 rats are cas-
trated and 10 of them receive a daily injection of testosterone while the other 10
are injected with oil in which testosterone is dissolved. After one or two weeks,
during standardized tests in which males are in the presence of a sexually recep-
tive female, the subjects treated with testosterone show active sexual behavior
whereas the control subjects injected with oil are inactive. This experiment
demonstrates that injecting testosterone causes the activation of sexual behav-
ior in male rats.
We have excellent reasons to believe that the same is true in humans; however,
the nature of the data supporting this conclusion is different. For obvious ethical
reasons, performing a castration in humans purely for experimental purposes is
forbidden, or even to treat subjects chronically with testosterone. Beyond that,
there are obvious methodological difficulties associated with obtaining a reliable
measure of human sexual behavior. The data indicate, however, that male sexual
behavior of the human species is, as in rats, influenced by testosterone. The argu-
ments supporting this argument are threefold.
Comparison with animal species provides a first type of argument. It is
known that human testes produce testosterone identical to that secreted in
rodents such as rats and mice and in most vertebrates. Testosterone circulates
in the blood of adult humans at concentrations similar to those observed in
other vertebrates, and postmortem anatomical studies have shown that the
human brain has receptors for testosterone similar to those of rats or monkeys,
and these receptors are located in exactly the same areas of the brain (Abdelgadir
et al., 1999; Fernández-Guasti et al., 2000; Pelletier, 2000; Kruijver et al., 2001).
This argument is somewhat indirect and may only convince biologists, but it is
strongly reinforced by the analysis of various clinical cases and the outcome of
their medical treatment.
These clinical studies provide a second type of argument, much more direct,
although they are prone to problems of interpretation. For example, there are
human cases of clearly insufficient development of the testes that produce very
low levels of circulating testosterone (less than 10% of values considered normal).
Some of these patients consult endocrinology clinics because they complain of
various sexual disorders, including a lack of desire and sexual motivation, a lack
of erotic fantasies, and weak erections. Clinical studies conducted in controlled
conditions have shown that treatment of these patients with testosterone signifi-
cantly increased all measures of activity and sexual motivation, such as the
number of nocturnal erections, thoughts and sexual fantasies, and masturbation
episodes (Davidson et al., 1979; Bancroft, 1995; Hajjar et al., 1997; Snyder et al.,
2000; Wang et al., 2000). Conversely, pharmacological manipulations that lower
levels of circulating testosterone induce a decrease in libido and various aspects
of sexual behavior in humans (Rosler & Witztum, 1998). These studies are gen-
erally carried out in double-blind conditions in which patients are treated either
by testosterone or by a control solution (vehicle injection), and neither the doctor
nor the patient knows during the study period who received testosterone or pla-
cebo. These studies demonstrate as clearly as possible in humans the behavioral
effects of sex steroids. It must be pointed out that the conclusions that can be
drawn from such studies are not necessarily generalizable to other physiological
situations. For example, other forms of sexual disability will not be improved by
added testosterone if there is not initially a very low circulating level.
The Effects of Sex Steroids in Humans 71
In addition, there are many questions that can only be definitively answered
with experiments that are unethical to carry out on humans, in particular the
detection of brain sites where steroids act to activate the behavior. For such ques-
tions, studies are limited in humans to the use of a third strategy, namely, analysis
of correlations between accidental brain damage and behavioral and endocrine/
physiological variables. When a correlation has been established, its meaning
must be sought by other methods, including comparison with animal studies.
A correlation between two variables does not in fact demonstrate that they
are linked by a direct causal relationship. For example, if we observe that a par-
ticular behavioral characteristic such as frequency of sexual fantasies is corre-
lated with the level of circulating testosterone (either within different individuals
at one time or in an individual over time), it is possible either that testosterone
stimulates behavior, or behavior activates the secretion of testosterone, or that
the two variables are stimulated or inhibited by a third independent variable,
such as the level of stress or relaxation of the individual. Only further studies
can help interpret the observed correlation.
A substantial part of this book is devoted to a review of various types of phys-
ical, functional, and behavioral variables that correlate with homosexuality.
These correlations can be interpreted schematically in three ways: either variable
A is the cause of homosexuality (case 1), or the appearance of A is the result of
homosexuality (case 2), or A and homosexuality are both the result of a cause
that remains to be identified (case 3). It should be noted that these interpreta-
tions are not mutually exclusive and a variable can, for example, represent a par-
tial cause of homosexuality but at the same time be reinforced by the expression
of this sexual orientation (simultaneous occurrence of cases 1 and 2). In the
absence of experimental studies that manipulate the incidence of homosexuality
in a given population (ethically prohibited, obviously), the interpretation of these
correlations will remain somewhat uncertain. The presence of these correlations,
however, clarifies and restricts the possible causes of homosexuality. This is par-
ticularly the case if multiple correlations are considered in parallel and if one
compares these results with those of animal studies. It is clear that only cases 1
and 3 are likely to explain the origins of homosexuality. Condition 2 reveals only
a more or less direct consequence of this orientation. Throughout the following
discussion I shall try to clarify the potential causal significance of various char-
acteristics to be analyzed. In many cases, this meaning is not known with cer-
tainty, so I shall discuss the relative likelihood of each of these solutions.
The remainder of this section is devoted to a brief review of the available evi-
dence that is consistent with the concept and demonstrates in many cases that
the effects of steroid hormones that have been identified in animals are also pres-
ent in humans, although they are sometimes partially masked by other mecha-
nisms. The human species is indeed the result of a long evolution, and there is no
reason to believe that the mechanisms of endocrine control of behavior that have
been identified in fish and mammals—and are still quite evident in primates—
have completely disappeared in our species. I obviously do not deny the specific-
ity of the human species. But as much as it is characterized by a spectacular
development of the cerebral cortex that has taken control over virtually all aspects
of behavior, this does not exclude the survival of more primitive underlying bio-
logical mechanisms. I consider first the biological mechanisms of sex determina-
tion that are identical in animals and in humans. I then consider the current state
of knowledge on the organizing and activating effects of sex steroids on physical
and behavioral characteristics.
THE SEX DETERMINATION IN HUMANS

At an early stage of development, the embryonic gonad or germinal ridge is
undifferentiated and bipotential. The subsequent development of the gonad into
testis or ovary is determined by the cellular expression or absence of a protein,
called testis-determining factor or TDF, encoded by the SRY gene present on
chromosome Y (Berta et al., 1990; Sinclair et al., 1990). Sex determination in
humans, as in mammals, takes place following entrance into the egg of a sperm
with a X or Y chromosome.
The TDF protein then regulates the expression of many genes, leading to cen-
tral cell proliferation at the expense of outermost layers of the germinal ridge so
that the previously undifferentiated gonad develops into a testis. If an individual
does not possess the Y chromosome (or has it, but the SRY gene is defective),
the TDF protein is not produced and development of the embryonic gonad does
not lead to the formation of a testis. In the presence of additional signals, ovaries
will be formed. Once differentiated, the gonads influence sexual differentiation
through their hormone production.
The data relating to the human species clearly indicate that the same mecha-
nisms are involved as in animals during differentiation of the germinal ridge into
a testis or ovary. Besides XY men and XX women, there are indeed people who
have additional sex chromosome. XXY, XXYY, XYY, or XXX and X0 (the 0 indi-
cates the absence of a second sexual chromosome) individuals have been identi-
fied in humans. The first group of anomalies of sex chromosomes (XXY, XXYY,
XYY) is invariably associated with a male phenotype, whereas the second group
(X0 and XXX) always produces female individuals. In other words, almost any
embryo that has at least one Y chromosome develops as male and almost anyone
who does not have a Y develops as female regardless of the other chromosomes
that are present (but see below). This observation, coupled with what we knew of
sexual differentiation in animals, led doctors to postulate as early as the 1950s
that the Y chromosome includes a gene that determines the sex of the embryo.
The identification of this gene had to await progress in molecular biology and
was carried out forty years later.
The British group of Andrew Sinclair was able to take advantage of the detailed
study of a small group of individuals in which the correlation between the pres-
ence/absence of a Y chromosome and the man/woman phenotype is not present
(Sinclair et al., 1990). There are indeed very rare cases in which there is a male
phenotype in XX individuals and, vice versa, a female phenotype in XY indi-
viduals. Researchers have found that in XX men, a small fragment of the Y chro-
mosome was transposed (translocated in scientific terms) on the X chromosome.
This translocation occurs exceptionally in one of the cell divisions that lead to
the formation of sperm. A small piece of chromosome “breaks” and joins erro-
neously to another chromosome, here the X chromosome. Conversely, there are
a few individuals with XY chromosomes that have a female phenotype. In these
individuals, the same small piece of the Y chromosome was broken and was lost
during the production of sperm. By studying the DNA contained in this little
piece of Y chromosome that when lost or translocated produces a reverse sex,
British researchers were able to identify in humans the SRY (sex-determining
region of the Y chromosome) gene that encodes the information to produce the
protein TDF (testis-determining factor), which then determines the sex of the
individual. The few cases of translocation or loss of this gene clearly confirm its
key role in humans and animals.
The equivalent gene (SRY) exists in mice, and experimental evidence of the
role of SRY in humans is linked in part to the comparison with the animal (see
Chapter 3). By genetic manipulations, it has been demonstrated in mice that
specific deletion of the SRY gene leads to the development of subjects that have
a completely female phenotype even though the rest of their genetic sex is male.
Conversely the “graft” on a non–sex chromosome of a female of the SRY gene
produces a female that will develop testes, and the hormones secreted by the
testes will masculinize the physical, functional, and behavioral traits of the
subject (De Vries et al., 2002; Arnold & Chen, 2009).
ORGANIZING EFFECTS OF STEROIDS ON PHYSICAL AND

BEHAVIORAL TRAITS IN HUMANS
Physical Traits
In the embryonic stage, all individuals have the tissues that are the precursors
of male and female reproductive tracts. Thus the fetus at an early stage has a
genital tubercle that can form either a penis or a clitoris and the genital folds
that can form the lips of the vulva in females or testicular scrotum in males
(Figure 3.5). At an early stage, the embryo also contains two sets of ducts that
connect the gonads to the outside of the body: the Wolffian and Müllerian ducts
(Figure 3.4).
In humans as in other male mammals, testosterone produced by Leydig cells
of fetal testes causes the differentiation of Wolffian ducts into epididymis, vas
deferens, and seminal vesicles. Sertoli cells will produce a peptide hormone, the
anti-Müllerian hormone, or AMH, which as its name suggests, induces rapid
and complete regression of Müllerian ducts (Figure 3.4).
In the female embryo, the ovary secretes little or no steroid hormones. The
production of testosterone is virtually absent, and if estradiol is produced, it is
in low concentrations that will bind to a circulating embryonic protein called
alpha-fetoprotein that prevents the action of the steroid at the intracellular level.
Consequently, Müllerian ducts spontaneously develop while Wolffian ducts
regress completely. Sexual differentiation of the internal reproductive organs
takes place along two independent axes: (1) demasculinization-masculinization
and (2) defeminization-feminization. For normal development of the male
reproductive system, the embryonic organ precursors undergo defeminizing
hormonal effects (regression of Müllerian ducts by anti-Müllerian hormone
produced in Sertoli cells) and masculinizing effects (development of Wolffian
ducts promoted by testosterone). The female reproductive tract will appear as a
result of feminization (development of Müllerian ducts) and demasculinization
(regression of Wolffian ducts) and does not depend a priori on any hormone
production. This dual mechanism directs the main changes leading to the acqui-
sition of physical gender.
While the internal reproductive organs differentiate through development
and regression of two separate drafts originally present in both sexes, this is not
the case for the external genitalia. They differentiate in effect from the same
embryonic structures (the genital tubercle and genital folds). Depending on
whether testosterone is or is not present at a critical stage, these will change in
the male or female external genitalia (Figure 3.5). The presence of testosterone
[specifically its androgenic metabolite, 5α-dihydrotestosterone (DHT) produced
by the enzyme 5α-reductase], induces the development of the male external
genitals by fusion of the folds to form a scrotum and the development of the
genital tubercle to form a penis. In the absence of testosterone, the genital folds
do not fuse and instead form the vaginal lips, while the genital tubercle develops
only slightly and turns into a clitoris. The same result is obtained in the absence
of gonad. Thus, in humans, as in other mammals, hormone production is not
required for normal development of the female reproductive system; hence it is
generally regarded as the “default” developing sex or neutral sex.
This difference between the development of internal and external structures
has important implications for individuals exposed for various reasons to
abnormal hormonal conditions. In a series of clinical conditions, it is indeed
theoretically possible to obtain adults who possess in parallel the internal sexual
organs typical of the male and female. In contrast, external structures are more
or less male or female but never both. The genital tubercle can develop to vary-
ing degrees ranging from a “normal” clitoris to a “normal” penis through all
stages of more or less enlarged clitoris, but there is never the simultaneous pres-
ence of two types of structures, since they derive from the same embryonic
tissue.
Behavioral Traits
It should also be noted that the principles that apply to sexual differentiation of
the anatomy of the reproductive system in animals and in humans are also
involved in the emergence of many differences affecting sexual behavior. It is
clearly established in animals that typical male behaviors are generated by mas-
culinization of the male under the influence of embryonic testosterone. Typical
female behavior is present in the absence of hormones, but typical female behav-
ior would be lost following embryonic exposure to testosterone (defeminiza-
tion). Typical male and female behaviors can be considered as independent
entities that depend heavily on different areas of the brain. It is therefore con-
ceivable that these two types of behavior can persist simultaneously in an adult
individual, and this condition has been observed in animals subjected to specific
hormonal treatment during their development. Sexual differentiation of behav-
ior does not occur, therefore, along a single axis from more masculine to more
feminine, but along two more or less independent axes involving structures con-
trolling male and female behavior. We shall see later how these concepts can be
applied to the human species.
PHYSICAL CONSEQUENCES OF ENDOCRINE ABNORMALITIES

IN HUMAN EMBRYOS
Experiments conducted in animals (mainly rats and mice but also monkeys)
have clearly established the role played by testosterone in sexual differentiation
of internal and external sex organs. Numerous clinical studies confirm that the
mechanisms described in the previous section are also valid in humans, in
whom testosterone plays a critical role in the development of sex organs of male
embryos. Several diseases in the embryo that affect either the concentrations of
circulating testosterone or the action of this steroid at the cellular level have
been characterized in detail. They are invariably associated with atypical devel-
opment of external genitalia and internal structures. I describe some of these
diseases here, because they demonstrate the profound role steroids play in
human development. Furthermore, these pathologies are also associated with
behavioral changes that I discuss in Chapter 9. Four types of diseases or clinical

conditions will be useful to examine in this context. The first two specifically
affect boys, the next two concern girls.
The Androgen Insensitivity of XY Embryos

To exert its physical and behavioral effects, testosterone binds to specific intra-
cellular receptors, the androgen receptors. These receptors are proteins that,
when occupied by testosterone, partially change structure and interact with
DNA to stimulate or suppress the synthesis of various mRNAs, which are then
translated into proteins that influence the cellular function. Hundreds of muta-
tions of the androgen receptor (changes in the structure of the corresponding
gene) have been described. These mutations lead to production of a receptor
protein that is either completely unable to fix testosterone or else fixes it with a
lower affinity. The result is a complete or partial insensitivity to androgens
(androgen insensitivity syndrome, AIS) whereby testosterone and other andro-
gens are unable to induce their physical, functional, and behavioral effects. It is
estimated that one child in 10,000 is affected by AIS, which is also incorrectly
called testicular feminization syndrome. The androgen receptor gene is located
on chromosome X.
If a woman who carries the AIS syndrome in heterozygotic state (presence of
the mutation on one of the two X chromosomes) has an XY baby, there is one
chance in two that he will be affected by the syndrome. In this genetic male,
testes develop under the influence of the SRY gene and often produce normal
secretion of testosterone and of the anti-Müllerian hormone that induces regres-
sion of Müllerian ducts (no development of oviduct or of uterus). Testosterone
has no effect, however. Depending on the specific mutation (complete vs. partial
AIS), internal genital structures derived from Wolfian ducts (e.g., epididymis)
will or will not be present. More conspicuously, the genital tubercle and genital
folds continue their development in the absence of testosterone action, as in a
typical anhormonal state normally present in female embryos. The tubercle
remains small and forms a clitoris, and the genital folds do not merge and
instead form the lips of the vulva (no testicular scrotum). External genitalia of
these individuals will be typically female at birth (Figure 6.1).
AIS XY girls are sometimes diagnosed at birth because of the presence of a
blind vagina or because the testes are palpable under the skin in the genital
area, where they remained blocked because testicular descent cannot complete
in the absence of a scrotum. Many cases are not identified and are simply
accepted as normal girls.
In summary, although male-typical hormones are present in normal con-
centrations in AIS XY subjects, testosterone is completely unable to act and
Figure 6.1 Photograph of four XY women (XY genotype) affected by the syndrome of
androgen insensitivity (AIS). Presumably normal levels of testosterone were present in
these individuals during embryonic life but could not act due to the mutation of the
androgen receptor. Completely female morphology has therefore developed. (Google
Image Search, AIS syndrome).
masculinize embryos. Their phenotype (appearance) will be entirely female.

This extremely precise disorder affects only one specific protein, the androgen
receptor, which very clearly demonstrates the effects of testosterone during
development and effectively illustrates the crucial role of this steroid on the
development of sexual morphology. Also note that the lack of masculinization
of male embryos affected by the androgen insensitivity is accompanied by sig-
nificant behavioral changes that affect both gender identity and sexual orienta-
tion. I return to this in later chapters.
5α-Reductase Deficiency in the Young Boy

As explained in the section on sexual differentiation in animals, testosterone
must be converted into 5α-dihydrotestosterone (DHT) in the skin of the genital
area to promote growth of the male external sexual structures (development of
the genital tubercle into a penis and fusion of genital folds to form a scrotum).
DHT is an androgen more potent than testosterone and is the only one capable
of masculinizing external genital structures; the embryonic testosterone con-
centration is insufficient to achieve this result.
The conversion of testosterone into DHT is catalyzed (promoted) by a protein
enzyme called 5α-reductase, which is expressed at high levels in the genital tuber-
cle and genital folds. In the early 1970s, Imperato-McGinley and her colleagues
identified a mutation of the 5α-reductase enzyme that makes it functionally
unable to produce DHT (Imperato-McGinley, 1994; Imperato-McGinley & Zhu,
2002). This mutation in the gene controlling the production of 5α-reductase is
recessive. It is only visible when present in a homozygous state (on both chromo-
somes of the pair). It has been observed in island societies where circumstances
make inbreeding unusually common; it was first discovered in the Dominican
Republic, but cases have been found subsequently in Papua New Guinea, Europe,
and the United States.
Male embryos (XY chromosomes) affected by this mutation are born with
external genital structures only partially masculinized (see Figure 6.2). They
develop apparently normal testes that secrete the anti-Müllerian hormone that
induces regression of structures derived from Müllerian ducts (fallopian tubes
and uterus). Testosterone also promotes the development of internal sexual
structures typical of the male, but at the level of external genitalia, the genital
Figure 6.2 Genital structures not masculinized in an XY individual (male) suffering

from a deficiency in 5α-reductase. Figure taken from Mishra, Reddy, and Chaturvedi
(Bombay Hospial Journal 46/02, Case #23), reproduced with permission of P. Kapoor,
Editor of the Bombay Hospital Journal.
tubercle either doesn’t develop or develops very little and remains about the size
of a clitoris. There is no fusion of genital folds: we observe at birth the presence
of lips surrounding an opening to a blind vagina. There is therefore no scrotum,
and testes are located under the skin around the lips of the blind genital opening
or in the inguinal canal.
In adolescence, dramatic increases in the circulating levels of testosterone
induce a partial masculinization of the genital structures. The skin of the genital
folds’ wrinkles darkens so it looks more like a scrotum, the testicles will continue
their descent and will be found in the genital folds, and the genital tubercle will
develop somewhat to resemble a small penis. The opening of the urethra remains
on the base of the “penis,” a medical condition called hypospadias. Pubertal
male testosterone also affects various other physical aspects: It will increase the
muscles and lower the tone of the voice. From a superficial point of view, these
individuals look as though they have changed sex at least partially.
Male individuals affected by this change are essentially raised as girls and
adopt a female sexual identity during childhood. In communities that are famil-
iar with this syndrome, affected boys are clearly identified at birth, however,
and the local language has a specific term to designate them (Guevedoche in the
Dominican Republic, which in Spanish means “eggs,” i.e., testes, and “twelve
years”; or Kwolu-aatmwol in Papua, which means “female thing transforming
into male thing”). One can imagine that those affected are to some extent con-
sidered intersex individuals. Intersex morphology at birth and partial change of
sex morphology at puberty are also associated with the changes in identity and
sexual role that I shall discuss in following sections.
Congenital Adrenal Hyperplasia in the Female Embryo

Congenital hyperplasia (excessive development) of the adrenal glands (CAH) is a
genetic disorder that affects one of the enzymes involved in the synthesis of cor-
ticosteroid hormones (cortisol) from cholesterol (often 17α- or 21-hydroxylase).
This synthesis of cortisol occurs in the adrenal glands. Cortisol is an important
hormone involved in immune responses, carbohydrate metabolism, and response
to stress. It is therefore not directly related to reproduction, but if its production
is interrupted by genetic defect (e.g., CAH), one observes in response a hyperse-
cretion of androgens (compounds similar to testosterone) in the affected embryos
(Hines, 2003; MacLaughlin & Donahoe, 2004).
CAH syndrome is an autosomal recessive disease (carried by non–sex chro-
mosomes). If both parents carry one defective copy of the gene, each of their
children, regardless of sex, will have one chance out of four to have two copies
of the mutated gene and be affected. It is estimated that one child in 16,000 is
affected by this syndrome. If the affected fetus is male, additional androgens
secreted by the adrenal gland will have little influence because they are simply
added to the large amounts of androgens produced by the testicles. In addition
to the metabolic problems associated with the absence of cortisol, we observe in
affected female individuals a more or less profound masculinization of the
external genital structures. In the most severely affected individuals, the genital
tubercle has developed into a fully formed penis with a size almost equal to that
seen in male babies. In addition there is a more or less complete fusion of the
genital lips that form a scrotum (Figure 6.3). There are, of course, no testicles.
The sex of these children is female. SRY is absent and consequently ovaries are
formed. The anti-Müllerian hormone was not produced and the Wolffian ducts
have developed into a uterus and oviducts. The internal reproductive organs are
those of normal girls.
This more or less complete masculinization of external genital structures in
CAH girls is usually detected at birth. These children will be treated during
their entire life by administration of glucocorticoids. This treatment, by feed-
back on the activity of the pituitary, also suppresses the pathological production
of androgens by the adrenal glands. Masculinized external genital structures
are often “corrected” surgically, removing part of the penis to reproduce the
clitoris and incising the genital folds to reopen the vaginal opening. These chil-
dren are then raised as girls as far as we can know. From a scientific point of
Figure 6.3 Genital structures in a masculinized XX individual (female) suffering from

adrenal hyperplasia (CAH). Image from Legros J.J. et al. 1974. Revue Médicale de Liège
29, February 1974, pp. 73–80; reproduced with permission.
view, CAH girls allow one to assess, as accurately as possible in humans, the
consequences of embryonic androgenization independent of postnatal endo-
crine changes (which are medically corrected). This helps in assessing the con-
sequences of embryonic androgenization on the behavior of girls who are
normally raised as such. [See the work of Melissa Hines, including (Hines &
Kaufman, 1994; Brown et al., 2002b; Hines et al., 2003; Hines et al., 2004).]
I shall return to this topic.
The Treatment of High-Risk Pregnancies by DES or Progestagen

There are also populations of girls whose mothers, during pregnancy, were
exposed for medical reasons (treatment of risk of spontaneous abortion) to syn-
thetic steroids such as diethylstilbestrol (DES) or acetate of medroxyprogesterone.
It was realized afterward that some of these treatments were associated with mas-
culinization of the reproductive function and behavior. These treatments were
stopped as soon as doctors became aware of their consequences. People unfortu-
nately exposed to this masculinizing action gave us another important source of
information about the long-term behavioral effects of embryonic steroids.
SEXUAL DIFFERENTIATION OF BEHAVIOR

The physical effects of testosterone during embryonic development that I have
just described leave, in my opinion, no doubt about the conclusion that in the
human species, mechanisms of endocrine control of morphological sexual dif-
ferentiation are identical in all respects to those reported in other mammals.
Knowing that in rodents and monkeys these early physical effects of testoster-
one on sexual differentiation are accompanied by the sexual differentiation of
many aspects of behavior, it is natural to ask whether human behaviors are also
irreversibly affected by embryonic exposure to testosterone in young males (and
its absence in the female embryo). The answer to this question is much more
difficult because, although the morphology develops under the influence of
hormones quite independently of external conditions, behavior is strongly
influenced by interactions with parents, siblings, and all social relationships of
young children. Influences that are not the same for babies of either sex (Hefez,
2007) will deeply affect the development of behavior. It is therefore very difficult
to find during the child’s development (and even more in the adult’s) clear traces
of a possible prenatal endocrine influence on the expression of behavioral traits.
Even though absolute proof of the existence of such effects is lacking, some
observations and clinical studies nonetheless suggest their presence.
A number of behavioral differences exist between men and women. Many of
these differences have a rather limited magnitude (see Chapter 5), however,
and seem to result mainly from the differential education of boys and girls.
Others have a much greater magnitude and relate in particular to sexual orien-
tation and sexual identity, two behavioral characteristics for which 90–99% of
the population shows a marked sex difference closely correlated to the biologi-
cal sex. The second part of this book is mainly devoted to the analysis of various
arguments that strongly suggest that sexual orientation is determined by bio-
logical factors acting predominantly during the embryonic period.
Before addressing these arguments, it is important to ask whether steroids
have organizing effects on behavior in humans and animals. In other words, does
the hormonal milieu, in addition to determining genital form, influence the
organization of behavior in a more masculine or feminine way? Given the impos-
sibility for ethical reasons of implementing any type of experimentation on the
subject, it is only through the analysis of clinical cases that we can try to answer
this question. Moreover, because behavior is strongly influenced by education
and social influences, we can hope to find a trace of an interpretable embryonic
hormonal influence only in clinical cases where this hormonal influence is acting
against the standards imposed by education.
For example, individuals affected by the syndrome of androgen insensitivity
(AIS) teach us very little about the possible role of hormones in the differen-
tiation of behavior. These individuals, with male sex chromosome (XY), have
testes that secrete testosterone, but the steroid cannot act due to a mutation in
the androgen receptor. Therefore they are born with female genitalia but are also
raised as girls. They thus assume as adults the sexual orientation and gender
identity of women, but it is impossible to know whether this is due to their edu-
cation as a girl or the lack of effect of testosterone in the male embryo. These
cases do demonstrate that the sex chromosomes are not very important as such:
XY individuals may well develop a female phenotype. It is the hormonal action
in the embryo that is critical for determining the phenotype at birth. Whether
female behavioral traits observed in adulthood are the direct consequence of this
embryonic hormonal action or a byproduct of the education of these subjects
raised as girls remains impossible to ascertain.
Girls with the syndrome of Congenital Adrenal Hyperplasia (CAH) are, from
this point of view, much more interesting and have therefore been the subject of
many behavioral studies. Females with CAH were indeed exposed in utero to a
hormonal milieu typical of the male (presence of large quantities of androgens of
adrenal origin) but are generally raised as girls after postnatal surgical correction
of their external genitals. If embryonic androgens affect the differentiation of
sexual behavior, it could be expected that these girls would be to some extent
behaviorally masculinized. And that is indeed what has been observed in several
studies (Hines, 2003; Hines et al., 2003; Hines et al., 2004). The author of these
studies has summarized the results in a book for the general public (Hines, 2004).
A classic behavioral difference between boys and girls concerns the type of
games that they choose freely in a controlled test situation. If we introduce indi-
vidual boys or girls in a room containing various types of toys, the girls spontane-
ously chose dolls while boys prefer to take cars or trains. Spontaneous games of
boys are also more violent than in girls. One might think that this sex difference
in children’s behavior is entirely cultural and driven solely by the differential treat-
ment of children by parents and society. In fact, it is not. It was demonstrated that
these spontaneous toy choices are clearly masculinized in CAH in girls who have
been exposed to abnormally high androgen levels during their life in utero but
not during the postnatal period (thanks to medical treatment). In addition, play
of CAH girls is more violent than in control girls not affected by prenatal andro-
genization (Figure 6.4) (Berenbaum & Snyder, 1995; Nordenstrom et al., 2002).
An interesting finding is that an offer of the same selection of toys to young
vervet monkeys (Cercopithecus aethiops sabaeus) produces the same choice dif-
ferences between sexes as found in humans: males are more interested in cars
or balls whereas females are more interested in dolls (Figure 6.5a) (Alexander
& Hines, 2002; Hassett et al., 2008).
600 Boy toys 600 Girl toys

Time spent playing with different
toys (Seconds)
400 400
200 200
0
Ctrl CAH Ctrl CAH Ctrl CAH Ctrl CAH
Girls Boys Girls Boys
Figure 6.4 The type of toy used predominantly by boys and girls is sexually
differentiated and influenced by the embryonic hormonal environment. Boys spend
more time with toys such as cars or buildings tools (Boys Games), and girls with dolls,
doll clothes, and kitchen utensils (Girls Toys). These preferences are significantly
masculinized (increased time spent with the boys’ toys) and defeminized (decrease of the
time spent with the girls’ toys) in CAH girls. The boys already have a lot of testosterone
and their preference is not affected by additional testosterone related to CAH status. The
results of the four types of subjects are represented by columns in which the intensity of
gray is proportional to the concentration of testosterone supposed to have been present
during embryonic life (according to Berenbaum & Snyder, 1995).
B Rhesus monkeys Human

“Boy” toys Boy toys
“Girl” toys Girl toys
Time spent playing (Seconds)
12
400
200
4
0 0
Males Females Boys Girls
Sex of experimental subjects
Figure 6.5 Young monkeys show the same preferences for different types of toys as
human children. A. Photographs showing the play of young vervet monkeys with a doll
(left) or a car (right). (Reproduced from Alexander & Hines, 2002, with permission.)
B. Sex difference in time spent playing with boy or girl toys in rhesus monkeys or in
humans (redrawn from Hassett et al., 2008; Berenbaum & Hines, 1992).
Recent experiments have now confirmed this result in another species of

monkey, the rhesus (Macaca mulatta). Young males received toys classically
considered in humans to be preferred by boys or girls: wheeled toys or stuffed
animals, respectively. Toys with wheels were used far more frequently by rhesus
males than by females, while young monkeys of both sexes played more or less
equally with stuffed animals (Figure 6.5.b) (Alexander & Hines, 2002; Hassett
et al., 2008). The comparison with the human data is striking. We also note that
in this study, in the rhesus as in humans, the sex difference of use of toys for
boys (cars) is much larger than the difference in the use of toys for girls (dolls,
stuffed animals).
One might ask why monkeys differentially use toys that are obviously not of
direct significance for them (cars or trains). The answer to this question should
probably be sought in the properties of toys studied. Toys with wheels can, for
example, allow games that are more active and could be preferred by males,
given their propensity to engage in these types of games [see Williams & Pleil
(2008) for a more detailed discussion on this subject]. These data suggest that
sexually differentiated preferences for characteristics of objects (color, shape,
movement) emerged very early in the lineage leading to human evolution. They
may have been selected by evolutionary pressures related to the differential
roles played by men and women and still be operant in the determinism of the
choice of toys shown by small children.
The notion that these preferences are imposed only by society is contra-
dicted by these observations. First, the monkeys had never been able to interact
with the toys before the test, so it is impossible that education played a role.
In addition, the CAH girls, although raised as girls, show preferences more or
less masculinized for games that seem consistent with the hormonal milieu to
which they were exposed in utero. The preferences observed are, instead, related
to the general nature of the objects (e.g., preferences of boys and male monkeys
for moving objects with wheels and more active games) (Hines & Alexander,
2008; Williams & Pleil, 2008) At the proximal level, these preferences appear to
be controlled, at least in part, by the degree of in utero exposure to testosterone,
as suggested by the study of CAH girls.
A study in Japan suggests, moreover, that there are sex differences in the types
of drawings made voluntarily by children and that these drawings are masculin-
ized in females affected by the CAH syndrome (Iijima et al., 2001). If we provide
young children, five to six years old, with a set of colored pencils, boys on aver-
age tend to draw mechanical or moving objects with dark or cold colors (blue,
green), while girls prefer drawing human beings (especially girls and women),
flowers, or butterflies with warm colors (yellow, red). The drawings of boys are
also more often drawings of objects seen from above, while girls tend to align
objects at ground level. An analysis of drawings made by girls affected by the
CAH syndrome has shown that their drawings did not have the usual female
characteristics but were heavily masculinized and included all the features nor-
mally observed in the drawings of boys. Again, these observations strongly sug-
gest that the embryonic hormonal environment plays a role in organizing this
aspect of behavior. It should be noted that during early childhood (from more
or less a year of age until the approach of puberty), levels of circulating testos-
terone are the same in boys and girls. Any influence of testosterone thus can
only be a long-term influence of hormonal differences that were experienced

during embryonic life.
The potential influence of the embryonic environment (masculinization) is
contra to the type of education received (as a girl). It is thus logical to assume
that the observed behavioral masculinization in CAH girls is due to their pre-
natal exposure to abnormally high levels of androgens. In most cases, these girls
are not completely masculine, and this aspect of their personality undoubtedly
reflects the postnatal influence of parental and social environment. The influ-
ence of education cannot be denied in humans (it is already clear in animals),
but these clinical cases suggest very strongly that the influences on the behavior
of the embryonic hormonal milieu that have been described in animals also
exist in humans. Based on the principle of the unity of life and evolution of
animal lineages, why would be otherwise?
Finally, the progress in psychological methods used to analyze the behavior
of young children now allows us to test their response to various stimuli from
an early age on, even during their first days of life, before any influence of edu-
cation can take place. These studies show that very interesting sex differences in
behavior are already present at that stage. In one study, British researchers from
the laboratory of Simon Baron-Cohen filmed more than 100 babies, only one
day old, while they were in the clinic being subjected to psychological tests.
During these tests, the researchers showed an image of one of the researchers
(JC), at a distance of 20 cm, in which she adopted a positive emotional expres-
sion while remaining silent, or the image of a moving object (ball suspended
from a 1 m stick) matched precisely for five aspects of the figure: color, size,
shape, contrast, and tridimensional aspect (a small ball of 3 cm was attached in
the middle of a mobile object at the same place as the nose). The mobile object
was moved mechanically over the child’s head at the same distance of 20 cm.
All elements of the face of JC were present on the mobile, but completely mixed
so that the presence of a human figure was no longer recognizable (Figure 6.6).
To preserve the neutrality of the experimenters, researchers were not notified
of the sex of the baby before the end of the tests and their analysis.
These studies have shown, quite surprisingly, that there is a significant differ-
ence between boys and girls in their response to both visual stimuli (Connellan
et al., 2000). The girls looked longer at the human face, but the boys looked
more at the mobile. This difference at birth recalls another difference observed
throughout the duration of life: on average, women interact more socially than
men via social smiles and by maintaining direct eye contact. Although aspects
of the methodology used in these studies have been criticized and its potential
impact on adult sex differences has been questioned (Barres, 2010; Fine, 2010),
the fact that such a sex difference has been observed at birth suggests the
existence of a biological difference related to prenatal factors. In addition,
Small ball
attached
here
Face of researcher (JC) Mobile object

Figure 6.6 Photograph of the stimuli used in the study of the gaze of one-day-old babies
(reprinted from Connellan et al., 2000 with permission).
other studies of the group of Simon Baron-Cohen have shown that the amount
of eye contact with parents of 12-month-old children was inversely correlated
to the concentration of embryonic testosterone measured by amniocentesis
between 14 and 21 weeks of gestation (Lutchmaya et al., 2002). There are there-
fore good reasons to believe that embryonic testosterone could be one of the
factors responsible for individual and sex differences that affect certain aspects
of social relationships (refer to the excellent popular books by Simon Baron-
Cohen on the subject (Baron-Cohen, 2004; Baron-Cohen, 2006)).
THE MULTIPLE PEAKS OF TESTOSTERONE DURING DEVELOPMENT:

POSSIBLE CONSEQUENCES FOR PHYSICAL AND
BEHAVIORAL DIFFERENTIATION
An extremely important fact to mention is that during embryonic development
of a normal male embryo and its postnatal life, testosterone is not secreted in
high concentrations on a continuous basis (Figure 6.7).
In male embryos, an early peak of concentration in early gestation is respon-
sible for morphological differentiation, followed by a period of slightly lower
concentration that covers the second and third trimesters of pregnancy. These
peaks do not occur in the female embryo. After a large peak immediately after
birth, the secretion of testosterone is very low throughout infancy and will show
a significant increase at puberty. At this point, it will trigger the final maturation
of genital organs, spermatogenesis, and development of pubic hair. This period
also corresponds to the onset of behavioral sexual activity.
It is firmly established that the differentiation of the genital structures of the
human embryo takes place during the first two months of gestation, whereas
the brain develops and differentiates during the second half of pregnancy. Thus,
if an event were to alter the endocrine profile between these two periods, it is
BIRTH PUBERTY
Embryonic life Childhood Adult life Aged men
Testicular Fetal Neonatal Endocrine Final Aging

differentiation testis activation rest maturation
Müller ducts Penile Development/maintenance Decreased libido

regression growth Sexual -Sex organs
Sexual dysfunction
? identity -Secondary sex characters
Male -Spermatogenesis
sexual Sexual -Libido, motivation
differentiation ? orientation
Figure 6.7 Schematic representation of the fluctuations in blood testosterone in men.
conceivable that the newborn may have a genital form (determined early in
gestation) that does not correspond completely with some traits of his person-
ality, such as his gender identity or sexual orientation, which are determined
much later.
It therefore seems possible that an individual who has experienced “normal”
conditions during the first part of his or her embryonic development has
genital structures of one sex but that, subsequently, hormonal changes occur
and may change his or her gender identity (transsexual) or sexual orientation
(homosexual). These characteristics of an individual are firmly established in
early childhood. The development of sexual orientation, however, does not
become evident until later in life (puberty). Various clinical and epidemiologi-
cal data suggest that sexual differentiation of these two characteristics could
begin during the second half of embryonic life (Swaab, 2007).
The presence or absence of testosterone at different critical periods during
development is not the only factor that can potentially generate discordance
between the genetic and gonadal sex of an individual, on the one hand, and
some of his or her behavioral characteristics, on another hand. To exert its
actions, testosterone must be transformed into active metabolites (estradiol,
DHT) that must bind to proper intracellular receptors and activate a cascade of
biochemical events that will eventually lead to the biological response. Localized
changes in any aspect of this reaction chain can therefore produce discordance
between the actions of the steroids at two loci. The profile of plasma testoster-
one concentrations could thus be perfectly normal in a male subject, but local
changes in testosterone action in the brain could prevent the full spectrum of
steroid action in this structure, resulting in discordant behavioral features.

I shall return to these ideas in the sections devoted specifically to homosexu-
ality (see Chapter 8).
SEXUAL DIFFERENTIATION OF THE BRAIN

A number of differences between the sexes have been identified in the human
brain (Chapter 5). These differences relate to the size of a sexually dimorphic
nucleus of preoptic area (SDN-POA) (Swaab & Fliers, 1985; Allen et al., 1989;
LeVay, 1991), the bed nucleus of the stria terminalis (BST) (Zhou et al., 1995;
Chung et al., 2002), and thick bundles of fibers connecting the left and right
sides of the brain (de Lacoste-Utamsing & Holloway, 1982) (the anterior com-
missure and the corpus callosum) [see (Dubb et al., 2003; Tuncer et al., 2005;
Ozdemir et al., 2007) for a potential challenge of this difference]. Difficulties in
obtaining human brains for histological analysis, combined with the relative
scarcity of endocrine diseases disrupting sexual differentiation, explain why,
to date, no study has specifically analyzed the contribution of sex steroids in
embryonic development on sex differences observed in the nervous system.
Two studies, however, have evaluated the postnatal ontogenesis of sex differ-
ences that affect the volume of the sexually dimorphic nucleus of preoptic area
and the bed nucleus of the stria terminalis and have shown that these structures
are not fully differentiated at birth. For example, the SDN-POA, as described by
Swaab and collaborators, at birth only contains 20% of the cells that it contains
at two to four years of age. This number of cells is believed to increase quickly
and in the same way in boys and girls between birth and two to four years. Only
after this period does a decrease in the number of neurons specifically affecting
girls create a sex difference in volume (Swaab & Hofman, 1988). These critical
events occur, therefore, in early childhood, when circulating levels of testoster-
one are the same in both sexes. Similarly, it was shown that the sex difference in
volume affecting the bed nucleus of the stria terminalis is not present in children
and is expressed only in adults (Chung et al., 2002).
Recall that in rats, the larger volume of the sexually dimorphic nucleus of the
preoptic area is determined exclusively by the prenatal and perinatal effects of
testosterone (Jacobson et al., 1981; Arnold & Gorski, 1984). This difference is
determined entirely prenatally in sheep. It is important to note that, in rats as
well, volumetric differences of the SDN are not necessarily present at birth
(Jacobson et al., 1980). They continue to develop during the first weeks of life
under the influence of increased neuronal death in females (Davis et al., 1995).
It is likely that early hormonal action sets in motion a series of mechanisms
that will continue to act during the postnatal prepubescent life in the absence
of significant concentrations of testosterone. This situation is perhaps not so
different from what happens in humans, and we have as yet no reason to believe
that the difference in volume of the SDN-POA present in humans develops by
mechanisms different from those described in rats or sheep. But we do not have
any argument beyond evolutionary continuity to confirm the identity of these
mechanisms.
Some authors (e.g. Vidal, 2007, p. 13) argue that the sexually dimorphic
nucleus of the hypothalamus would be unable to control sexual orientation.
C. Vidal announces a size of 50 microns of this sexually dimorphic nucleus in
women and homosexuals and 100 microns in heterosexual men (cubic microns,
I assume). Although small (between 0.05 mm3 in homosexuals and 0.10 mm3
in heterosexuals), the human SDN contains a significant number of neurons,
approximately 1,800. It is generally accepted that each neuron in the human
brain on average makes 100 to 1,000 connections with other neurons. This means
that the SDN is directly connected to a number of neurons ranging between
18,000 and 180,000, themselves part of a network that is branching in an expo-
nential manner. Such a network is capable, I believe, of managing complex infor-
mation. Excluding a priori that it could be a significant part of the mechanisms
that control sexual orientation is not based on any rational argument, even if the
author of this book was trained as a neurobiologist and director of research at the
Pasteur Institute. The denial of the role of the brain is here based on a priori
excluding the idea that there may be significant differences between the male
and female brains. If the ultimate goal of this claim is laudable (fight against sex
discrimination), the means are not scientifically acceptable.
7
The Effects of Sex Steroids in Humans

Activating Effects
FORMS AND FUNCTIONS

There are numerous effects of sex steroids (androgens, estrogens, and progesta-
gens) during the postnatal life in man and woman. These effects are impossible
to ignore and widely known to the general public. At puberty, because of the
significant increase in circulating testosterone concentrations in young males,
the penis increases in size and the testes grow larger and start producing sperm.
In parallel, muscles develop, pubic and armpit hair appears, and the tone of
voice drops (deeper voice) due to the thickening of the vocal cords. All these
effects are, entirely or in part, caused by testosterone. If testosterone is not pres-
ent due to a failure of testicular function or following an accidental or voluntary
castration (e.g., Italian castrati, eunuchs for harems), none of this maturation
related to puberty will occur.
From a theoretical point of view, it should be noted that many of these effects
are irreversible. Some researchers thus consider puberty to be a second phase of
organization of physical form and function that complements the sexual dif-
ferentiation that occurs during the months before and after birth (Sisk & Zehr,
2005; Ahmed et al., 2008).
In young females, puberty is marked by the onset of ovarian activity, which,
for the next 30 to 40 years, produces estrogen and progesterone in a cyclical
fashion in association with the cycle of ovulation and menstruation. Puberty

also brings the appearance of pubic hair and (later) armpit hair, a development
of the lips of the vulva, and a deepening and thickening of the walls of the
vagina. It is also the time when breasts are developing that in a few years will
reach their adult size. Ovarian estrogens are largely responsible for these changes
in the girl during puberty, but androgens secreted by the adrenal glands also
play a particular role in the development of the hair. If for any reason ovarian
activity and menstruation do not appear, the changes of form just described
will not be observed.
The role of androgens in the development of hair in women (and men, of
course) is also well established by the fact that the alterations of adrenal activity
associated with increased production of androgens always result in an increased
hairiness affecting the same area, the face (“bearded woman”). During meno-
pause, which corresponds to the cessation of ovarian activity, there is also often
a slight increase in the secretion of androgens by the adrenal glands, often
accompanied by a slight development of facial hair (beard and mustache).
BEHAVIOR
If the reality of the effects of sex steroids on the morphology (including genital)
and the physiology of reproduction does not appear to be contested, there is in
contrast a widespread reluctance to accept the idea that sex steroids could play
a role in control of behavior in general and sexual behavior in particular. This
resistance is particularly important in Latin societies. For instance, in Men and
Women: Do We Have the Same Brain? Catherine Vidal wrote, “In animals, the
action of hormones on the brain induces mating behavior and mating periods
are associated with ovulation of the female. Sexuality and reproduction go hand
in hand. In contrast the human being escapes this determinism” (Vidal, 2007).
There is, in my opinion, no statement more false than her last sentence.
Human sexual behavior is very complex, and its control involves multi-
ple cognitive aspects (experience, perception of social expectations of the
environment, individual preferences, social conventions, varied religious and
philosophical influences, etc.), the importance of which relates to the major
development of the cerebral cortex that characterizes our species. Accordingly,
the mere presence of testosterone in the blood and brain of a man and of estra-
diol in a woman will not automatically lead them to have sex (mating) within
minutes of their first meeting. It is also not the case in animals. The specific
studies conducted in particular in rats and mice are showing the influence of
multiple variables on sexual behavior in these species, such as prior experience,
learning, individual preferences, and the effect of an audience on the social
behavior achieved. Only in the study of instinctive reactions in invertebrates
such as insects do we find to some degree the automatic causal reactions
between hormones and behavior that is erroneously considered as being typical

of animal sexuality.
The decision to engage in sex with a sexual partner in the human species
does not depend solely on circulating testosterone levels of men and estradiol
in women, but it will certainly be influenced by these hormonal factors in
a more or less direct manner (see below). Sexual motivation in humans remains
under hormonal control even if it is partially emancipated, as is partly the case
in the animal lineage in parallel with the development of the brain and espe-
cially of the cortex. If sexual motivation were entirely the result of learning
and a social construction, we should then wonder why no society has ever man-
aged to put its expectations in line with the behavior of individuals. Sex outside
marriage is discouraged or prohibited, for example, by many human societies.
In some cases, adultery is still punishable by death, sometimes by methods that
could be very discouraging, such as lapidation (stoning). The fact is, however,
that sex outside marriage remains present even in these societies.
I believe that sexual motivation in humans and animals is essentially the
result of the action of sex steroids on the brain and in particular on preoptic,
hypothalamic, and limbic (amygdala) regions. Philosophical or social contin-
gencies have an important role in determining whether the individual will
adopt a behavior in agreement with this motivation. Religious or social prohibi-
tions can, for example, prevent an individual from having an extra-marital rela-
tionship. Here cortical activity takes precedence over the impulse originating
in hypothalamic or limbic areas, but the motivation remains. This cannot be
emphasized too strongly! Innumerable literary works examine the conflict
between the sex drive and social prohibitions consciously managed by the
cortex. This motivation is also modulated by many other factors (e.g., lack of
sexual desire in a situation of stress), but the main determinant remains hor-
mones in humans and animals. Many arguments support this thesis.
EVOLUTIONARY ARGUMENT
The brain regions that are known to control sexual behavior in rodents and
primates have changed very little from a physical or neurochemical point of
view between mammals and humans. Humans possess almost all the nuclei
(clusters of neurons) of the preoptic area, hypothalamus, and limbic systems
that control sexual behavior in the rat brain. In addition to this anatomical
consistency, the same neurotransmitters as in animals and steroid hormone
receptors (androgen and estrogen receptors) that have been characterized in
other vertebrates are also found in these regions of the human brain. The distri-
bution of these receptors is very consistent from fish to mammals (Kelley &
Pfaff, 1978; Morrell & Pfaff, 1978), and man is no exception (Kruijver et al.,
2001; Kruijver et al., 2002; 2003). The cellular and neurochemical “machinery”
underlying the expression of sexual behavior in vertebrates is therefore intact in

humans. Its persistence indicates that there are evolutionary pressures that
maintain it in place and function. If, as I claim here, these neural circuits con-
trol sexual motivation under the influence of ovarian and testicular steroids,
the nature of this selective pressure becomes immediately obvious.
TEMPORAL AND INDIVIDUAL CORRELATIONS

The sex steroids are not present in constant quantities throughout life. If they
play a significant role in the control of sexual activity, one would expect a posi-
tive correlation between changes in the time of the concentrations of steroids
and frequency of sexual activity. This is indeed what has been observed in many
studies of both men and women.
Measurement of the blood levels of testosterone of a large number of men
during their lifetime yields a series of reproducible values that are positioned
on a generally inverted U-curve (see Chapter 6, Figure 6.7). The ascending
phase of this curve corresponds precisely to the puberty and onset of sexual
activity. The peak of this curve is between 20 and 60 years and corresponds to
the maximum of sexual activity in humans. Then begins a gradual decrease of
circulating levels of steroid associated with the well-documented decline in
sexual activity with age (Vermeulen et al., 1972). There is wide individual varia-
tion around this mean curve, and it would be ridiculous to suggest that changes
in concentration of steroid alone explain the distribution of sexual activity
during the life of man. The beginning and end of this activity, however, are
linked quite closely to changes in blood concentration.
The relationship between hormones and behavior can also be greatly
strengthened by studying individual variations in periods of instability. Various
researchers have analyzed the relationship between increased blood levels of
testosterone at the time of puberty and the appearance of male sexual behavior.
Since a continuous and accurate record of this behavior is impossible, research-
ers have relied on more indirect, rough measures, such as the age of first sexual
relationship or the likelihood of having had sex at a given age. Even with mea-
sures this imprecise, a very strong connection appears between hormone and
behavior. Figure 7.1 illustrates the relationship between salivary testosterone
(salivary concentrations are a fairly accurate reflection of blood levels), expressed
as deviations from the average values (mean plus or minus one, two or three
standard deviations, a measure of variability), and the probability at a given age
that a boy has already had sex. There is an excellent correlation between these
two measures, suggesting that testosterone levels higher than average are asso-
ciated with greater sexual activity (Halpern et al., 1998).
6
Coital risk
0
45.2 120.4 195.6 270.7 345.9
−1SD Mean +1SD +2SD +3SD
Testosterone in saliva (pmole/liter)
Figure 7.1 Relationship between the concentration of testosterone in saliva and
the probability of occurrence of complete sexual (coital) behavior in adolescents
(age 12–13 years at baseline). Saliva samples were collected at regular intervals over
a period of two years during which the subjects filled out questionnaires about their
weekly sexual activity. The average risk of coital sexual behavior is adjusted to 1 for
average values of testosterone (120.4 pmol/l). It increases in an exponential manner
when concentrations of testosterone increased by 1, 2, or 3 standard deviations (SD)
above average (redrawn from Halpern et al. 1998).
This type of relationship has been observed both in transversal studies of

a large number of individuals of the same age (Udry et al., 1985) and in longi-
tudinal studies that followed a smaller number of subjects during their develop-
ment (Halpern et al., 1998). These relationships also appear to be independent
of the societies and cultures in which the boys live. Similar correlations between
testosterone levels and age at first sexual relationship have recently been identi-
fied in a population in Zimbabwe (Campbell et al., 2005).
In women, a similar relationship between the individual values of blood con-
centrations of testosterone and the onset of sexual activity has been identified
by the same methods (Halpern et al., 1997). Testosterone appears to be involved
also in the activation of sexual motivation in women; this hormone is pre-
scribed in combination with estrogen to increase sexual desire in women whose
ovaries have been removed for medical reasons (Figure 7.2). Controlled studies
demonstrated an increase in sexual desire, sexual fantasies, and sexual arousal
in response to treatment with androgens combined with estrogens. In contrast,
estrogens alone had no effect (Sherwin & Gelfand, 1987).
5 Injection Injection
6
Sexual excitation
Sexual desire
4 5
E+A E+A
4
3
3
2 E E
2
1 1
Ctrl Ctrl
0 0
0 1 2 3 4 0 1 2 3 4
Weeks after injection
Figure 7.2 Evolution over time of various aspects of sexuality in ovariectomized women
treated with estrogen alone (E) or estrogen combined with androgens (E+A), or
receiving a control (Ctrl) treatment (redrawn from Sherwin & Gelfand 1987).
Also, since ovarian activity is cyclical in women, it is possible to analyze the

relationship between hormonal changes and spontaneous behavior. These stud-
ies have considered various aspects of sexual activity, including motivation
(desire) but also sexual attractiveness. It must be admitted that many results
in this context have proved to be difficult to reproduce or not reproducible.
Some studies have reported the existence of a peak of sexual activity in women
at the time of ovulation [for details see LeVay & Valente (2006) p. 142]. This
peak could represent a remnant of estrus in animals. In many species, sexual
activity of the female is indeed restricted to the periovulatory period. Peaks
of activity, however, have also been reported in humans at other times of the
cycle (premenstrual period) and have no obvious functional or evolutionary
interpretation. These periovulatory peaks, when observed, are probably the
result of hormonal changes associated with the cycle, because some studies
indicate that these behavioral changes are not observed in women who undergo
hormonal contraception that makes their hormonal environment constant
during most of the ovarian cycle.
The reasons underlying the variability of the results of these studies are not
always identified but include the possibility of inadequate characterization of
the hormonal status of women studied (e.g., a single dose of blood unrepresen-
tative of the average hormonal state because of rapid fluctuations of hormone
levels, hormonal status inferred from the stage of the menstrual cycle but not
confirmed by direct measurements of plasma levels), the obscuring of the cor-
relation of hormone-influenced behavior by nonhormonal factors (daily prob-
lems interfering with sexual desire, unrecognized medical problems, etc.), or
vague or biased recording of actual sexual activity.
Even if better-controlled studies confirm the presence of a periovulatory peak
of sexual activity among women, the interpretation of this peak will remain dif-
ficult because it could result equally from either an increase in the sexual moti-
vation of the women or an increase in their attractiveness. Studies have attempted
to isolate these two interpretations by documenting (with questionnaires),

in stable couples, the frequency of sexual activity that was initiated by the
woman (and supposedly related to her motivation) or initiated by the man
(reflecting hormone-related changes in the woman’s attractiveness), but it
appears that the variability in these studies makes it impossible to draw repro-
ducible conclusions.
In a recent study, U.S. researchers adopted a very creative protocol to quan-
tify potential changes during the menstrual cycle of attractiveness of a group
of women in a situation that would minimize uncontrolled extraneous vari-
ables (Miller et al., 2007). They recorded the tips earned by seminude (topless)
women dancing in nightclubs (“lap dancers”) and correlated the tips with the
stage of the women’s menstrual cycle. These women dance on a podium several
times each evening to an audience, but under these conditions they will collect
only small tips. During the remainder of the evening, they walk in the room
and offer to “dance” on the laps of customers during a period of about 3 minutes.
During this activity they usually receive a larger tip, which often varies between
$10 and $20 U.S. but is not fixed by any rule. These “dances” typically involve
intensive rhythmic contacts between the pelvic region of women and the penis
of the man, who is dressed and cannot touch the dancer with his hands. They
represent the most intense form of paid sexual interaction that is permitted
by U.S. law. Customers are able to assess the attractiveness of the dancers
through very intimate verbal, visual, tactile, and olfactory sexual contacts. This
type of situation is likely to more accurately evaluate attractiveness than in situ-
ations that were used previously and were based on the evaluation of photos or
the voice of individuals tested.
The researchers asked the dancers to record the total amount of their earn-
ings in the night club and their working time each day for 60 days. They were
also to indicate the dates of their menstruations. Data from 18 women with
regular menstrual cycles (28–30 days) were finally analyzed. Seven of them were
using hormonal contraception, the other 11 did not use contraception and had
not used it during the three months preceding the study. As indicated in Figure
7.3, gratuities earned by women during their fertile period (periovulatory) were
twice as high as what they earned in luteal phase (the part of the menstrual cycle
following ovulation) or during menstruation. These variations in gains were sig-
nificant and were not observed in women using the contraceptive pill, which
equalizes the circulating hormone concentrations during most of the cycle.
These data support the idea that sexual attractiveness in women varies during
the ovarian cycle and that these variations in attractiveness are due to fluctua-
tions associated with the hormonal cycle. Further research should be under-
taken to test the reproducibility of these findings and identify the nature of
signals that modulate attractiveness (verbal, visual, or olfactory).
400
Normal
300 cycles
Dollars earned
200
Pill
users
100
0
Menstrual Fertile Luteal
Cycle phase
Figure 7.3 Variations during the menstrual cycle of female attractiveness assessed
indirectly by the dollar amount of tips obtained by seminude dancers in cabaret.
A peak gain is observed in ovulatory phase (fertile) compared to menstrual and luteal
phases. This temporal variation is not present in women using an oral contraceptive
and who therefore have a constant level of estrogen and progesterone during the cycle
(redrawn from Miller et al. 2007).
THE ANALYSIS AND TREATMENT OF CLINICAL CASES

Numerous clinical studies have been conducted on different types of patients
suffering from various sexual problems. These studies reinforce the notion that
male sexual activity is controlled at least partly by testosterone. For example,
there are patients with pathologically low development of the testes, which
therefore produce abnormally low testosterone concentrations (hypogonad-
ism). Many of these subjects have a low sexual activity associated with a low
frequency of nocturnal erections, episodes of masturbation, and orgasms. Julian
Davidson showed that treatment of these subjects by a single injection with a
form of testosterone that is released gradually into the blood (testosterone
enanthate) increases all the measures of sexual activity included in this study,
proportional to the dosage, while restoring more normal blood levels of testos-
terone (Davidson et al., 1979).
Many similar studies have since confirmed this conclusion, including a blind
clinical study (the subjects did not know which treatment was administered),
again on hypogonadic men, that showed that after two years of treatment with
testosterone, over 80% of subjects considered that their libido had improved,
while less than 10% held this view in the control group (Hajjar et al., 1997)
(Figure 7.4).
Same
80
Better
Worse
60
% of subjects
40
20
0
Control Testosterone
Figure 7.4 Percentage of male subjects suffering from hypogonadism in which sex life
had improved, deteriorated, or remained the same after two years of treatment with
testosterone or a control treatment (according to Hajjar et al. 1997).
The role of testosterone has also been demonstrated in initially normal

patients in a very elegant study conducted at the Veterans Administration
Medical Center in Seattle. For 6 weeks, male volunteers were treated with an
antagonist of GnRH (gonadotrophin releasing hormone or gonadoliberin).
This hormone, normally produced by the hypothalamus, stimulates the synthe-
sis of the two gonadotropic hormones (LH and FSH) in the pituitary gland,
which in turn act on the testes to produce secretion of testosterone. In the
chronic presence of an antagonist (receptor blocker) of GnRH, levels of circu-
lating testosterone drop rapidly. The treated subjects observed a strong parallel
decrease of their interest in sex. They had fewer sexual fantasies, masturbated
less often, and the frequency of sex decreased. A fraction of these subjects were
simultaneously treated with testosterone in sufficient quantity to maintain their
circulating concentrations at normal physiological levels, and they accordingly
experienced no decrease in sexual motivation (Bagatell et al., 1994).
There is also a biomedical literature that indicates that the treatment of sex-
ual offenders by androgen antagonists or GnRH antagonists that suppress
either the action or the secretion of testosterone diminishes the sex drive.
However, in many of these studies, the decrease of sexual motivation was
a condition for release from jail of the offenders. The objectivity of their state-
ments about their sexual motivation may therefore be questioned, and the value
of the conclusions drawn from these studies is questionable. It is also interest-
ing to note that even if the sexual motivation of these offenders is decreased
(which is questionable, as we have seen), orientation remains the same and they
are still attracted by children (boys, girls, or both), thus indicating that sexual
orientation is not controlled by the action of hormones in adulthood, as I dis-
cuss in Chapter 8.
Nevertheless, all studies of subjects with spontaneously or experimentally
very low levels of testosterone agree in demonstrating that the addition of tes-
tosterone increases sexual activity or sexual fantasies (Davidson et al., 1979;
Hajjar et al., 1997; Snyder et al., 2000; Wang et al., 2000). In contrast, improve-
ment is not observed following administration of testosterone to subjects suf-
fering from lack of sexual desire if the concentration of circulating testosterone
is normal. Testosterone does not affect erectile potential in response to erotic
stimuli (Carani et al., 1992). It would only be involved in the activation of sexual
motivation.
8
Sex Differences Suggest Homosexuality Is an

Endocrine Phenomenon
Two major types of biological nonexclusive explanations have been advanced to

explain homosexual sexual behavior. Either it is controlled by hormonal factors
that are known to play a role in the control of animal behavior and its sexual
orientation, or it depends on genetic factors that work independently of hor-
mones or by altering their production or action. I consider in this chapter and in
Chapter 9 arguments that suggest the existence of hormonal control mechanisms
of homosexuality.
HORMONES ARE “NORMAL” IN ADULT HOMOSEXUALS:

NO DIFFERENCE IN ACTIVATION
Knowing that sex steroids play a key role in the activation of sexual behavior in
animals and in humans, researchers have logically thought that disturbances in
circulating concentrations of these hormones could be involved in the control
of homosexuality. In animals and to some extent in humans, testosterone acti-
vates the male-typical sexual behavior, and estradiol combined with progester-
one activates female-typical sexual behavior. One could imagine that the
presence of excessive levels of estradiol in some men is responsible for their
sexual attraction to other men (a female characteristic) and that, vice versa, high
concentrations of testosterone induce female sexual attraction toward other

women (a feature normally observed in men). This old theory has been regarded
as plausible by some scientists, and it still enjoys a certain popularity in the
public, but it is obviously inconsistent with animal studies that have established
that the type of behavior produced by males or females is not conditioned by the
type of hormones to which they are exposed in adulthood but rather by the sex
of their brain, which is determined during ontogenesis (organizing effects of sex
steroids, see Chapter 3).
Since accurate and reproducible assays of sex steroids became available during
the 1960s and 1970s, many studies have tested this hypothesis by comparing the
concentrations of circulating sex steroids in men and women in relation to their
sexual orientation. The result of this work is very clear: no hormonal difference
seems to exist between homosexual or heterosexual males or between homo-
sexual or heterosexual women (Meyer-Bahlburg, 1984). Sexual orientation is
not linked in humans to an inadequate activation by sex steroids considered
typical of their sex. Gay men have levels of circulating testosterone that are per-
fectly normal, and the same applies for the concentrations of estradiol and pro-
gesterone in homosexual women. This conclusion is by no means surprising
in light of the knowledge accumulated in animal studies. There used to be a
confusion between the type of behavior patterns expressed by the animal (male-
typical or female-typical sexual performance) and its orientation (toward a
same-sex or opposite-sex partner). When this distinction was clearly estab-
lished, it became clear that sexual orientation is not affected by treatment with
sex steroids in adulthood. Only lesions of the preoptic area or hormonal manip-
ulations made during ontogeny (embryonic or early postnatal) can achieve this
goal (Chapter 4). Furthermore, the circulating testosterone levels are not altered
in homosexual sheep as compared to subjects with a heterosexual orientation. It
is therefore not surprising that the circulating hormone levels are perfectly
normal among gay men and women. This also explains why attempts to change
sexual orientation through hormone treatments have always failed.
CORRELATES OF HOMOSEXUALITY THAT SUGGEST

PRENATAL HORMONAL DETERMINISM
Knowing that in animals sexual differentiation of reproductive behavior and of
sexual orientation is controlled by the action of embryonic steroids, researchers
began looking at the logical consequences. Among the first to do so was endo-
crinologist Günter Dörner, who proposed a theory of homosexuality based on
hormonal embryonic imprinting (Dörner, 1969). According to this theory,
human fetuses destined to become gay in adulthood would experience an
abnormal sexual differentiation due, for example, to exposure to abnormally
Sex Differences Suggest Homosexuality Is an Endocrine Phenomenon 103
low testosterone concentration in boys or a too high level of this hormone

for girls or to a modified response of the brain to these hormones (see also
Chapter 9).
Various arguments of a correlative nature suggest that sexual orientation in
humans could be determined, at least in part, by the early hormonal imprint-
ing (difference in the organization by steroids). The homosexual attraction in
humans is the attraction normally present in a heterosexual woman (attraction to
men) and vice versa. Homosexuality could easily be conceived, from a theoretical
point of view, as the result of the sexual differentiation of this trait being reversed.
Thus one can imagine that a lack of masculinization or incomplete masculiniza-
tion of this behavioral characteristic results in adult men with a homosexual or
bisexual orientation (a sexual orientation toward men, “normally” characteristic
of females). Conversely, a female embryo’s excessive exposure to androgens
would result in an abnormal masculinization of the resulting behavioral trait in
adulthood and in the male characteristic of a sexual attraction toward other
women.
Knowing that the genital structures differentiate much earlier in embryonic
life than the brain (see Chapter 6), it is conceivable that in a given individual,
hormones can induce a “normal” development (corresponding to the genetic
sex) of genital structures and that later during embryogenesis, hormonal changes
can occur such that the differentiation of the brain and sexual orientation do
not occur as expected. This model is perfectly consistent from the theoretical
point of view and would be in complete agreement with animal studies that
have been described previously.
For fairly obvious reasons, it is impossible or very difficult to know what hor-
monal milieu an adult was exposed to during embryonic life. Measures of the
concentrations of circulating testosterone or estradiol in the embryo are associ-
ated with a significant risk (taking blood may cause an interruption of preg-
nancy), and this risk is taken only when required by serious medical reasons but
in no case to make a scientific study, whatever its interest.
However, taking advantage of measurements made during amniocenteses pre-
scribed for medical reasons, several years ago English researcher Simon Baron-
Cohen begun analyzing the relationship between testosterone levels faced by the
human embryo and various behavioral characteristics related mainly to social
skills (what he calls the predisposition to empathy vs. systematization) of young
children. Very interesting correlations have been found (Baron-Cohen, 2006).
However, the number of subjects is limited and so far these studies have only
looked into young children.
Extending such studies to homosexuality involves facing two issues. There is
the very long latency (often exceeding 20 years) between the embryonic hormonal
event supposed to induce homosexuality and its consequences, which are not
seen until adulthood. There is also the limited frequency of homosexuality (2 to

8% of the population). It would be necessary to follow in detail the hormonal
changes of a large number of randomly selected embryos, retain such data for
over 20 years, find these subjects in adulthood, and obtain information on their
sexual orientation (not always openly stated) in order to draw direct conclusions
from this study. No researcher has yet had the ability to conduct such a long study,
so this type of data does not exist.
We are therefore forced to rely on indirect evidence of what the hormonal
embryonic milieu of homosexual individuals might have been. Many traits
are irreversibly affected by embryonic hormones in animals and in humans.
Researchers have thus been able to identify many characteristics that are different
among men and women (see the next sections) as a result of a differential expo-
sure of male and female embryos to embryonic sex steroid. Many studies have
sought to evaluate these characteristics in a comparative manner in populations
of homosexual and heterosexual subjects to determine whether homosexuals
were exposed to atypical hormonal conditions during embryonic development.
Many positive results have been gathered.
Homosexuality is not simply a different sexual orientation; it is accompanied
by many changes. These changes usually concern characteristics that are sexu-
ally differentiated in the heterosexual population, related for example to the
functioning of the inner ear and of specific parts of the brain. There is also a
significant change in some cognitive abilities of individuals, including sexually
differentiated aptitudes not related to sexuality, such as verbal skills or the abil-
ity to visualize three-dimensional space, not to mention differences in the length
of bones and even in some brain structures. All the differences that were detected
suggest the existence of hormonal differences in homosexual individuals during
their infancy or embryonic life.
These studies have another significant interest. They indicate that homosexual-
ity is not only a change in sexual orientation but is also reliably associated with
physical (e.g., relative size of index and ring finger of the hand), functional (activ-
ity of the inner ear), and behavioral (various cognitive skills) differences unrelated
to sexual behavior, though at the behavioral level the reproducibility of results
seems lower and their interpretation more complex. These associations indicate in
an indirect but fairly convincing manner that homosexuality is not a lifestyle
choice to anywhere near the degree that most people believe it is. It is difficult to
imagine how life choices could induce a change in the relative size of the fingers or
a differential functioning of the inner ear. These differences obviously precede the
appearance of homosexual orientation, and the most straightforward and plausi-
ble interpretation is that the same cause (probably hormonal) is responsible for
all of them. I now describe in more detail these differences between homosexual
and heterosexual individuals that are not directly related to sexual behavior.
COGNITIVE AND BEHAVIORAL DIFFERENCES

NOT DIRECTLY RELATED TO SEX
There are many differences in cognitive abilities between men and women.
These differences often have rather limited amplitude (see Chapter 5), but many
studies have shown that, on average, men tend to have better results than women
in mathematical reasoning and in the performance of visual-spatial tasks.
In contrast, women are better than men on tests measuring verbal ability, speed
of calculation, the recognition of facial expressions, and memorizing the loca-
tion of objects (Kiumura, 1999; Baron-Cohen, 2004).
Education clearly plays a major role in the genesis of these differences
(Chapter 5) but there are data showing that the behavioral differences observed
in adulthood result, at least in part, from prenatal differences in hormone
concentrations between male and female embryos—that is, the differences
partly result from the organizing effects of steroids (Collaer & Hines, 1995).
The cognitive differences between men and women would be the result of
biological differences and cultural differences that are likely to be mutually
reinforcing, but the relative importance of these two types of effects remains
to be determined [see Baron-Cohen (2004) for a more detailed discussion of
this issue].
If we consider the possibility that homosexuality is due, at least in part, to
an atypical embryonic hormonal milieu, it is then logical to ask whether the char-
acteristics of cognitive functioning that are sexually differentiated (i.e., differ
between men and women) are different as well in homosexuals compared to het-
erosexuals. Several studies have been devoted to this subject and have provided
positive responses.
Tests for Which Men Are on Average Better than Women
Visual-Spatial Tasks
Several studies have shown that homosexual men have performances inferior
to those of heterosexuals in the execution of many tasks, including visual-
spatial mental rotation tasks, evaluation of the orientation of a straight line, and
aiming at a specific target (Hall & Kimura, 1995; Neave et al., 1999; Rahman
& Wilson, 2003b). Gay men generally have performances similar to those of
women or intermediate between men and women. Some studies, however, have
not shown such differences between homosexual and heterosexual males for
some of these characteristics (Gladue et al., 1990; Tutle & Pillard, 1991).
Studies conducted in women, which according to this theory should identify
masculinized performances in lesbians, have yielded results that are less clear.
A recent study indicates that homosexual women perform mental rotations
slightly better than heterosexual women, but the difference only affects the
speed of execution of the task, not its accuracy (Rahman & Wilson, 2003b).
Aggression
In general, men are more aggressive than women, and two separate studies have
indicated that gay men are less physically aggressive than heterosexual men
(Ellis et al., 1990). No difference in aggression has been identified between les-
bians and heterosexual women (Gladue & Bailey, 1995). One can imagine that
the basal level is very low, so a small increase can pass unnoticed.
Tests for Which Women Are on Average Better than Men
Remembering The Location of Objects

A study based on a large number of subjects has shown that gay men localize
objects more efficiently than heterosexuals, a finding that is consistent with this
feature not having been fully masculinized (that is, increased to male levels by
embryonic testosterone) in homosexuals. However, no difference was observed in
this study between homosexual and heterosexual women (Rahman et al., 2003a).
Verbal Fluency
A 1991 study indicated that homosexual men have a higher verbal performance
than heterosexuals, a finding that conforms with the notion that they were not
completely masculinized and therefore are more feminine from this point of
view (McCormick & Witelson, 1991). Two subsequent studies have not been
able to confirm this difference (Gladue et al., 1990). In addition, a recent analy-
sis based on a large number of subjects reported that in verbal fluency tests,
both gay men and gay women have performances that are atypical for their
gender (Rahman et al., 2003a). Thus, in one of these tests, gay men were better
than all other groups (heterosexual men and women regardless of their orienta-
tion). In another test, gay men and heterosexual women were better than lesbi-
ans and heterosexual men. These data indicate that for these features, gay men
are more similar to women, whereas lesbians are more like men.
Feature Not Usually Differentiated: Manual Lateralization

Several studies have shown that both gay men and gay women tend not to
be right-handed (instead, either left-handed or ambidextrous) significantly
more frequently than heterosexual individuals (Lalumiere et al., 2000; Lippa,
2003b). This feature is not generally regarded as sexually differentiated in the
heterosexual population, in which there are as many right-handed or left-
handed or ambidextrous subjects among men as among women (Lippa, 2003b).
However, the existence of a predominant development of the right side of the

body among men, whereas the left side would be favored in women, has been
suggested (Baron-Cohen, 2004).
A recent meta-analysis of five independent studies also confirms the exis-
tence of a higher probability among homosexual men of not being right-handed
(Blanchard & Lippa, 2007). This increased frequency of non–right-handedness
is interesting from the point of view of the ontogenesis of homosexuality because
the preferential use of one hand over the other is an individual trait observable
before birth (Hepper et al., 1991), even if a reversal of lateralization is possible
after a trauma, such as that associated with a difficult birth. The higher inci-
dence of non–right-handedness among homosexuals of both sexes would be
consistent with the idea of a prenatal determination of this orientation.
These differences in cognitive abilities or in lateralization related to homo-
sexuality are, of course, consistent with the idea that an atypical embryonic hor-
monal milieu may have influenced the two types of variables in parallel. However,
the hormonal contribution to these cognitive traits is probably limited (the role
of the postnatal environment is important here), and many differences have a
low reproducibility. These cognitive changes observed in homosexuals only
moderately support the hormonal theory of homosexuality.
PERIPHERAL PHYSICAL DIFFERENCES

A good number of physical traits are also different between men and women
and/or are known to be permanently influenced by embryonic hormones. Some
of these traits are modified in homosexuals, thus suggesting again that they were
exposed in utero to abnormally high or low concentrations of sex steroids. Two
groups of reproducible studies on such characteristics are of interest, although
not fully conclusive.
Index/Ring Finger Length Ratio

For most women, the index (finger 2 or D2) is almost as long as the ring finger
(finger 4 or D4), so that the ratio of length D2:D4 is very close to 1 (0.973 on
average). In males, the D2:D4 ratio is smaller (around 0.955) (see Figure 8.1).
This sex difference in the relative length of fingers probably reflects differ-
ences in the embryonic hormonal milieu (androgen concentrations higher in
male fetuses than females) at the time of development and growth of the fin-
gers. It is known that testosterone modulates the growth of long bones in the
human embryo, as in other vertebrates. This notion is consistent with the fact
that girls exposed to high androgen levels due to congenital hyperplasia of the
adrenal glands (CAH, see Chapter 6) have a masculinized D2:D4 ratio, a lower
D2 D4
Men Women Homosexual

women
0.975
0.970
0.965
D2:D4 ratio
0.960
0.955
0.950
0.945
Men Homosexual Women Homosexual

men women
Figure 8.1 Ratio of lengths of the index (D2) and ring finger (D4) and sexual
orientation. The ratio D2:D4 is higher in women than in men. It is close to the male
level in homosexual women, thus suggesting they may have been exposed to abnormally
high levels of testosterone during embryonic life. This ratio is not consistently changed
in gay men as compared to heterosexual men.
ratio than in “normal” women (Brown et al., 2002b). A recent study also indi-
cates that the D2:D4 ratio is higher in XY women affected by the complete
androgen insensitivity syndrome (AIS) than in control XY males (Berenbaum
et al., 2009). Moreover, this sex difference in the relative length of fingers is
found in various animal species as well (Brown et al., 2002a; Romano et al.,
2005) and is in animals significantly affected by early treatments with andro-
gens (Lutchmaya et al., 2004; Romano et al., 2005; Manning et al., 2006).
Together, these facts strongly suggest that the D2:D4 ratio is indeed affected by
prenatal androgen levels [see Breedlove (2010) for a recent review], but it has
been pointed out that the real world might actually be more complex than sug-
gested by this simple conclusion. The ratio might be related more directly to the
ratio of androgens to estrogens and might reflect a difference in adiposity rather
than a true difference in finger lengths (Wallen, 2009).
Several independent groups have shown that the D2:D4 ratio is significantly
smaller (and therefore similar to the ratio observed in men) in lesbians than in
heterosexual women, which is in perfect agreement with the theory that assigns
a role for prenatal hormones in the development of homosexuality (McFadden &
Shubel, 2002; Rahman & Wilson, 2003c; Kraemer et al., 2006). This difference
was recently confirmed by a meta-analysis of multiple studies involving thou-
sands of subjects (Grimbos et al., 2010). It was also shown that only lesbians who
identify as masculine (known as “butch”) have this decreased D2:D4 ratio (Brown
et al., 2002a). Another study also confirmed these results independently, showing
that among monozygotic twins (true twins), where one is homosexual and the
other not, the lesbian sister had a smaller (masculine) D2:D4 ratio compared
with the heterosexual sister (Hall & Love, 2003). This work also shows that these
variations of the D2:D4 ratio are independent of genes, since the study is of twins
that have an identical genetic heritage. A more recent study based on a larger
number of subjects, however, indicates a genetic contribution in conjunction
with a contribution of the prenatal environment in the determination of the
D2:D4 ratio (Gobrogge et al., 2008). These studies therefore imply that if testos-
terone is a key factor that determines this ratio, twins may be exposed to different
concentrations of testosterone, which so far has not been experimentally tested.
However, it is also possible that the twins could be exposed to the same concen-
tration of steroids but be differentially sensitive (e.g., differences in the density of
androgen receptors in the intracellular metabolism of the hormone, etc.; see
Chapter 3). To my knowledge, only one study has not replicated the relationship
between the D2:D4 ratio and sexual orientation in women (Lippa, 2003a).
Four similar studies were conducted among men, in whom it could be
expected that homosexuality is associated with a more feminine (higher) D2:D4
ratio, with conflicting results (only one of the four studies conducted among
men jibes with the predictions of the prenatal hormonal theory).
Differences in the Length of Various Long Bones

The growth of the bones of the hand seems to be under the control of embryonic
steroid hormones, which would explain the sex difference of the D2:D4 ratio and
the modification of this ratio in homosexual women, which would be due to their
exposure to an abnormally high concentration of androgens during fetal life.
Pursuing this idea further, researchers looked at the lengths of various long bones
in males and females and the changes in the relationships between these lengths
in homosexual individuals (Martin & Nguyen, 2004). This study was conducted
to test the hypothesis that these ratios could be a reflection of the embryonic
hormonal environment (androgenic and/or estrogenic) and thus provide infor-
mation on the potential hormonal causes of a reversal of sexual orientation.
It was shown that the length of the bones that become sexually dimorphic in
infancy was significantly different between homosexual and heterosexual indi-
viduals, while the bones that become different between men and women after
puberty are not modified according to sexual orientation. Thus, people who have
a sexual preference for men (heterosexual women and homosexual men) have a
smaller growth of the bones of the arms, legs, and hands than subjects who have
a sexual preference for women (heterosexual men and homosexual women).
These data support the idea that gay men have experienced a reduced exposure
to sex steroids during development, and conversely that the homosexual women
were exposed to higher concentrations of sex steroids than heterosexual women,
or alternatively that the sensitivity to these hormones is different in relation to
sexual orientation.
Before this major study based on over 500 subjects (Martin & Nguyen, 2004),
other less systematic work had already identified in males differences in ratios
of the lengths of various parts of the body associated with sexual orientation
[see references in Martin & Nguyen (2004)]. It is therefore quite likely that the
results of this study are reproducible.
Other studies have identified differences compatible with the existence of
endocrine disruption during development in homosexuals, but they appear to
be less reproducible at the present stage of knowledge. These studies should
certainly be pursued.
PHYSIOLOGICAL DIFFERENCES
Oto-Acoustic Emissions
It was discovered in the late 1970s that the inner ear, in addition to its obvious
function in hearing, also emits sounds in the form of barely audible clicks that
can be recorded by placing a sensitive microphone in the canal of the ear. The
subjects do not usually hear these sounds, which are called oto-acoustic emis-
sions (OAE). They are produced either spontaneously or in response to external
short noises (e.g., clicks; Figure 8.2). It is interesting, for the purposes of this
book, that oto-acoustic emissions are different in men and women. Women
produce more oto-acoustic emissions, with a greater amplitude, than men do in
the same situation. This difference is already present during childhood. The
functional significance of oto-acoustic emissions is unclear (they seem to be an
indirect consequence of the functioning of the inner ear), but it is thought that
the sex difference that affects this sound production is largely the result of the
prenatal exposure to androgens in male embryos (McFadden, 2002; 2011).
This hypothesis is supported by studies of OAE in animals. These OAE are
indeed found, mutatis mutandis, both in sheep and in monkeys, and they are
25
Sound pressure (dB)

15
−5
−15
2 4 6 8
Frequency (Hz)
Amplitude of induced OAE Number of spontaneous OAE
0
16 (Clicks at 69 dB)
12
Men Homosexual Women Homosexual

men women
Figure 8.2 The oto-acoustic emissions (OAE) are sexually differentiated and
significantly masculinized in homosexual females. The upper figure shows the
distribution of frequencies and amplitudes of a spontaneous OAE recorded directly
into the ear canal. The individual has a peak of emission at 1,700 Hz. Some individuals
have such peaks at multiple frequencies. The lower figure represents the average
number of spontaneous OAE and OAE amplitude evoked by clicks sounds
measured in the right ear of hetero- or homosexual males and females
(according to McFadden, 2002; 2008).
also sexually differentiated (female OAE are more frequent and of greater
amplitude than in males). OAE were studied in hyenas because in this species,
females are strongly masculinized by androgens during fetal life and immedi-
ately postnatally, so that their external genital structures are at first sight not
different from those of males. The clitoris is greatly enlarged and, for a naïve
observer, seems to have the same form as the male penis. Correspondingly, it
was noted that in this case, the OAE of the females do not have a greater mag-
nitude than those of males. This magnitude is even slightly lower. The OAE of
hyenas are affected by prenatal androgens, because if pregnant mothers are
treated with an antiandrogen (compound that blocks the action of androgens at
their receptor), an amplitude OAE well above normal is observed in treated
young (McFadden, 2008).
Furthermore, in sheep that show the classic sex difference in OAE (female >
male), it was shown that prenatal treatment of female embryos with testoster-
one considerably reduced the magnitude of their OAE (McFadden et al., 2009).
Finally, in both species (sheep and hyena), castration in adulthood does not
affect the OAE, thus confirming that the sex difference affecting this character-
istic reflects the physiological prenatal hormonal milieu but not activation by
steroids in adulthood. In humans also, the sex difference in OAE seems to result
mostly from prenatal organizing effects of steroids, but there is also some evi-
dence for limited activational effects of steroids on this physiological response
(McFadden, 2011). It has been shown, for example, that OAE are masculinized
in women taking oral contraceptives.
Researchers at the University of Texas at Austin found that lesbians and bisex-
ual women have oto-acoustic emissions and acoustic evoked potentials that are
partially male: they emit significantly fewer OAE than heterosexual women, and
these OAE have a lower amplitude (McFadden & Pasanen, 1998; 1999; McFadden,
2002). These data are consistent with the hormonal theory of homosexuality
stating that female fetuses destined to become homosexual or bisexual have
been exposed to abnormally high levels of androgens. These same researchers
also demonstrated that women who had a twin brother and therefore had poten-
tially been exposed to slightly higher levels of testosterone during embryonic life
(testosterone produced by the twin would have diffused to the female embryo)
had masculinized OAE, thus confirming sensitivity to androgens of the response
in humans. However, the same researchers found no difference between the
OAE of homo- and heterosexual males.
Evoked Acoustic Potentials and Alarm Response (“Startle Response”)

The same group of researchers in Texas also showed that the evoked acoustic
potentials (electrical responses induced in the brain by short auditory stimuli)
are also different in men and women, and in (partly masculinized) homosexual
or bisexual women compared with heterosexual women. By contrast, other
characteristics of these evoked acoustic potentials were hypermasculinized in
homosexual men when compared with values observed in heterosexual men
(McFadden & Champlin, 2000). This study is again consistent with the idea that
lesbians were exposed to abnormally high levels of androgens during fetal life
but that, contrary to what one might intuit, gay men have been masculinized as
much or even more than heterosexual men. Indeed, if we remain in the context
of an interpretation based on differences in the concentrations of testosterone
that embryos have been exposed to, we should conclude that if lesbians were
exposed to concentrations higher than heterosexual women, they would have
been masculinized and therefore present a male-typical sexual orientation
(preference for women), which is consistent with the theory. In contrast, these
data would suggest that gay men were exposed in utero to testosterone at above-
normal levels. No definitive explanation for these paradoxical observations
about gay men has been provided to date, but several potential mechanisms
have been suggested (McFadden, 2011). Since it is a recurring paradox that
I have noted before, it may be that we still have an imperfect understanding of
some hormonal effects.
An independent study in London has also focused on the analysis of another
physiological response known to be sexually differentiated—the particular fear
response called the alarm or startle response. Following a loud noise, there is a
blink of the eyes, but this response is partially inhibited if the loud noise is pre-
ceded by a lower noise. This inhibition by the preceding sound (prepulse inhibi-
tion) is usually less pronounced among women than among men. The group of
Rahman and his colleagues recently showed that inhibition of the blink by a low
alarm is stronger among lesbians than among heterosexual women (Rahman
et al., 2003b). This response is masculinized in agreement with the prenatal hor-
monal theory of homosexuality. However, here again, the authors have not found
this difference in inhibition between homosexual and heterosexual men.
Positive Feedback in Response to Estradiol

In rats and many other mammals, ovulation is induced in females by a gradual
increase in plasma estradiol during the follicular phase of the cycle. When these
concentrations reach a threshold, they induce a positive feedback resulting in a
significant peak of release of luteinizing hormone (LH), which is responsible
for ovulation, among other functions. The LH peak can be artificially induced
in females by a single injection of a high dose of estrogen, but males are unable
to produce such a peak of LH in response to estrogen. This neuroendocrine sex
difference is organized during the embryonic and perinatal period by the same
hormonal mechanisms that differentiate sexual behavior. The male embryos are
masculinized and defeminized by early exposure to testosterone, whereas the
female phenotype develops in the relative absence of hormones (Chapter 3).
The injection of testosterone in female embryos of rats induces the loss of the
LH peak in response to estradiol, but males castrated immediately after birth
are able to produce such a peak.
Having suggested that human male homosexuality could be the result of
inadequate masculinization during embryonic life, Dörner wondered whether
the positive feedback of estrogen on LH secretion might be different in homo-
sexual and heterosexual males (Dörner, 1969). In women, an estrogen injection
at an appropriate time of the ovarian cycle induced a strong increase of circulat-
ing levels of LH, whereas this response was not observed in men. In keeping
with his theory, Dörner showed that gay men react to an injection of estrogen
by a significant increase in blood levels of LH. This increase is lower than that
observed in women, but it is still significantly higher than among heterosexual
men (Dörner, 1972; 1976; 1980).
These publications raised a huge controversy, in part because of the research-
er’s views concerning preventive treatments that he suggested should be estab-
lished to prevent the emergence of gays, but also because the positive feedback
of estrogen on LH that is quite evident in the rat is apparently not present in
monkeys (Baum et al., 1985), a finding that raises the question of its existence
in humans. Many researchers have thus doubted the reality of the effect identi-
fied by Dörner. In an attempt to resolve this controversy, American neuroendo-
crinologist Brian Gladue therefore decided to test the reproducibility of the
effect. In a study published in Science in 1984, Gladue and colleagues clearly
demonstrated the reproducibility of the effect identified by Dörner. After a
single injection of Premarin, a compound with estrogenic action, women
showed a major increase in LH 72 to 96 hours later (Gladue et al., 1984). This
increase was not observed in men, but again an increase of intermediate mag-
nitude was observed in men who reported having had a homosexual orienta-
tion during their life (see Figure 8.3).
This LH response to the injection of an estrogenic compound looks similar,
but is not identical, to the positive feedback observed in the control of ovulation
in rats [see for discussion Baum et al. (1985)]. It is important that there is a dif-
ference in this endocrine response in men according to their sexual orientation.
However, based on available data, it is impossible to know whether this differ-
ence concerns a sexually differentiated neuroendocrine mechanism, which
would suggest that homosexuals were exposed to “abnormal” endocrine condi-
tions during their ontogeny. Alternatively, the differential response of LH to
estrogens might just reflect a stable but unidentified difference in adult testicular
Women
200
LH (% of baseline)
150 Homosexual
men
100
Men
50
Base 24 48 72 96
line Hours after
injection
Premarin
injection
Figure 8.3 Changes in blood levels of luteinizing hormone (LH) in response to injection
of a single dose of an estrogenic compound (Premarin) in heterosexual men and women
and in homosexual men. The injection induced in women after 72–96 hours an increase
in LH that was not present in men but was seen in an attenuated form in homosexuals
(according to Gladue et al., 1984).
activity that could interfere with the positive and negative feedback on LH
(Baum et al., 1985; Gladue, 1985).
Further studies should be conducted on this issue, but the fact remains that
all this work indicates the existence of a (neuro)endocrine difference between
homosexual and heterosexual men. We cannot conclude, at present, that this
difference concerns a neuroendocrine mechanism, located most likely in the
hypothalamus, that is known in rats to be differentiated (defeminized) during
ontogenesis after exposure to testosterone rather than an aspect of testicular
physiology associated in a less specific or even unknown manner to the differ-
entiation of sexual behavior and brain. In either case, these data demonstrate
the existence of a difference in the function of the hypothalamic-pituitary-
testicular axis associated with homosexuality.
Brain Activation in Response to (Putative) Pheromones

In many species of mammals, some body odors emitted by males or females
specifically attract sexual partners and stimulate (or inhibit in some situations)
hormonal activity underlying reproduction. These odors, grouped under the
term pheromones, are generally the product (more or less transformed) of sweat
glands or may be present in urine, feces, or vaginal secretions. It was tradition-
ally accepted that pheromones were detected primarily by an independent, aux-
iliary olfactory sensory system, known in the literature as the vomeronasal organ
or organ of Jacobson or accessory olfactory system (Keverne, 1999). More recent
studies tend to show, however, that both olfactory systems, the main and the
auxiliary, play a role in the perception of sexual pheromones.
The existence of pheromones in humans is controversial, and many research-
ers believe that the human vomeronasal organ is vestigial and completely non-
functional. Contrary opinions have been expressed. A detailed discussion of
this controversy goes far beyond the scope of this book, but the interested reader
may consult more or less recent journal articles on the subject (Foidart et al.,
1994; Meredith, 2001; Halpern & Martinez-Marcos, 2003; Wysocki & Preti,
2004). However, it is reasonably well established that in humans, some com-
pounds present either in armpit sweat or in the urine can be detected, con-
sciously or not, and can influence responses such as choice of location (chairs in
a waiting room; which of a series of identical urinals to use) and/or can modify
the physiology and mood of subjects, even their sexual motivation. Experiments
highlighting these effects are numerous and clearly not easy to replicate. They
do not usually determine whether perception involves the main olfactory system
or the vomeronasal system.
This problem is of little importance for what concerns us here, namely the
possible difference of reaction to pheromones depending on heterosexuality or
homosexuality. It was shown that the male rat brain is activated by the odor of
bedding soiled by females, but no brain activation was observed in males exposed
to bedding soiled by a male (Chapter 4). A perinatal treatment with a compound
that blocks the aromatization of testosterone into estradiol irreversibly trans-
forms a young rat into an adult male who is bisexual or homosexual and who will
display behavior of sexual receptivity when placed with another male (Bakker
et al., 1993a). In these subjects with a reversed sexual orientation, brain activity
will correspondingly be stimulated by the smell of litter soiled by a male rat.
Early hormonal treatment has therefore changed the brain activity in response to
olfactory stimuli from an individual of the same sex (Bakker et al., 1996a).
In recent years, through advances in medical imaging techniques that can now
detect which brain areas are activated during the presentation of a specific stimu-
lus (among other things), researchers were able to identify a similar phenomenon
in humans. Though this idea is disputed, two compounds derived from steroids
are considered by some to be pheromones in humans: 4, 16-androstadien-3-one

(AND) derived from testosterone in men, and the estrogenic compound estra-1,
3,5 (10),16-tetraene-3-ol (EST) in females. AND was identified in armpit sweat
in humans, while EST is present in urine in women. The effects of these com-
pounds appear to vary depending on the dose and the study protocol, but they
have been shown to act on the autonomic nervous system and to alter mood and
sexual arousal depending on context. Using two separate techniques of medical
imaging, positron emission tomography (PET) and functional nuclear magnetic
resonance (fMRI), a group of researchers from the Karolinska Institute in
Stockholm have asked whether these compounds produced a differential activa-
tion of the brain in men and women and, if so, whether this activation was
affected by the sexual orientation of the subjects. In a first step, differential brain
activations in the two sexes have been identified (Savic et al., 2001). In males,
exposure to EST (but not to AND) induced activation of the anterior hypothala-
mus, a brain area clearly involved in the control of male sexual behavior. Among
women, a hypothalamic activation was detected after exposure to AND but not
to EST (Figure 8.4).
More recently, researchers have analyzed the association of these differential
brain activations with the sexual orientation of the subjects. They were able to
show a profoundly altered activation in homosexual men. Specifically, the pre-
optic area and hypothalamus were significantly activated by the perception of
AND (a male stimulus) in gay men as in heterosexual women, but not in hetero-
sexual men. In contrast, exposure to EST, a female stimulus, activated hypotha-
lamic areas in heterosexual men but not in heterosexual women or homosexual
men (see Figure 8.4) (Savic et al., 2005; Berglund et al., 2006).
Similar results have been observed in parallel in homosexual women in whom
hypothalamic areas are not, contrary to what is seen in heterosexual women, acti-
vated by the male odor AND. One does not see in lesbians an activation by the
female olfactory stimulus, EST (Berglund et al., 2006). These results show that the
brains of homosexual men and women are differentially activated by odors with
sexual connotations compared to the brains of heterosexual subjects of the same
gonadal sex. The interpretation of these results is complicated by the fact that the
olfactory stimuli were used in the experiments at high concentrations (crystalline
form of the steroids), presumably much higher than what could ever be encoun-
tered in the real world. It is therefore difficult to ascertain whether similar differ-
ential reactions would be observed with biologically relevant concentrations.
Furthermore, the data do not indicate, whether the differential perception of these
odors with a sexual significance is the cause or the consequence of the sexual ori-
entation of the subjects tested. One can indeed imagine that the smells produced
by an individual can activate brain areas involved in sexual motivation among
homosexuals of the same sex and therefore encourage a homosexual attraction,
Heterosexuals Homosexuals
Men Women Men Women
AND − + + −
EST + − − −
Figure 8.4 Exposure to potential pheromones differentially activates the hypothalamic
area of males and females, and the response is modified based on their sexual
orientation. The upper figure presents a schematic view of a sagittal section of human
brain showing the region (arrow and black square) where metabolic activations were
detected by medical imaging. The table indicates whether these hypothalamic activations
are present (+) or absent (–) depending on the sex of the individual, his or her sexual
orientation, and the compound used. AND = compound produced by men; EST =
compound produced by women (according to Savic et al., 2005; Berglund et al., 2006).
although these odors would have no effect among heterosexuals. Conversely, it is

also conceivable that being gay produces a sensitization to certain type of odors
from individuals of the same sex and that this learning is reflected at the level of
brain activity detected by medical imaging. By analogy with animal data, how-
ever, one might choose the first of these interpretations.
DIFFERENCES IN BRAIN STRUCTURE

As we have seen in the section on animal studies (Chapter 3), several limbic and
hypothalamic nuclei involved in the control of sexual behavior are sexually dif-
ferentiated. They are either larger in males than in females [sexually dimorphic
nucleus (SDN) of preoptic area, bed nucleus of the stria terminalis (BNST)] or
more developed in females than in males [antero-ventral nucleus of the anterior
hypothalamus (AVPV), ventromedial nucleus of the hypothalamus (VMN)].
Researchers have questioned whether homologous nuclei, or at least nuclei

located at the same places of the brain, were sexually differentiated in humans
and whether the volume of these nuclei was modified according to the sexual
orientation of the individuals. It is known that the volume of several of these
nuclei (SDN, BNST, AVPV) is determined in rats by the action of sex hormones
during the embryonic or immediately postnatal period and cannot be modified
during adulthood. If that were the same in humans, the volume of these nuclei
could then represent a reliable marker of the hormonal milieu in which an indi-
vidual developed. If these nuclei were modified according to the sexual orienta-
tion of individuals, they would provide an additional argument concerning the
validity of the hormonal hypothesis of sexual orientation.
The experimental material necessary to conduct such studies is very difficult
to obtain. Structures that interest us here have a small volume and still cannot be
identified by the in vivo imaging techniques currently available [magnetic reso-
nance imaging (MRI) or positron emission tomography (PET)]. Only studies of
postmortem tissue can provide useful information, and they have only been
performed so far by a limited number of researchers with the few samples avail-
able in the human brain banks around the world. Some very interesting results
have nevertheless been identified. They mainly concern the suprachiasmatic
nucleus (SCN), the anterior commissure, and the sexually dimorphic nucleus of
preoptic area (SDN-POA).
The Suprachiasmatic Nucleus

Morphometric analysis performed on a group of 18 heterosexual men who died
of various causes and ten homosexual men who died of various diseases related
to AIDS have shown that the SCN was significantly larger (1.7 times) in homo-
sexuals than in heterosexuals (Swaab & Hofman, 1990). This nucleus in homo-
sexuals also contained about two times more neurons than in heterosexuals.
Thinking that the difference could be related to the indirect consequences of
AIDS, these researchers also analyzed the SCN of six heterosexual men who
died of AIDS and found that these subjects had a perfectly normal size SCN
typical of heterosexuals.
This was the first study to demonstrate a neuroanatomical difference related
to sexual orientation. As such, it deserves to be mentioned because it provided
an important argument in favor of the idea that homosexuality is associated
with specific biological changes. However, this neuroanatomical difference does
not provide information about the mechanisms that lead to homosexuality.
Although SCN, which is mainly related to the control of circadian rhythms,
seems to be involved in the control of certain aspects of reproduction (Swaab &
Hofman, 1990), there is no data suggesting a direct involvement in the control
of sexual orientation (Kruijver et al., 1993). And since there is also no evidence
that the volume of the SCN is sexually differentiated (greater or smaller in men
than in women), the difference observed between homosexual and straight
men does not provide any argument that might suggest a lack of masculiniza-
tion of the brain.
The Anterior Commissure

The anterior commissure is the bundle of fibers that connects the two cerebral
hemispheres in the anterior hypothalamus. A study of the brains of 90 subjects
(heterosexual men and women and homosexual men) showed that this connec-
tion is significantly more developed in women than in men when measured in
the mid-sagittal plane (Allen & Gorski, 1992). Moreover, the size of this anterior
commissure is larger among homosexual men than among heterosexual men and
even than in women (Figure 8.5). The anterior commissure is in fact 34% bigger
in homosexual men than in heterosexual men and 18% larger than in women.
This neuroanatomical difference associated with genital sex, and in men also
associated with sexual orientation, could at least partly explain some gender
differences observed in various cognitive tasks (such as visual-spatial skills and
verbal tasks) and sex differences of functional lateralization (male brains seem
more strongly lateralized than female brains). Furthermore, gay men are quite
similar to women for some of these criteria (see earlier parts of this chapter),
and the large size of their anterior commissure is perfectly correlated with their
15
Mid-sagittal surface (mm2)
12
Woman Men Homosexual

men
Figure 8.5 Size of the brain anterior commissure estimated from its surface in the
mid-sagittal plane in men and women and in homosexual men (redrawn from Allen &
Gorski, 1992).
cognitive abilities. However, some studies have failed to confirm these differ-
ences in cognitive abilities between heterosexual and homosexual men, so they
must be treated with caution even though many aspects of experimental proto-
cols, and in particular the selection of subjects, are likely to explain these fail-
ures of replication.
It is important to note here, once again, that the neuroanatomical difference
in gay men affecting the size of the anterior commissure, a structure that has no
direct link with sexuality, is an additional strong argument suggesting that
homosexuality is not the result of a choice, but instead a complex phenotypic
change that goes far beyond the field of sexuality and has in all likelihood bio-
logical bases beyond the control of the individuals.
The Corpus Callosum

In 1982, biologist Christine de Lacoste-Utamsing and her colleague Ralph
Holloway published an article based on the postmortem study of human brains
showing that another connection between the left and right cerebral hemispheres
differs as a function of gender. The corpus callosum is by far the most important
interhemispheric connection in humans (de Lacoste-Utamsing & Holloway,
1982). de Lacoste-Utamsing and Holloway showed that the shape of a part of the
corpus callosum called the splenium was sexually differentiated, and was larger
in women than in men. This work received enormous attention, and since the
structure involved is large, it is possible now to view it by brain imaging in living
subjects, thus making study much easier. Several studies have therefore addressed
this question. Some could not reproduce the difference between sexes originally
identified, while others have confirmed it. The existence of this difference remains
controversial at present, but I shall not go into detail on the debate because it
remains peripheral to the topic of this book.
However, a relationship between homosexuality and the shape of the corpus
callosum has been identified recently and deserves our full attention. As I explained
at the beginning of this chapter, male homosexuality is associated with a signifi-
cant change in the manual lateralization: gay men are more often not right-handed
(left-handed or ambidextrous) than heterosexuals. This unusual lateralization of
the use of hands is itself normally correlated with increased size of the corpus cal-
losum, especially its posterior part, called the isthmus. Researchers have therefore
questioned whether the unusual lateralization in homosexual men (greater fre-
quency of left-handers and ambidextrous) was associated with a difference in the
size of the isthmus of the corpus callosum. To obtain results that are exclusively
related to sexual orientation and are not simply a reflection of the manual lateral-
ization, they compared the size of the isthmus by magnetic resonance imaging
in 12 homosexuals and ten heterosexuals who were all strictly right-handed.
They observed that the isthmus of the corpus callosum was significantly greater
among homosexuals, thus indicating that they have, like women, a less pro-
nounced brain asymmetry than do heterosexual men. Right-handed homosexu-
als thus have a cerebral functional asymmetry less developed than in heterosexual
men, and the motor lateralization is partially dissociated in these homosexuals
from the cerebral lateralization. Moreover, during this same study, researchers
were able to show that statistical analysis could correctly identify the sexual orien-
tation of the subject in 21 cases out of 22 based not only on the size of the isthmus
of the corpus callosum but also on the basis of different cognitive tests already
mentioned in this chapter. These data demonstrate an association between a neu-
roanatomical structure and cognition, and join many other results that indicate
the existence of anatomical differences between the brains of homosexual and
heterosexual men that indirectly support the hypothesis of a biological basis to
sexual orientation (Witelson et al., 2008). Remember also that being left-handed
or right-handed is a characteristic that appears early in development (Hepper
et al., 1991). This feature cannot be a personal choice. The fact that it is correlated
with sexual orientation suggests that sexual orientation is determined early.
The Sexually Dimorphic Nucleus of the Preoptic Area

As explained earlier, homosexuality can be induced experimentally and irrevers-
ibly in the laboratory rat by altering the hormonal milieu of the embryo or the
young pup just after birth (Bakker et al., 1993a; Bakker et al., 1996a). This change
in sexual orientation in rats is associated with a decrease in the volume of a group
of cells located at the base of the brain (SDN-POA), which is normally larger in
males than in females and develops to a female size in male rats that have been
made “gay” by the early hormone treatment (Houtsmuller et al., 1994).
One or more sexually dimorphic nuclei are also present in the preoptic area of
the human brain. A preoptic nucleus significantly larger in men than in women
was first identified by Dick Swaab and colleagues at the Netherlands Institute for
Brain Research in Amsterdam in 1985 (Swaab & Fliers, 1985). This difference
is not present at birth but gradually develops during the first 10–15 years of life
and persists until an advanced age (> 80 years), although its amplitude tends to
decrease with age (Swaab & Hofman, 1988). A comparative analysis of the size of
this nucleus in the brains of homosexual men, however, identified no difference
in volume compared to heterosexual men, although this study highlighted a dif-
ference in the volume of another nucleus as a function of sexual orientation of
the subjects (see preceding section on the suprachiasmatic nucleus).
A few years later, researchers from the laboratory of Roger Gorski (who had
identified the SDN-POA of rats) at the University of California, Los Angeles, ana-
lyzed again the human preoptic area and described four distinct cell condensations
(nuclei) in this region. They called these nuclei the INAH 1 to 4 for the interstitial
nuclei of the anterior hypothalamus. Two of these nuclei were found to be larger in
men than in women (INAH 2 and 3) (Allen et al., 1989). It was shown a little later
that the INAH 3 was significantly smaller in homosexual men than heterosexual
men. Its average size was in fact similar to the size of INAH 3 observed among
women (LeVay, 1991) (Figure 8.6).
Bill Byne’s independent study based on different brains confirmed the reduced
size of INAH 3 in male homosexuals compared to heterosexuals, though the
magnitude of the difference observed in this replication was lower than in the first
study of LeVay and thus not statistically significant (Byne et al., 2001). This study
also showed greater cell density (more cells per unit volume) in the INAH 3 of
homosexuals than in heterosexuals. The nucleus of homosexuals thus has a cel-
lular composition identical to that of heterosexuals (same number of neurons),
but the neurons are closer to each other on average, possibly because they have
H 4
INA2 3
1
SO
3rdV
OC
AIDS
0.20 Other
INAH 3 volume (mm3)
0.15
0.10
0.05
0.00
Women Men Homosexual
men
Figure 8.6 Schematic representation of the human hypothalamus illustrating the
position of the four interstitial nuclei of the anterior hypothalamus (INAH) in relation
to the 3rd ventricle (3rdV), the optic nerve chiasma (OC), and supraoptic nucleus (SO).
The lower figure represents the values of the volume of INAH 3 measured in women,
men, and gay men who died of AIDS or of another cause. The horizontal bars represent
the average for each group of subjects (redrawn from LeVay, 1991).
formed fewer synapses during development. This observation and its interpreta-
tion, however, need to be confirmed.
Many critics have focused on the fact that many gay brains used in these stud-
ies came from individuals who had died of AIDS. It was therefore argued that the
small size of the INAH 3 in these subjects arose from either AIDS or the inten-
sive medical treatments to which these patients had been submitted. In the pres-
ent state of knowledge, this critique is unfounded. Indeed, LeVay and Byne were
perfectly aware of this problem and whenever possible included homosexuals in
their samples who had died of a cause other than AIDS, as well as heterosexuals
who died of AIDS. These data have not provided an argument to suggest that
AIDS and its treatment can reduce the size of the SDN (see Figure 8.6). Thus, in
the study of LeVay, the size of the SDN of heterosexuals who had died of AIDS
was larger than the average size of that of homosexuals. In addition, the study of
Byne and colleagues included nine heterosexual men who had died of AIDS and
22 who had died of other causes. A difference in volume of INAH 3 could not be
detected between these two subgroups. It should also be recalled here that in
rats, SDN size is determined during early life and can no longer be affected by
any known treatment in adulthood.
It was also said that the number of brains used in these studies was too
limited. This objection is clearly based on a misreading of the results. LeVay’s
study was based on the brains of 41 people, including 19 homosexuals, which
given the difficulty inherent in obtaining such samples is a respectable or even
remarkable sample size. The difference he observed in the volume of INAH 3
according to gender and sexual orientation was quite significant (p < 0.00014,
or less than 2 chances in 10,000 of obtaining a difference of such an amplitude
by random fluctuations in sampling). So it is clearly a misunderstanding of the
meaning of statistical tests to say that the observed difference could simply be
caused by fluctuations of sampling. The result of LeVay was by and large repro-
duced independently by Bill Byne, which reinforces its validity.
The presence of a smaller SDN in male homosexuals (IANH 3) acquires an
especially important meaning when connecting this observation to the animal
studies that have been described in detail previously. Remember the following:
1. The nucleus is located in the center of the preoptic area, which plays
a key role in controlling male sexual behavior (Nelson, 2005).
2. The lesion of the SDN in several mammalian species induces a change
in male sexual orientation from strictly heterosexual to either
homosexual or bisexual (Paredes & Baum, 1995).
3. The larger size of SDN in males compared to females is determined
exclusively by the action of sex steroids during embryonic and
postnatal life (masculinization by testosterone and its estrogenic
metabolite, estradiol) (Jacobson et al., 1981; Arnold & Gorski, 1984;

Rhees et al., 1990). Hormone treatments in adulthood have no effect
on the size of the nucleus.
4. Perinatal treatment with an aromatase inhibitor that affects the sexual
orientation of male rats (Bakker et al., 1993a; Bakker et al., 1993b) in
parallel reduces the size of the SDN of the preoptic area (Houtsmuller
et al., 1994).
5. In sheep, a smaller SDN of the preoptic area is associated with
homosexual orientation of male sexual behavior (Roselli et al.,
2004b), just as in humans. This nucleus also expresses less aromatase
in homosexual sheep than in those who are sexually attracted to
females. Recent experiments also indicate that, as in rats, the size of
this nucleus in sheep is controlled by the action of testosterone during
embryonic life (Roselli et al., 2007).
If these results of experiments on animals are transposed to the human SDN

(INAH 3), the difference in the size of the nucleus observed in homosexuals is
an argument to support the theory of an early hormonal origin of homosexual-
ity. Homosexuality in the human male would be the result of the early exposure
to low concentrations of testosterone. Small SDN in homosexuals would be the
signature of this atypical embryonic hormonal environment and, given the role
of the SDN in animals, could even be considered to be one cause—maybe the
cause—of homosexuality.
However, there are a number of considerations that should temper the
tendency to extrapolate too quickly from this finding.
1. The homology (strict identity) between the preoptic SDNs of rats,

sheep, and humans is not established so far. The preoptic area has a
complex architecture and in humans, in particular, four INAH were
identified and only one is changed among homosexuals (INAH 3).
The work of Swaab had identified another sexually dimorphic nucleus
in the human anterior hypothalamus (Swaab & Fliers, 1985), which is
likely to be equivalent to INAH 1 (LeVay, 1991), but INAH 1 was not
shown to be larger in men than in women (Allen et al., 1989).
Differences in sampling (different age of subjects in both studies)
almost surely explain this discrepancy (Garcia-Falgueras & Swaab,
2008), but before we have a replication of these studies, this
interpretation remains uncertain. It is therefore possible that the SDN
identified in different animal species and humans are merely similar,
not strictly homologous. Additional anatomical and neurochemical
studies, admittedly complex but nevertheless possible, should help
answer this objection. But even if a strict neuroanatomical homology

were established between the human INAH 3 and the SDN in rats and
sheep, this would not demonstrate that all the characteristics of these
nuclei must be identical (see 2 and 3).
2. No studies are available (and probably will never be, for obvious
logistical reasons) to demonstrate that the size of the human INAH 3 is
determined exclusively by embryonic hormones. This is the case for the
SDN in rats and also in sheep, but to date nothing ensures that the
same conclusion is true in humans. The volume of most dimorphic
nuclei in mammals is determined irreversibly by embryonic hormones
acting during a defined critical period, but there are some exceptions in
which the hormones can still change the size of a sexually dimorphic
nucleus in adulthood [e.g., part of the amygdala (Cooke et al., 1999),
the medial preoptic area in rat (Bloch & Gorski, 1988a; b)]. A recent
study has also demonstrated that the size of INAH 3 is slightly reduced
in human males castrated in adulthood for medical reasons (prostate
cancer). Although the difference was not statistically significant, it
remains possible that the volume of INAH 3 is at least partly
determined by hormonal status in adults (Garcia-Falgueras & Swaab,
2008). In addition, the study of Bill Byne suggests that the small size of
the INAH 3 in gay people would not be accompanied by a decrease in
the number of neurons but would rather result from their greater
density (Byne et al., 2001). This seems different from what is known in
rats, where the smaller SDN of females is due to enhanced neuronal
death during development. Adult females have fewer neurons in SDN
than males. It is therefore important to replicate this observation of Bill
Byne in order to further assess how far the human–rat comparisons
can be pursued in this area.
3. Although the human INAH 3 is located in roughly the same place as
the SDN of rats, manipulation (e.g. lesion) has not been, nor will be
for obvious ethical reasons, carried out to confirm that INAH 3
indeed controls sexual orientation in humans.
4. The studies of LeVay and Byne cannot answer the question of
whether the smaller INAH 3 in homosexuals is the cause or the
consequence of their sexual orientation. When causal relations
between brain and behavior are addressed, one is spontaneously led to
think that the changes in structure or function of the brain determine
changes in behavior rather than vice versa, but numerous studies have
demonstrated that, contrary to beliefs that were widespread until very
recently, the structure of the brain is plastic. For example, it was
shown that if rats are raised in environments enriched with a more
complex spatial structure, the size of their hippocampus (a brain

area heavily used for orientation in space) and the complexity of its
neurons are increased as compared with control rats. Even in humans,
it is established that the use of a specific area of the brain leads to its
hypertrophy in a manner similar to the increase of the mass of a
muscle that is highly used by an athlete.
One could thus imagine that the small size of the SDN in homosexuals is
induced by an aspect of their behavior or lifestyle, rather than being the cause of
their orientation. It has been observed that many homosexuals living in New
York and California, from which came the subjects in LeVay’s study, had sexual
activity involving large numbers of relationships with many partners. One might
therefore think that the smaller size of SDN could be the result of that intense
and diversified sexual activity. This argument cannot be formally rejected, but
all cases of behavior-induced plasticity in the nervous system that have been
described both in animals and in humans never concern the hypothalamus and
limbic system but instead are almost always related to the cerebral hemispheres.
These brain areas seem to react to plastic life experiences of the subject, whereas
the hypothalamus appears to be controlled by intrinsic physiological mecha-
nisms based largely on hormonal changes. It is not excluded that in the future
researchers will discover changes in the hypothalamus induced by experience,
but in the current state of knowledge, data do not support this hypothesis.
In a minimalist interpretation, one can imagine that the small size of the
homosexual INAH 3 represents a signature of the early hormonal environment
in which these individuals have developed, but the size of the nucleus is not
responsible for the sexual orientation of the individuals. In other words, the
small size of the INAH 3 and the sexual orientation of the concerned individual
would have both been induced by the presence of abnormal embryonic hor-
monal conditions (probably an overly low concentration of testosterone), but
these two features would not be directly related in a causal manner.
Alternatively, if we consider that lesions of the preoptic area (including SDN) of
rats or ferrets change their sexual orientation, it is also possible that there is actually
a causal chain that, from a low level of embryonic testosterone, leads to the develop-
ment of a small INAH 3 that is the cause of the homosexual orientation. To defend
this hypothesis further, it should be previously determined by what mechanism a
small SDN/INAH 3 can be the cause of a reversed sexual orientation.
Finally, we should note that even if we consider the latter interpretation of
these data sympathetically, the fact remains that the size of INAH 3 and cellular
changes that must necessarily accompany the changes in the volume of this
nucleus cannot be the only cause of homosexuality. Indeed, although the average
volume of INAH 3 in homosexuals is statistically smaller than in heterosexuals,
there is some overlap between the volumes observed in both groups. In other
words, the smallest of the INAH 3 observed in heterosexuals have a volume
smaller than the largest nuclei observed in homosexuals. The size of this nucleus
thus cannot be the sole cause of homosexuality. This feature may predispose to
sexual orientation, but not produce it by itself.
PROBLEMS AND GENERAL INTERPRETATION

If we review all the studies presented in this section, it is clear that homosexual-
ity in men and women is significantly associated with a range of physical, func-
tional, and behavioral characteristics that are modified from what is normally
seen in heterosexual individuals. These characteristics are not changed in all
studies, however, and are not always observed in both sexes. The question arises,
therefore, of why these correlations with independent markers of homosexual-
ity are not stronger and more reproducible. Various factors probably contribute
to obscuring the relationship.
1. Homosexuality is a complex phenomenon that may have multiple

causes that are not the same in all individuals. Prenatal hormonal
causes may concern only certain subjects, so the traits under study
would be changed only in some homosexual individuals.
2. The classification of subjects as homosexual, heterosexual, or bisexual
is often based on questionnaires or spontaneous statements. These
answers may be biased for a variety of reasons, and the classification
of individuals included in the studies could be partially wrong, which
would dilute the relationship with the so-called “markers” of
homosexuality.
3. The studies conducted so far have been unable to determine whether
the hormonal difference supposed to lead to homosexuality is the
embryonic concentration of circulating testosterone or the brain
sensitivity to the action of this steroid. If the level of circulating
testosterone is affected during a critical period of development of the
sexual orientation, orientation will be affected, as well as all other
androgen-dependent responses that are developing at the same time. If
instead it is the sensitivity of the brain to testosterone that is changed but
the circulating levels are normal, we can expect to find adult homosexual
individuals in whom most physical or functional characteristics are
identical to those of heterosexuals. Only neural or behavioral
characteristics that depend on the same mechanism of sensitivity to
androgens as sexual orientation will be affected. Both options are not
exclusive. It is possible that for some homosexuals, there are correlations
with peripheral changes (if the circulating testosterone was affected) but
that these correlations are not present among other subjects (for whom
brain sensitivity to testosterone is changed).
4. The dose-response relationships that link the concentration of
circulating testosterone during embryonic development with the
mechanism controlling sexual orientation and its various correlates
are not known and could be different. Studies of animal behavior and
physiology have shown that among a group of responses controlled by
hormones, some are affected by lower doses than others. If we
imagine that human sexual orientation is more sensitive to small
variations in testosterone levels than other hormonal responses
described above (to which it is sometimes correlated), we can easily
obtain homosexual individuals in whom the trait associated with
homosexuality is affected (strong hormonal changes) and others for
whom it is not (more subtle hormonal changes).
5. In the same way, the critical period during which sexual orientation
and its various correlates are determined might be partially different,
as is the case for various responses studied in animals. Hormonal
events occurring at a specific stage of development can affect one
response without affecting the others significantly. Since we know
nothing of these critical periods, this possibility is difficult to assess.
All the interpretations set out in items 3 to 5 of this list are derived
directly from studies on animal models and could easily be tested on
animals. In humans, obvious ethical reasons prohibit deliberate
manipulations of embryonic hormones, and it is therefore much more
difficult to obtain information on the mechanistic aspects of
hormonal action during ontogeny.
6. Finally, any prenatal endocrine phenomenon that leads to
homosexuality is obviously not completely deterministic but simply
modulates the probability of occurrence of this orientation. It is
possible and even likely that to determine sexual orientation in
adulthood, it is necessary that various hormonal and biological
changes accumulate in the same subject or that the predisposition to
homosexuality induced by embryonic hormones is strengthened by
not-yet-identified aspects of interaction with parents, teachers, or
society in general. It is possible that a predisposition to homosexuality
that would be controlled by prenatal biological factors can only be
manifested in particular circumstances of the environment. These
have not been identified. If these “social” influences exist, their
importance is probably less than that of the embryonic hormonal
factors, since they are so far much better identified.
9
Sexual Orientation in Clinical Cases
The previous chapter was devoted to the description of correlations identified

between homosexuality and changes in sexually differentiated features. Although
alternative interpretations are possible in some cases (and some cases only), the
likely interpretation of these differences associated with homosexuality is that
they result from hormonal changes that occurred during embryonic life that
affected, in parallel, the sexual orientation and the development of characteris-
tics normally differentiated between men and women. A real test of this causal
interpretation would be to manipulate a group of randomly selected human
embryos with steroid hormones and to analyze their sexual orientation 20 years
later. This experience is of course morally proscribed. Nevertheless, there are
various conditions described in detail in Chapter 6 that allow us to understand
the impact on the sexual orientation of spontaneous hormonal changes in the
embryonic environment in humans. Many studies have been devoted to this
topic.
These variations are often called “invoked experiences,” but they are in fact
only pseudo-experiments because they do not affect individuals randomly, and
one can never be sure that the hormonal variable that is studied is the only
change compared to a group of control subjects and is therefore responsible for
the observed effects on sexual orientation. This being said, given the ethical lim-
itations associated with experimentation in this area, these pseudo-experiments
Sexual Orientation in Clinical Cases 131
probably represent the best possible source of information. The information that
was collected using this approach will be considered here in detail.
EFFECTS OF PRENATAL STRESS

The first hormonal theory of homosexuality was developed by Günter Dörner,
already referenced in this book several times. It was based on the results of
animal studies conducted by Ingeborg and Byron Ward and their colleagues
from Villanova University in Pennsylvania, showing that when pregnant rats are
stressed (immobilization in a highly illuminated area), the sexual differentiation
of young males from these mothers is significantly affected (Ward, 1972).
In rats, the distance between the genitals and the anus is smaller in females than
in males. However, males born from these stressed pregnant rats have at birth
a shorter distance between the base of the penis and anus than normal males.
In adulthood, the males show atypical sexual behavior: they mount females less
frequently and are even capable of presenting female-typical behaviors such as
lordosis in response to sexual advances of other males (Ward & Ward, 1985). The
volume of their sexually dimorphic nucleus of the preoptic area (SDN) is also
lower than normal and thus closer to values typical of females (Anderson et al.,
1985; Kerchner & Ward, 1992).
All these changes clearly reflect a partial lack of masculinization and defemini-
zation. The analysis of the endocrine status of these embryos under stress has also
revealed that their blood level of testosterone was reduced and, in addition, that
the aromatase activity (conversion of testosterone into estradiol) was inhibited in
their preoptic area (Weisz, 1983; Ward, 1984; Jimbo et al., 1998). These hormonal
changes induced by stress and the associated increase in circulating levels of cor-
ticosterone were clearly responsible for their incomplete sexual differentiation
(see Chapter 3).
Based on these results, and given that male homosexuality could be primarily
considered as a lack of masculinization (like a heterosexual woman, a gay man
is attracted to men), Dörner hypothesized that gay men should be born to moth-
ers who were stressed during their pregnancy. He then performed retrospective
studies by interview to research whether the mothers of homosexuals had expe-
rienced stressful events during their pregnancy with a higher frequency than
mothers of heterosexual control subjects. The data collected showed that there
was a significant peak in frequency of gay men among cohorts of boys born in
Berlin between 1942 and 1946 (Dörner, 1980; Dörner et al., 1980; Dörner et al.,
1983). This increase did not appear be linked to an increased reporting and
detection of homosexuals linked to the appearance around 1970 of a more toler-
ant attitude toward homosexuality. Indeed, this peak is transient and is specific
to pregnancies having taken place during World War II.
Mothers interviewed in this study lived in Berlin. It is thus easy to imagine

that many of them had experienced extremely stressful events during pregnancy.
A more detailed analysis of the questionnaires also indicated a relationship
between highly stressful life events that mothers remembered and the likelihood
that their boy born in this period be homosexual. Less than 10% of mothers of
heterosexual boys had been exposed to stressful events, but more than 30% of
mothers of homosexuals had been placed in such situations. Mothers of bisexual
men occupied an intermediate position (Dörner et al., 1983). These data there-
fore suggest that stress experienced during pregnancy may increase the likeli-
hood of homosexual orientation in males, probably due to an interference of
corticosteroids (stress hormones) with the production or action of androgens or
their estrogenic metabolites as described in the rat.
Animal studies by Ward and colleagues had not shown an effect of stress
on sexual orientation but rather on the type of behavior (male or female) per-
formed by the subjects. This was declared by some to contradict Dörner’s results
and conclusions. As I have explained before, these are two different dimensions
of sexuality that should not be confounded. Since that time it was also shown
that (1) the sexual orientation of rodents is controlled by hormonal embryonic
mechanisms similar to those that control the expression of male-typical or
female-typical behavior (see Chapter 4), and (2) that maternal stress affects the
volume of the sexually dimorphic nucleus of the preoptic area and this nucleus
seems to be involved in the control of male sexual orientation (Paredes et al.,
1998). These data thus gave new importance to the observations of Dörner.
However, more recent studies have failed to reproduce the correlation between
maternal stress and homosexual behavior observed by Dörner (Schmidt &
Clement, 1990; Bailey et al., 1991) or they only produced equivocal data (Ellis
et al., 1988). We can wonder, however, whether the importance of stress experi-
enced by mothers during these more recent studies is really comparable to the
stress that mothers were exposed to while living in Berlin during the Second
World War during the intensive bombing of the city. In the present state of knowl-
edge, the theory of prenatal stress as a factor to explain homosexuality does not
seem firmly established in humans, even if it is applicable to the rat. These two
species could obviously react differently to stressful situations from the endo-
crine and/or behavioral point of view. Additional studies would certainly be
useful.
ANALYSIS OF WOMEN WITH ADRENAL HYPERPLASIA

Another source of information on the potential role of embryonic androgens in
the control of sexual orientation may be found potentially in the study of girls with
congenital hyperplasia of the adrenal glands (CAH) who have been exposed
in utero to abnormally high levels of androgens (see Chapter 6). We have already
seen that these girls display a masculinization of certain character traits during
childhood, such as the type of toys used or drawings made freely (Berenbaum
& Snyder, 1995; Berenbaum et al., 2000; Iijima et al., 2001). They also show a
high level of physical activity, which is typical of boys.
There is an increase in CAH women of the probability of commitment or desire
to engage in a homosexual relationship in comparison to a population of control
females or to unaffected sisters of these androgenized women (Money et al., 1984;
Dittmann et al., 1992; Zucker et al., 1996). Although there is an incidence of
female homosexuality of 10–12% in a control population (Kinsey et al., 1953), or
somewhat below 5% (Mosher et al., 2005), a study by John Money and his col-
leagues reported an incidence of homosexuality or bisexuality of 37% in CAH
women (Figure 9.1). Recent studies have produced similar figures (Hines, 2006;
Meyer-Bahlburg, 2009; Meyer-Bahlburg et al., 2008). A review by Meyer-Bahlburg
and collaborators also presented a summary of 17 studies on the subject in which
most studies showed an increased incidence of homosexuality or bisexuality in
CAH girls.
The prenatal endocrine change corrected at birth is thus associated with a
significant reduction of conventional heterosexual orientation. The explanatory
interest of this medical condition is that postnatal treatment of the subjects
Sexual orientation in CAH women
40
Homosexual orientation (%)
30
20
10
0
Homo. CAH
(Kinsey)
Bi-/homo. Refused Hetero.
to answer
Figure 9.1 Effect of prenatal androgenization linked to the syndrome of congenital
adrenal hyperplasia (CAH) on sexual orientation. One can observe in affected women a
significant increase in the percentage of bisexual or homosexual orientation compared
with control subjects and a large number of subjects refusing to answer questions
relating to this topic (drawn from data in Money et al., 1984).
(as a girl) is expected to act in the opposite direction of the prenatal hormonal
influence (masculinization). The masculinization of a behavioral trait (female
homosexuality, or sexual attraction for women, a trait normally male) should
thus in theory result from prenatal hormonal effects.
There are two possible alternative explanations, however, to this increased
incidence of homosexual or bisexual orientation. First of all, the change of
sexual orientation could be caused indirectly by the masculinization of the
external genital structures. Although these were surgically corrected at birth,
they still do not have an ideal structure in some women and therefore allow
little or no penetrative heterosexual relationships. This could be one reason for
engaging in homosexual rather than heterosexual relationships. It is also pos-
sible, theoretically, that the increase in nonheterosexual behavior and attraction
in CAH women is induced indirectly by the reaction of the parents of the girl
with masculinized genital structures. Even though these genital structures are
surgically corrected at birth and adequate hormonal treatment is in place to
block the secretion of androgens, the parents may not educate their daughter in
the same way as an unaffected girl. However, one might reasonably assume that
if parents have a modified reaction toward CAH girls, they should intentionally
promote the feminine behavioral traits of these girls. Homosexual attraction in
women is a trait normally found in men, and therefore one should expect that
the action of parents would decrease this aspect of behavior. Since several stud-
ies have observed increased homosexual orientation in CAH girls, one would,
in this interpretation, be lead to believe that parental efforts systematically pro-
duce an effect opposite to the intended effect. This would be paradoxical, to say
the least.
Therefore, the most logical interpretation of the change of sexual orientation
observed is that it was caused by the action of prenatal androgens. Of course,
adrenal hyperplasia is a rare disease and cannot play a role in determining the
sexual orientation of the majority of lesbians. Also note that the effect of prenatal
androgens in CAH girls, even if it is very significant, has moderate amplitude.
Only about 30% of the population of affected girls is not strictly heterosexual,
and statistical calculation of the effect size of this difference (a measure of the
difference with a control population taking into account the difference between
means and the normal variation of this characteristic; see Chapter 5 for the def-
inition of that term) demonstrates that it is five times smaller than the effect of
the adrenal hyperplasia on children’s games, for example (Hines, 2006). That
said, the results nonetheless fit within the expected parameters.
The potential limits to the effects of embryonic androgens on sexual orienta-
tion might include the existence of limited periods of sensitivity to androgens
that overlap only partially with the times during which these hormones are
present in high concentrations in the blood (problem of time-response); the fact
that children’s games are more sensitive to androgens than sexual orientation
(problem of dose-response); or finally that the part of the sexual orientation
controlled by androgens is limited, and other prenatal or postnatal factors,
including environmental or social and parental influences, are required. There
is no way to discriminate between these interpretations at present.
TREATMENT OF PREGNANT MOTHERS WITH DES

Between 1939 and 1960, about two million pregnant women were treated with
diethylstilbestrol (DES) in Europe and the United States to prevent an unwanted
abortion. This treatment was not only ineffective but also damaging in that this
substance, with its estrogenic activity, produces a slightly elevated risk of cervi-
cal cancer in girls exposed during their embryonic life. Studies suggest that, in
addition, these girls have a greater likelihood of bisexuality or homosexuality
(Ehrhardt et al., 1985; Meyer-Bahlburg et al., 1995; Swaab, 2007). Thus, the first
of these studies showed that 24% of 30 women who were exposed to DES during
their embryonic life were classified on the Kinsey scale (see Chapter 1) to scores
of 2 to 6 (bisexuality or homosexuality), whereas none of 30 women used as
controls in these studies (matched on as many variables as possible) were classi-
fied as bisexual or homosexual (Kinsey scores between 2 and 6) (Figure 9.2).
Twelve of these girls had a sister not exposed to DES, and comparison of these
Sexual orientation in DES women
40
Homosexual orientation (%)
30
20
10
DES Ctrl DES Sisters

Figure 9.2 Effect of prenatal treatment with diethylstilbestrol (DES) on female sexual
orientation. One can observe in treated women a significant increase in the percentage
of individuals that are bisexual or homosexual as compared with control subjects. This
increase is also reflected in the comparison of the subpopulation of women exposed to
DES and their nonexposed sisters (drawn from data in Ehrhardt et al., 1985).
two populations also showed an increased incidence of bisexuality or homo-

sexuality in subjects exposed to DES in utero (5/12 vs. 1/12).
A study by Meyer-Bahlburg and colleagues in 1995 similarly showed an
increase of most measures of homosexual orientation (ideas associated with
masturbation, dreams, attraction, relationships) in a population of DES women
as compared with their controls, as well as in a subpopulation of sisters of whom
only one of the two women had been exposed to DES before birth. In contrast,
a more recent study suggests that girls exposed to DES before birth have a
slightly lower probability (compared to control subjects) of having sex with a
same-sex partner (Titus-Ernstoff et al., 2003). The origin of this discrepancy is
unknown, but it could be related to the fact that this study considers only the
actual sexual activity, whereas previous studies also analyzed other aspects of
sexuality (fantasies, dreams) and used a graded scale (Kinsey scale in 7 points)
rather than a dichotomous classification (presence/absence of a homosexual
relationship).
If the effect of DES is real, which remains difficult to confirm given that this
treatment has now been abandoned for a long time and the exposed subjects
are becoming less and less available, this would indicate that estrogens as well
as androgens (testosterone) are able to masculinize sexual orientation. This
possibility is all the more likely given that in animals, testosterone exerts many
of its effects on sexual differentiation after conversion into estradiol by aro-
matase in the brain (see Chapters 3 and 4). Furthermore, the potential effects of
stress on human sexual differentiation described earlier in this chapter might
be induced by inhibition of brain aromatase, as is the case in rats. It should be
noted that studies conducted in rhesus monkeys and a number of data from
clinical studies indicate, however, that sexual differentiation of the brain in
humans would be under the direct influence of androgens (not aromatized),
which would make the effects of DES much more difficult to interpret.
DEFICIENCY IN 5 α -REDUCTASE
As we saw in Chapter 6, a genetic deficiency affects an enzyme called 5α-reductase
that converts testosterone into dihydrotestosterone (DHT). The patients are
exposed during fetal life to estrogen and androgen (testosterone) but not DHT.
At birth, affected boys have poorly masculinized or nonmasculinized external
genitalia, and they adopt a female gender identity in their childhood. Under the
influence of massive secretions of testosterone that occur during puberty, their
genitals are partially masculinized, and these children usually change their sexual
identity in adulthood to live as men. This change of identity and gender role has
even been observed in some individuals who had been married very young to a
man as a girl and who remarried with a woman after puberty.
These studies have been criticized because children deficient in 5α-reductase

are usually identified at birth. Terminology exists to name them in cultures
where they live in substantial numbers (Dominican Republic and Papua), and
one could therefore imagine that they are not really (fully) raised as girls, but
rather as intersex individuals because their family knows that masculinization
will occur at puberty (Rubin et al., 1981). Furthermore, it was also suggested
that the sex change occurring at puberty was motivated by the privileges granted
to men (especially in terms of independence, ability to undertake studies, etc.)
in these relatively “macho” societies (Rubin et al., 1981).
Imperato-McGinley and her colleagues have shown that among 18 subjects
from the Dominican Republic affected by that change who had completely
female external genitalia at birth and had therefore been brought up in the firm
belief that they were small girls, 17 have adopted a male sexual identity after
puberty and became sexually attracted to women (Imperato-McGinley et al.,
1991). The same authors have also reproduced these results in 5α-reductase-
deficient subjects in Papua New Guinea, which is a much less indulgent society
concerning the possibility of a change of gender identity and sexual orientation.
Also in Papua, most affected children who had been raised unambiguously
as girls changed their identity and sexual orientation after puberty (Imperato-
McGinley et al., 1991).
Of course, many questions remain open regarding these studies. If we imag-
ine that the change of identity and sexual orientation observed at puberty is the
result of effects of androgens during fetal life (organizing effects) and puberty
(activating effects) that are able to counteract the social experience and educa-
tion of all young children almost completely, then we are led to believe that
the action of testosterone on gender identity and sexual orientation is due to the
hormone itself or its estrogenic metabolites and that conversion to DHT, the
active metabolite at the peripheral level, is irrelevant for the behavioral responses.
Also, the idea that these children are truly and fully raised as girls will always be
impossible to prove. The fact remains that, as in the case of adrenal hyperplasia,
if education has a role, it always leads to the opposite effect of what is desired by
parents and thus also the opposite effect to be expected by those who say nurture
trumps nature.
CLOACAL EXSTROPHY
Cloacal exstrophy is a rare, complex, genito-urinary malformation occurring
during embryonic development that results in the birth of XY males who, in
addition to various malformations of the pelvis, have no penis. This is not an
endocrine disease, in that the testes are apparently normal both physically and
functionally. In many cases, the XY individuals are assigned at birth a female sex
both at the legal and social levels. They are also subjected to corrective surgery,
including removal of the testes and vaginoplasty. Several recent studies have fol-
lowed the psychosexual development of affected subjects, and one study observed
that in a significant number of cases (sometimes up to 55%, 8/14), adult subjects
chose to adopt a male identity and gender role (Reiner & Gearhart, 2004; Meyer-
Bahlburg, 2005). A clear masculinization of games played by these XY children
reared as girls was also present [preference for highly physical sports (football)
or aggressive sports (karate)] (Schober et al., 2002; Reiner & Gearhart, 2004).
In many cases, a typical male sexual orientation (attraction to women) was also
observed.
These data again suggest that hormonal imprinting of the embryo by andro-
gens could be responsible, at least in part, for the determinism of sexually dif-
ferentiated characteristics such as gender identity and sexual orientation. This
would result in a significant increase in the change of gender identity and sexual
orientation later in life in subjects exposed in utero to androgens, despite the
fact that they have subsequently been raised as girls. This change, however, does
not concern all individuals. If this type of change is consistent with a role of
prenatal androgens, it also indicates that androgens are probably not the only
determinant and the postnatal environment is potentially also involved. I will
come back to this idea.
OTHER ACCIDENTAL CHANGES IN THE EMBRYONIC

HORMONAL MILIEU
The four conditions affecting sexual orientation described above contribute to
the explanation of this feature because the assumed role of hormones goes against
the supposed role of education. It is therefore possible to differentiate these two
influences. There are also other endocrine changes that affect sexual orientation
and put it into conflict with the genetic sex of the affected individuals, but these
clinical cases are less interesting for the purpose of this work, because in these
cases, the effect of education is consistent with the effect of the endocrine disease
and therefore one cannot easily distinguish their respective roles.
Thus, males (XY) affected by complete androgen insensitivity are born, as
detailed in Chapter 6, with completely female genitalia. They are then raised as
girls and in general adopt a female gender identity and a female sexual orienta-
tion (they are sexually attracted by men). Detailed monitoring of a group of these
patients indicated that they were very satisfied with their sexual identity and
their sex life in general (Wisniewski et al., 2000). However, it is impossible in
these cases to know whether the gender identity and sexual orientation as women
are the result of the postnatal education or of the lack of action of testosterone
during embryonic life.
Similarly, various chromosomal abnormalities that have been discussed

previously (XXY or XYY boys and XO girls, see Chapter 6) provide little infor-
mation for understanding the control mechanisms of sexual orientation because
the missing or supernumerary chromosomes contain many genes whose absence
or presence could directly or indirectly influence sexual orientation. It is thus
impossible to link sexual orientation with any certainty to a specific genetic or
hormonal factor.
IN CONCLUSION
I have in this chapter reviewed a number of medical conditions that are all
related, at least in part, to sexual orientation at odds with the genetic sex of the
affected subjects. In several of these conditions, a sexual orientation was finally
adopted that was in opposition to the sex assigned at birth, but was consistent
with the type of presumed hormonal exposure during intrauterine life (attrac-
tion to men given absence of androgens and to women given presence of andro-
gens during embryogenesis). Most of the effects described here were reproduced
in independent studies on populations of unrelated subjects, which adds to their
credibility. It must be recognized that, occasionally, these effects could not be
replicated in other research. This is not necessarily surprising given the com-
plexity of these problems, including the selection of subjects and their controls
and the difficulties associated with an accurate determination of gender identity
and sexual orientation in subjects whose education and hormonal history were
potentially disturbed.
Taken together, these data suggest that embryonic hormones could, in humans
as in animals, play a significant role in determining sexual orientation. This con-
clusion is consistent with the results presented in the previous section indicating
that homosexual orientation is often associated with a change in physical, func-
tional, or behavioral traits supposed to differentiate themselves under the influ-
ence of prenatal sex steroids.
The most convincing medical cases suggesting a role of embryonic hormones
on sexual orientation are those in which hormonal effects and education are sup-
posed to pull in opposite directions, such as in congenital adrenal gland hyper-
trophy, prenatal DES exposure, 5α-reductase deficiency, and cloacal exstrophy.
As in any clinical study, it is impossible to be completely sure that the sex in
which subjects were educated was entirely consistent with what was desired.
It would be surprising that in all cases, parents have unintentionally encouraged
the development of a sexual attraction and sexual identity that would be con-
trary to their intentions (XX girls with adrenal hyperplasia raised as girls who
become attracted to girls, XY boys with 5α-reductase deficiency or cloacal exstro-
phy raised as girls but adopting male gender identity and male sexual orientation
in adulthood). The data therefore suggest a contribution of the hormonal embry-

onic milieu in agreement with what has been described in rodents.
However, it should also be noted that the medical problems discussed in this
chapter are not associated with a reversal of sexual orientation in all subjects,
except for androgen insensitivity syndrome (but in this case it is impossible to
separate the role of embryonic hormones from the role of education). In all stud-
ies of medical conditions where hormonal effects and education are supposed to
act in opposite directions, only a fraction of individuals (often 20–40%) show a
reversal of sexual orientation in agreement with the hormonal influence. One
should thus imagine that the prenatal hormonal environment only predisposes to
sexual orientations (and identities) but does not fully determine them. I discuss
this issue after considering the possible genetic influences.
10
A Genetic or Immunological Mechanism

Underlying Homosexuality?
If one accepts that homosexual orientation develops in whole or in part follow-

ing an abnormal exposure to sex steroids during embryonic life, the question
arises as to the origin of these anomalies. Three hypotheses have been advanced:
(1) lowered circulating levels of testosterone in the embryo could follow an
external event, such as an intense and chronic stress suffered by the pregnant
mother, which would prevent the complete masculinization of a male embryo
and thus induce homosexuality at adulthood, (2) random variations of hor-
mone concentrations could take place in which the extremes would be exposed
to concentrations sufficiently low (in men) or high (in women) as to induce a
disturbed phenotype, (3) genetic differences could induce abnormal circulating
levels of hormones, disruption of the brain’s response to these hormones, or a
direct effect on sexual orientation independently of any hormonal changes.
The first of these possibilities (prenatal stress) was considered in the previous
chapter, so I shall not return to it here. “Random” fluctuations of androgen
concentrations to which male or female embryos are exposed are also possible
(hypothesis 2). No biological parameter is actually adjusted to a specific value
and, on the contrary, they fluctuate around a mean value. One can therefore
imagine that, although male embryos are exposed to average concentrations of
testosterone that are greater than those found in female embryos, there is an
overlap between the values observed in the two sexes. In parallel, there may be
a minimum concentration of testosterone to establish a male typical sexual

orientation, i.e., a sexual attraction toward women (Figure 10.1). Under these
assumptions, it is obvious that individuals exposed to testosterone levels in
the intermediate zone of recovery of male and female concentrations could
logically adopt a homosexual orientation.
Even if we could draw blood to measure the concentrations of circulating
testosterone in a human embryo without a risk of abortion, we would still need
a clear idea of the critical moment when the differentiation of sexual orientation
takes place to predict when assays of testosterone concentrations in embryos
should be performed to predict their orientation, and this information is not
available. It is therefore extremely difficult to evaluate this theory based on
spontaneous fluctuations in concentrations of embryonic testosterone and to
determine to what extent it can explain the observed cases of homosexuality.
Anyway, a hormonal explanation could never be final because it always leads
to the next question, which is to identify the cause of unusual hormonal condi-
tion (“random” fluctuations within the physiological range considered in hypoth-
esis 2 or genetic change in hypothesis 3). Intense research efforts have therefore
been undertaken to attempt to identify individual genetic differences (muta-
tions or specific genes inherited from parents) that influence sexual orientation
in humans either directly or by the hormonal mechanisms described previously.
Attraction Attraction
for men for women
Threshold
concentration
Number of subjects
Women Men
Heterosexual Heterosexual
women men
Low High
(T Concentration during
the critical period)
Homosexual Homosexual
men women
Figure 10.1 Model showing how the fluctuations around the average concentration of
testosterone during embryonic life could lead to a homosexual orientation in a fraction
of the population.
A Genetic or Immunological Mechanism Underlying Homosexuality? 143
Given the complexity of the phenomenon studied, it is extremely likely that

sexual orientation is not determined in a strict causal manner by a single gene.
If this were the case, it is very likely that this gene would have been discovered
by now.
It should be noted that some genes have a marked influence on all aspects of
sexuality. Thus, certain recessive mutations in the gene controlling the synthesis
of the androgen receptor lead to a complete lack of masculinization of the
genetically male (XY) subjects who are affected (see Chapter 6). These individu-
als have female-typical external genitals, sexual identity, and sexual orientation.
This orientation is homosexual if it is defined in relation to genetic sex (male),
but it is in agreement with the physical and social sex and with the sex of educa-
tion. This condition of androgen insensitivity illustrates the importance of cer-
tain genes in determining sex, but it is not a model of homosexuality in the
conventional sense in which sexual orientation is in opposition to physical and
sexual identity (and often education).
One can easily imagine the existence of more subtle genetic mechanisms
(control of predispositions) depending on the interaction of several genes that
may nevertheless play an important role in determining sexual orientation as
such. Many studies indeed support the existence of such mechanisms.
THE CONCORDANCE OF SEXUAL ORIENTATION IN

MONOZYGOTIC AND DIZYGOTIC TWINS
A genetic contribution to homosexuality was initially postulated on the basis
that homosexuality tends to occur more frequently in some families than in
others. Gay men on average have more gay brothers than heterosexual men.
So in a given population, if a son is homosexual, between 20 and 25% of his
brothers will also be homosexual, whereas if a son is heterosexual, the probabil-
ity that his brothers are gay is only 4 to 6% (Diamond, 1993; Rahman & Wilson,
2003a). In parallel, lesbian girls have a higher probability (about 10% more) of
having a sister that is also homosexual than do heterosexual girls. This correla-
tion could of course result from psychosocial factors independent of genetics,
but studies of twins allow us to reject this interpretation.
Many studies have actually compared the concordance of sexual orientation
in matched pairs of twins. They all show that there is a much better concordance
of orientation in true monozygotic twins than in dizygotic twins (fraternal twins
born from different ova and sperm) (Figure 10.2).
The first study on this subject reported a rate of 100% concordance among
identical twins but only 10% among fraternal twins. In other words, all monozy-
gotic gay twins had a gay twin, but that was the case for only 10% of heterosexual
twins. The selection of subjects in this study was probably suspect, but several
Concordance of sexual orientation of twins
100 Monozygotic
Dizygotic
80
% of concordance
60
40
20
0
1 2 3 4 5 6 Mean
Studies
Figure 10.2 Percentage of concordance of homosexual orientation among true twins
(monozygotic) or false (dizygotic) twins observed in six independent studies involving a
total of 270 pairs of true and 271 pairs of false fraternal twins and average of the results.
Note that a concordance of 100% does not indicate the presence of 100% of homosexuals
in the population, but only that if in a twin pair one of the subjects was homosexual, the
other was as well (according to data in Diamond, 1993).
more recent studies using more sophisticated methods of selection have resulted
in a qualitatively similar conclusion showing a greater concordance of sexual
orientation between true than in fraternal twins. Taken together, these studies
indicate that 50–60% of the variance in sexual orientation in men is genetic in
the socio-cultural conditions typical of Western societies (LeVay & Hamer,
1994; Rahman & Wilson, 2003a; Swaab, 2007). These studies may still be influ-
enced by bias in the recruitment of subjects. Two recent very strict studies based
on very large populations of Australian and U.S. twins show a concordance
between monozyotic twins that is substantially lower, equal to about 30%.
A separate analysis of Australian data also shows a concordance of 26% for men
but 58% for women (see Rahman & Wilson, 2003a for more detail).
Whatever the precise value hiding behind these estimates, it is clear that all
studies show a better concordance in real than in fraternal twins. This difference
should be genetic. Identical twins in fact developed from the splitting of a single
fertilized egg and therefore share (in first approximation) exactly the same genetic
material, whereas dizygotic twins are not more similar genetically than siblings
born at different times. This genetic identity is likely to cause the better concor-
dance of sexual orientation. The only alternative explanation would be to assume
that the two types of twins are raised in a different way (that is partially the case)
and that these educational differences induce a better concordance in sexual
orientation. Given the lack of data demonstrating a role of education on sexual
orientation (Green, 1978), this interpretation seems unlikely at present. Absolute
demonstration of the genetic interpretation of these data could be obtained by
comparing two types of twins separated at birth, but due to the extreme rarity of
this combination of already minority events (frequency of twins, homosexuality,
separation at birth), this study will probably always remain impossible.
MATERNAL TRANSMISSION AND THE X Q 28 REGION

OF THE X CHROMOSOME
Studies of the potential genetic transmission of homosexuality through analysis
of genealogical trees show that sexual orientation in men tends to be transmitted
through the matriarchal line. In other words, a homosexual man has a higher
probability of having gay men among his ancestors on the maternal side, but not
on the paternal side. This type of inheritance may, of course, be explained in a
simple way if one or more genes contributing to the emergence of homosexuality
are located on the X chromosome (the only one inherited in a systematically dif-
ferent way from the father and mother). This maternal transmission was observed
in three independent studies, but was not found in a fourth one.
The situation for women’s sexual orientation seems more complex. Studies
have identified an increased rate of nonheterosexuality (the term used to group
homosexual and bisexual women) in girls, nieces, and cousins of the paternal
lineage of lesbians. Such a transmission could also be consistent with a link to
X chromosome, but it could come from the father as well as the mother. Other
interpretations are also possible, and the interpretation of these data remains
difficult (Rahman & Wilson, 2003a).
Molecular studies in gay men have subsequently attempted to identify the
region of the X chromosome that might contain the gene(s) potentially linked
to male homosexuality. On the basis of the genealogical tree of 114 families with
gay sons, Hamer and colleagues confirmed the higher incidence of homosexu-
als in the maternal line (Hamer et al., 1993). In 40 families selected on the basis
of containing two homosexual brothers and no indication of nonmaternal
inheritance, they showed a correlation between homosexual orientation and the
transmission of polymorphic markers located on chromosome X in more than
half of the subjects tested. A detailed statistical analysis of these associations has
finally shown a linkage with markers located in the subtelomeric region of the
long arm of the sex chromosome, a region called Xq28. This association with
the region Xq28 was found in two subsequent studies (Hu et al., 1995; Sanders
& Dawood, 2003) but not in a fourth one (Rice et al., 1999). It should be noted
100
Xq28 alleles
p=0.0001 Not shared
80 Shared
p=0.04 p=0.04 p=0.0001
60
Percentage
n.s.
40
20
0
Hamer Hu Sanders Rice Total
et al. et al. et al. et al. q28
1993 1995 2003 1999
Figure 10.3 Four independent studies indicate a link between the transmission of
male homosexuality and genetic markers located in the Xq28 region of chromosome X.
The diagram on the right shows the location at the subtelomeric end of this set of genes
on chromosome X (redrawn from Bocklandt & Vilain, 2007).
that this last study confirmed the link with maternal inheritance but had a lower
statistical power and was therefore less likely to identify a significant link with
Xq28 (Bocklandt & Vilain, 2007).
A comprehensive analysis of all available data, including negative results,
however, indicates that approximately 64% of gay brothers have common alleles
in the Xq28 region of chromosome X (Figure 10.3) (Bocklandt & Vilain, 2007).
This association is significantly associated with a probability of less than 0.0001,
which means there is less of a chance in ten thousand that the association
appeared by chance in the data and does not reflect a real connection. That
being said, it should be made immediately clear that this chromosomal region
remains quite broad and could potentially contain many genes. We are still far
from the identification of the responsible genes.
RESEARCH, SO FAR UNSUCCESSFUL, OF CANDIDATE GENES

To date, attempts to identify in the Xq28 region one or more genes that may
contribute to the determinism of sexual orientation have all been unsuccessful
(Bocklandt & Vilain, 2007; Ngun et al., 2011). Other approaches have been used
to try unravel the genetic mechanisms potentially involved. A recent study has
particularly been interested in the phenomenon of gene inactivation in the X
chromosome. The male cells indeed contain a single copy of the X chromosome,
while female cells contain two. So women (females) should theoretically pro-
duce a double amount of proteins whose genes are located on the X chromo-
some as compared to what is seen in men (male). To counteract this, each cell of
a female embryo inactivates randomly one of the two X chromosomes. This
inactivation is random in most cells, and it remains present in all cells that derive
from a given cell in the adult tissues: all derived cells have one or the other X
chromosome inactivated. Comparing this inactivation in mothers of homosex-
ual or heterosexual sons, Bockland and collaborators have recently shown that
inactivation showing an extreme asymmetry was present more frequently in the
first group (13 of 97 = 13% in mothers of homosexuals) than in the second (4 of
103 = 4% in mothers of heterosexuals) and even more frequently in mothers of
two or more gay boys (10/44 = 23%) (Bocklandt et al., 2006).
It is not known whether this unusual pattern of X chromosome inactivation
is partly responsible for the homosexual orientation of the sons or is a conse-
quence of an unidentified mechanism that directs this orientation. One possible
explanation would be that one or more genetic factors that influence sexual
orientation could also alter the survival of cells in the mother (white blood cells
or stem cells), leading to a selection of cells that inactivate the one or the other
allele. The fact remains that these data indicate that a mechanism that affects
male sexual orientation is visible in the blood of the mothers. The unusual inac-
tivation of the X chromosome among mothers of homosexuals also reinforces
the idea that this chromosome influences sexual orientation.
The influence of the X chromosome is only partial and it is therefore likely
that other genes located on nonsex chromosomes (autosomes) are also involved.
A gene-linking study taking into account the entire genome was undertaken
recently through advances in techniques of molecular biology (Mustanski et al.,
2005). Unlike the studies focused on the X chromosome that only considered
subjects with proven maternal transmission, this study did not use any criterion
for exclusion of individuals. By mathematical techniques too complex to explain
here, the study allowed for the identification of the maternal and paternal con-
tribution to the transmission of homosexuality by the measure of a score called
Lod (logarithm of odds). The largest parental genetic contributions to sexual
orientation have been identified on chromosome 7 (position 7q36, maternal
and paternal inheritance roughly equal) and chromosome 8 (position 8p12,
again equal inheritance from both parents). An exclusively maternal effect was
also identified on chromosome 10 (10q26). A similar study on a larger scale
(more than 1,000 gay and heterosexual controls) is under way to confirm these
results (see Bocklandt & Vilain, 2007; Ngun et al., 2011 for details).
The existence of a site of exclusively maternal heritability on chromosome

10 could lead to reinterpreting the results for the transmission of homosexual-
ity through the maternal line. These results were interpreted as reflecting a
transmission supported by the X chromosome, which is always inherited by a
boy from his mother. It is also possible, however, that the transmission from
mother to male child occurs by an autosome (chromosome 10 here) that would
be altered by a change that does not affect the sequence of DNA, only its level of
expression (called an epigenetic change). This involves changes (acetylation or
methylation) of DNA and of proteins (histones) that surround it and results in
the repression of the expression of certain genes. The repression is not necessar-
ily identical on both chromosomes of a pair, which could explain these cases of
exclusive maternal transmission through autosomes.
In an interesting way, this may also explain why there is not a complete con-
cordance between orientation of monozygotic (true) twins. Although these
true twins have in common all of their DNA, there might be differences in DNA
methylation leading to differences in gene expression during critical periods of
development (Fraga et al., 2005). The differences between monozygotic twins
that are commonly attributed to specific effects of the environment could in fact
be the result of a different epigenetic influence. This hypothesis opens up a new
extremely large field of investigations.
THE OLDER BROTHERS EFFECT

The variable that has been linked in the most solid manner to male homosexu-
ality is undoubtedly the number of older brothers born to the same mother.
Although the mechanism by which this occurs has not been formally identified
to date, it is likely that it involves at least in part a genetic contribution, and
I therefore consider it in this chapter.
For over 20 years, Ray Blanchard and his colleagues at the University of
Toronto and a few researchers working independently have shown that in
humans, there is a highly reproducible correlation between the number of older
brothers that a given subject (in this context, called the proband) has and the
probability that he is homosexual. This effect has been called the “older brothers
effect.” An analysis of 14 independent studies representing more than 10,000
subjects found that for each additional older brother an experimental subject
has, his probability of being gay increases by 33%. This does not of course mean
that 33% of boys who have a brother born before them are gay, but that the like-
lihood of developing this orientation is 33% higher than in the general popula-
tion. If we consider that the base percentage is 10%, the second son of a sibling
will have a 13.3% probability of being homosexual (10 + 33%), the third 17.6%
(13.3 + 33%), the fourth 23.4% (17.6 + 33%), etc. (Figure 10.4). The probability
Older brothers effects

80
% of homosexuals 70
60
50
40
0 1 2 3 ≥4
Number of older brothers
Figure 10.4 Relationship between the number of older brothers born to the same
mother and the percentage of gay men in a population of 302 gay men and 302
heterosexual matched subjects used as controls. Because the homosexual and
heterosexual subjects were matched for this study, there was 50% of homosexuals
in the population studied, which is of course much higher than what is found in
a normal population (redrawn from Blanchard & Bogaert, 1996).
that a fourth boy in a family is gay is thus about twice higher than for the first-
born boy (Blanchard, 1997; Blanchard, 2004). Such a correlation could have
many causes and interpretations. But given the robustness of the phenomenon
and because it has been known for many years, it was possible by statistical
analysis to exclude most of these alternative interpretations, and the authors of
these studies now focus on an immune theory (Bogaert & Skorska, 2011), which
I discuss below. Let us first consider all the most likely factors that have been
considered but do not appear to explain this correlation.
It has been shown that the effect of older brothers is not observed for younger
brothers (no increase in homosexuality, according to the number of brothers
born after the proband) nor for sisters whether they are born before or after the
proband. The effect does not depend on the number of brothers that were raised
at the same time as the proband (a social effect that would be derived from an
infancy spent with a lot of other boys) but only on the number of brothers pre-
viously born from the same mother (Bogaert, 2006). This effect is not found in
girls. There is no increase in female homosexuality associated with the number
of brothers or sisters born before (or after) the subject. All these data suggest a
specific link with the sex of the children and that the sequence of births rather
than the number of boys in a family is the relevant factor. This tends to show
that the cause is not an educational factor but rather a biological factor associ-
ated with multiple pregnancies. This notion is further strengthened by more
recent analyses showing that the effect of older brothers (born before) is still
observed if the subjects considered are raised in different families (because of
divorce and reconstituted families), but that this effect does not occur for half-
brothers (born to a different mother) or adopted brothers (Blanchard et al.,
2006). These data therefore exclude, as effectively as possible in a psychological
study in humans, the influence of postnatal educational effects of numerous
siblings.
Furthermore, we know that parental age affects some characteristics of chil-
dren, the best-known example being the rate of trisomy 21 (a genetic disorder
that is one of two causes of Down syndrome), which increases with parental
age. Since boys who have many brothers were born to mothers (and fathers) on
average older than subjects with fewer older brothers, one would also have
thought that the age of the parents was the explanatory factor. However, given
the large samples (more than 10,000 subjects available), detailed statistical anal-
ysis (by multiple correlations and partial regressions) could be used to reject
this idea.
Having rejected over the years all the alternative hypotheses that could explain
this effect, the only remaining explanation at present is based on a type of
immune response of the pregnant mother. This interpretation considers that the
mother who carries male embryos develops a progressive immune response
over successive pregnancies against these embryos, which are considered a for-
eign body that secretes foreign proteins in greater quantities than female embryos
(Blanchard & Bogaert, 1996; Blanchard & Klassen, 1997). This is the hypothesis
of progressive maternal immunization. The accumulation of antibodies is simi-
lar to that seen if a Rhesus-negative mother sequentially carries in her womb
several Rhesus-positive embryos. The first embryo is almost unaffected, but the
following ones are quite severely attacked by the maternal antibodies and must
often be prematurely removed from the uterus by Caesarean section or some-
times have to undergo an complete change of blood at birth to prevent damage
caused by maternal antibodies. The immune hypothesis to explain the effect of
older brothers assumes that mothers who had boys previously produce antibod-
ies against unidentified male proteins and that these antibodies affect the devel-
opment of certain aspects of the brain that are involved in determining sexual
orientation.
Consistent with this hypothesis, it was shown that the birth weight of boys
with older brothers is significantly lower than the weight of matched boys with
the same number of older sisters (Blanchard & Klassen, 1997; Blanchard, 2001).
Moreover, within this population of subjects with older brothers, gay men had a
lower body weight at birth than heterosexual men (about 6 ounces of difference,
which is low but significant in the sample). These differences may be explained
by the effects of hypothetical antibodies produced by the mother against the
antigens typical of the male.
To mediate such an immune effect on the development of sexual orientation,
a protein should theoretically fulfill a number of criteria (Bogaert & Skorska,
2011). Namely: (1) it should be expressed more or less specifically in the brain,
since sexual orientation is obviously controlled by the brain and homosexuality
is not associated with any major change in body structure or function, (2) it
should be expressed at the surface of brain cells in order to be accessible to
maternal antibodies, (3) it should be a protein specific to males, (4) fetal mate-
rial containing this protein should enter the maternal circulation and should
induce the formation of antibodies in the mother and (5) the concentration of
these antibodies in the maternal circulation should increase with the number of
male embryos that the mother has been exposed to. A small number of proteins
have been identified that fulfill most of but not all these criteria. However, there
is as of this date no formal proof that these proteins actually play a role in the
control of sexual orientation and in particular in the increased incidence of
homosexuality associated with the older brother effect. It is therefore premature
to discuss these proeins here, but the interested reader can conveniently find a
full discussion of this topic in a recent review (Bogaert & Skorska, 2011).
In summary, the mechanism underlying the effect of older brothers, and more
specifically its immune aspects, remains unknown to date. It is possible to imag-
ine an effect of the antibodies (which are known to penetrate the embryo) on the
expression of genes and their interaction with the environment that are indis-
pensable for the construction of an individual. However, there is no evidence of
interaction with a genetic system, and the possible immune effects could result
only from interactions with membrane proteins. Nevertheless, the effect of older
brothers is without any doubt an argument of great weight supporting the idea
that sexual orientation is significantly determined before birth.
EVOLUTIONARY SIGNIFICANCE OF HOMOSEXUALITY

Before closing the chapter, it is necessary to answer an objection frequently
made about genetic theories of homosexuality. According to this objection,
homosexuals do not reproduce, so if this orientation is controlled by one (or
more) gene(s), this (these) gene(s) should disappear over the generations.
But it is wrong to say that homosexuals do not reproduce. It is important to
recognize that some homosexuals change their apparent orientation during
their life for various reasons often linked to social pressures (they hide their
homosexuality until later in their life). During their apparent “heterosexual”
phase, they may get married and eventually have children.
However, it remains clear, as proven by quantitative studies, that homosexuals

reproduce less than heterosexuals. How is this lower reproductive success com-
patible with the maintenance of one or more genes controlling homosexuality?
Several possible answers have been made to this apparent contradiction (Rahman
& Wilson, 2003a; LeVay & Valente, 2006).
It is just possible that the gene or genes controlling homosexuality confers a
reproductive advantage yet unknown, which would offset the negative effect on
reproductive success of homosexuality as such. It is known that some genes
have negative effects when they are present in the homozygous state (that is, on
both chromosomes of a pair) but they confer a benefit on survival and thus
reproduction when present on a single chromosome of the pair (heterozygous).
The best-known example of this is sickle cell anemia, which causes a serious
disease of the blood in the homozygous state but is associated with a resistance
to malaria in the heterozygous state. Heterozygous individuals, because of the
rules of genetics, are significantly more frequent than homozygous individuals,
so that the gene for sickle cell anemia persists in the population, especially in
malaria-infested areas where the advantage for survival is important. A gene
favoring homosexuality could similarly be associated with an unidentified
reproductive advantage in the heterozygous state (where homosexuality would
not be apparent) but a clear decrease of the breeding success in the homozygous
state.
It is also important to note that male homosexuality can be interpreted as an
extreme attraction for males. If we imagine that the gene(s) controlling this trait
of character has (have) a similar effect in women, the deleterious effect on male
reproduction could be offset by an increased reproduction of women carrying
the gene. Being more attracted to men, they would have a more intense sexual
activity, resulting in a larger number of offspring. A fairly recent study showed
that relatives (mothers, grandmothers, aunts) on the maternal side of homo-
sexual men have more children than relatives of heterosexuals (Camperio-Ciani
et al., 2004). Furthermore, this difference is not found in the paternal lineage,
which is in complete agreement with the genetic studies described above that
indicate that homosexuality would be inherited by genetic transmission from
the mother, not the father. This study also confirmed that homosexuals have
more gay relatives on the maternal than on the paternal side and that homo-
sexuals are more frequent among younger brothers of a family than among
older brothers (the older brothers effect).
The group theory of evolution (kin selection) also provides a hypothesis to
explain the persistence of genes for homosexuality. Homosexual individuals who
have no children themselves could help their close relatives (brothers, sisters,
cousins) to raise their children who are genetically related. Under the more diffi-
cult conditions of survival that prevailed during human evolution, this additional
assistance could have favored the survival of children and thus contributed to the
transmission of genes promoting homosexuality. Many animal societies have
been identified in which some individuals do not reproduce, but simply help raise
the offspring of their close relatives. This reproduction, based on a few individuals
helped by others closely related to them genetically, is almost always observed in
difficult environmental conditions (e.g., deserts) and is expected to be maintained
because the nonbreeding individuals still spread their genes through the repro-
duction of their close relatives. This seemingly altruistic behavior would be still
“interested” from an evolutionary point of view.
It is also possible that genes controlling homosexuality are only harmful to
reproduction and therefore have a tendency to disappear on a continuous basis.
They would be maintained in humans by repeated mutations that would fre-
quently make them reappear in a more or less similar form. Sites on the chromo-
somes are known that are actually subject to very frequent mutations, but so far
none has been associated with homosexuality.
It is clear that as long as the gene or genes favoring homosexuality have not
been identified, it remains impossible to test these different interpretations. They
are presented here simply to show that the existence of a gene to a phenotypic trait
obviously associated with a decreased reproductive success is not inconceivable.
11
General Conclusions
At the end of this journey into the world of neurobiology in relation to sexual-
ity, it is important to summarize the accumulated knowledge and the questions
that remain unanswered, and then see how scientific information collected over
the past 20 or 30 years can and should probably change certain aspects of med-
ical practice and also, in a much more important manner, the general attitude
of society toward homosexuals (and also probably transsexuals).
WHAT CONTEMPORARY NEUROBIOLOGY TEACHES US

A dozen important points should be recalled to understand the real scope of
current knowledge.
In Animals
1. The action of hormones on the hypothalamus and preoptic area in
rodents and primates precisely determines the type of sexual
behavior (male or female) expressed by an individual. This
determination is largely independent of the genetic sex of the
animal. It is controlled by the type of hormonal imprinting
experienced during embryonic and/or immediately postpartum life
General Conclusions 155
in interaction with the hormones that are secreted by the testes or

ovaries in adulthood.
2. The embryonic sex steroids also differentiate the size of several brain
structures, including the SDN-POA and, once acquired, this size is
typical of one sex and cannot be changed in adulthood by steroid
hormones.
3. Male sexual orientation is determined by the action of sex steroids
in the preoptic area. It is modified (inverted) by hormonal
treatments during early development in young animals (the same
treatments that are changing the type of behaviors that will be
exhibited later), and thereafter orientation seems to be a stable
characteristic of the individual. This orientation can be manipulated
at will by injections of steroids in a developing individual, but not in
adulthood.
4. The only case of spontaneous strict homosexuality in animals is
observed in sheep. Homosexual male sheep have a sexually dimorphic
nucleus of the preoptic area the same size as that of a female. The size
of this nucleus is determined by the action of testosterone during
embryonic life.
In Humans
5. The sex steroids and their receptors in the human brain are similar
or identical to what they are in animals. One can find in the human
species all the embryonic effects of sex steroids on genital structures
that are identified in animals. Effects of steroids are also present in
the activation of sexual behavior even though they are, as might be
expected given the complexity of human sexuality, more nuanced
than in animals.
6. The sexual differentiation of genital structures during embryonic life
precedes by several months the differentiation of the brain, the organ
controlling behavior. It is therefore possible that hormonal changes
occurring between these two embryonic periods can induce
discrepancies between physical sex and aspects of sexual behavior
that are sexually differentiated (orientation, gender identity).
Such discrepancies could also originate in a differential sensitivity
of responses to steroids (dose-response difference) or in the
existence of different sensitive periods to the action of steroids.
7. Homosexuality is not simply a different sexual orientation; it is
accompanied by complex physical, functional, and behavioral
changes. Homosexuality therefore affects not only a particular aspect
of sexual behavior but also multiple sexually differentiated traits

that are not related to sexuality. Male (female) homosexuality is a
complex change, often in a female (male) direction, of multiple
features that are sexually differentiated. These characteristics
include variables that possibly could be secondarily affected by
homosexuality (e.g., responses in cognitive tests) but also physical
(ratio of length of fingers D2:D4) or functional (oto-acoustic
emissions produced by the inner ear) variables. It is hard to see how
these variables could be influenced by sexual orientation.
8. A physical difference associated with homosexuality of particular
interest is the difference that affects the size of the sexually
dimorphic nucleus of preoptic area (SDN-POA). It is larger in
heterosexual men than in women and that of male homosexuals
is the size of a female’s. The mechanism that controls the
development of this nucleus in humans is unknown, but it does
not seem to significantly depend on the hormonal status in
adulthood. In rats and sheep, the size of the nucleus is determined
irreversibly by the action of embryonic sex steroids, and its lesion
in adult male rats changes sexual orientation.
9. Many diseases that affect the functioning of the endocrine system
during fetal life are associated with more or less profound changes
of sexual orientation in men and women. In a limited number of these
pathologies, children are raised as if they belonged to a sex opposed to
the potential hormonal influence they have been exposed to in utero.
A substantial proportion of subjects exposed to such “experiments of
nature” or pseudo-experiments develop a sexual orientation (and
sometimes a sexual identity) in opposition to the sex of the education
they have received. This suggests an important role of embryonic
hormones in determining these behavioral characteristics.
10. Various arguments, including many studies of the sexual orientation
of twins, clearly indicate a genetic contribution (but not an absolute
determinism) of sexual orientation. If one twin is homosexual, the
probability that the second is also gay is much higher in “real” twins
originating from one egg and thus sharing the same genetic material
than among the “false” fraternal twins.
11. The genetic contribution to male homosexuality is apparently
inherited preferentially through the maternal line. This observation
led to the identification of a region of the X chromosome whose
variability is significantly associated with sexual orientation. This
association has been identified several times independently, but to
date one or more genes linked to homosexuality still have not been
isolated. The genetic contribution to sexual orientation is most likely

multigenic and therefore very difficult to identify.
12. The most reliable factor identified for the development of homo-
sexuality in the male is the presence in his family of older brothers
born to the same mother. In these circumstances, the likelihood of
homosexuality increases by 33% for each older brother and is
accompanied by a small but statistically significant decrease in weight
at birth. These effects do not appear to be explained by differences in
education or family and are presumably the result of the accumulation
in the mother, during successive pregnancies, of antibodies against one
or more proteins expressed specifically by boys but not yet identified.
Taken together, all these data thus strongly suggest that hormonal, genetic,
and possibly immunological factors acting to a large extent during the embryonic
life or during early infancy play an import role in the determination of sexual
orientation.
THERE IS NO PSYCHOANALYTIC, PSYCHOLOGICAL, OR

SOCIOLOGICAL EXPLANATION OF HOMOSEXUALITY
In contrast, the alternative explanations of homosexuality that are based on
psychoanalysis, psychology, or sociology are perhaps attractive at first glance,
but the rare quantitative studies that were conducted to test them provide,
to our knowledge, no support. Psychoanalytic interpretations are mainly based
on series of anecdotes related to the analysis performed by Sigmund Freud
himself, whose scientific honesty was seriously questioned (Van Rillaer, 1980;
Bénesteau, 2002; Dufresne, 2007).
It seems well established that the sexual and social experiences of early child-
hood and adolescence have little or no effect on the development of homosexu-
ality. Theories of homosexuality derived from psychoanalysis, behaviorism, or
social constructivism, attributing a major role to early sexual experiences or
relationships with parents, did not find any support in controlled scientific
studies and are in fact at odds with many facts of observation. For example,
children raised by homosexual couples are not more likely to become homo-
sexual than children raised by heterosexual couples. The absence of the father
has no effect on the occurrence of male homosexuality (no increase among
children raised by a single mother), and the frequency of homosexuality is not
larger (it is in fact similar in all human societies) in some societies in New
Guinea, where homosexual experiences are the rule among the adolescent boys,
or in boys who have been raised in boarding schools where homosexual rela-
tions were present on a regular basis during adolescence.
These negative results do not necessarily imply the complete absence of influ-
ence of the postnatal environment on the development of sexual orientation,
but available data strongly suggest that the environment alone cannot deter-
mine homosexuality or heterosexuality, though it is quite conceivable that it
may interact with prenatal biological factors that I have outlined to allow their
full expression. Studies that could potentially identify such interactions should
probably be based on extremely large numbers of subjects, given the low ampli-
tude of the effects studied.
WHAT ARE THE KEY POINTS THAT SHOULD BE REMEMBERED?

Even taking into account the limitations described in earlier chapters, it seems
undeniable that the balance between biological and cultural factors that may
explain homosexuality very much favors the biological factors acting predomi-
nantly during embryonic life. Human (and animal) homosexuality is the result
of an interaction between hormonal and genetic embryonic factors with per-
haps a minor contribution of postnatal social and sexual experiences.
Studies on twins show a significant genetic contribution that could explain
at least 50% of the variance in sexual orientation in Western societies. Various
prenatal endocrine disorders affect orientation, but in no known case is there
is percentage of induced homosexuality greater than about 30%. Finally, the
observed associations between homosexuality and the modification of sexually
differentiated characteristics are often statistically significant, but they are also
related to a definite level of variability. Taking the size of the SDN of the preop-
tic area once again as an example, this nucleus is twice smaller on average in
homosexuals than in heterosexuals. However, at the individual level there are
heterosexuals who have an SDN whose size is close to the gay average, and con-
versely there are homosexuals who have a relatively large SDN that therefore
falls within the normal distribution for heterosexuals. The association between
size of the SDN and sexual orientation is statistically validated, but it does not
apply for each individual.
It is thus clear that none of the biological factors identified to date is able
by itself to explain homosexuality. Three potential explanations are therefore
possible. Either there are different types of homosexuality—some have a genetic
origin, others a hormonal origin, still others result from the older brothers
effect or from biological factors not yet identified—or the effects of different
biological factors that have been identified interact with each other in a variable
manner in each individual and it is only when several of these predisposing
factors are combined that homosexual orientation is observed, or finally, all
the biological factors that I have described only produce a predisposition to
become homosexual, and these predispositions can only develop in a specific
set of psychosocial contexts that are not yet identified. But if this postnatal con-
text is actually an important permissive factor, it is surprising that a quantitative
study has been unable so far to identify aspects of the environment that are
limiting.
Current knowledge does not allow us to discriminate between these interpre-
tations. However, it is clear that biological factors acting during prenatal life
play a significant role in determining sexual orientation and that homosexuality
is not, for most people, a choice of life. This orientation is often or always
an evidence that imposes itself on the individual during his or her teens or life
as a young adult. The recognition of a nonconventional sexual orientation is
very often the occasion of significant psychological suffering. Remember the
suicide rate that is three times higher during adolescence among homosexuals
as compared to the general population. By contrast, the heterosexual orienta-
tion develops spontaneously, often while the individual does not truly realize it.
It is not a matter of choice here. One does not choose to be homosexual any
more than one chooses to be heterosexual. We can choose to accept this orienta-
tion, to act accordingly, and to reveal it or not to the society, but the orientation
itself is not in any way a deliberate choice. Given the complexity of the human
person, this does not mean of course that there is not a minority of gay men for
whom this sexual orientation is a choice of life influenced by past experiences
and/or motivated by various reasons ranging from curiosity to perversity.
There are probably sex perverts among homosexuals, just as there are among
heterosexuals, but homosexuality itself is not a perversion. A large proportion
of homosexuals are born with that sexual orientation, which is revealed to them
in a very progressive way during development and is often accepted at the price
of a significant psychological distress. It is for most of them not a choice (how
could a homosexual change by personal choice his D2:D4 ratio or the function-
ing of his or her inner ear?). Homosexuality is due neither to a perversity nor to
inadequate parents. It is a biological variation of a complex behavioral trait
whose control is obviously multifactorial.
WHAT ARE THE CONSEQUENCES?

It was thought for a long time that the young child is a tabula rasa and that it
is possible through education to impose a sexual identity and sexual orientation
independent of the genetic sex (Money & Ehrhardt, 1972). The exemplary
case of John/Joan and many other studies that were reviewed clearly suggest
that this is not the case and that sexual identity and sexual orientation are,
at least in part, determined before birth. The clearest demonstration of this pre-
natal control is provided by the occurrence of cloacal exstrophy. It seems that
the hormonal embryonic imprinting or possibly the genetic sex itself have an
influence on gender identity and sexual orientation. The results of animal

experiments associated with correlations derived from human clinical data
thus suggest that people are born gay, they do not become homosexual, con-
trary to a widespread idea. This finding has important implications in two quite
different areas.
UNDERSTANDING DOES NOT MEAN ACTING: THE EUGENIC RISK

It is important to clarify that the purpose of science in general and this book in
particular is not to identify the biological mechanisms that control the emer-
gence of homosexuality in order to control this aspect of human sexuality. The
goal here is simply to understand the functioning of the animal and human
brain and advance the objective knowledge of human nature. I believe that a
better understanding of the mechanisms that control sexual orientation (homo-
sexual as well as heterosexual) should have positive effects on how society views
homosexuals and should help their parents understand that they are not respon-
sible for the sexual orientation of their child.
At the same time we must not ignore the consequences of this research and
we must seriously ask ourselves whether knowledge of these mechanisms would
be likely to lead to eugenic abuses, which could either propose abortion of
embryos for which there is a increased risk of homosexuality or in the future
propose treatment that would potentially change this prenatal predisposition.
This risk is of course present as soon as a society is exposed to ethical drifts and
no longer controls the use of its science and technology. Several remarks are in
order at this level, however.
First, it should be noted that even if we now have converging information
strongly supporting the idea that homosexuality is controlled by prenatal hor-
monal and/or genetic factors, we are still far from understanding the details of
these mechanisms to thus be able to modify these processes to obtain a specific
result. It is likely that this relative ignorance will persist for many years and we
will therefore not be able to “choose” the desired sexual orientation for a future
child.
Furthermore, for ethically deviant societies (e.g., Nazi or communist coun-
tries), an understanding of the determinism of homosexuality is not necessary
and was not necessary in the past to persecute homosexuals and even some-
times condemn them to death. That is still the case in countries such as Iran or
Afghanistan. Societies as well as social, religious, or medical groups who want
to hurt homosexuals do not need to understand the mechanisms that control
this sexual orientation to combat homosexuality. Their ignorance is sufficient,
and if the truth is likely to prevent the deeds of such extremist groups, it cannot
support them.
It also should be noted that understanding does not necessarily mean want-
ing to control. Understanding is part of science; the choice of the use of knowl-
edge is a question of morality and politics. It is very important that scientists be
involved in the political choices that are made, particularly when they concern
the use of science and technology, but the two aspects of human activity should
not be confused. It would be foolish to abandon the pursuit of knowledge
because it could later be misused. If this were the case, then we would have had
to abandon the discovery of the structure of the atom. This knowledge has
brought significant benefits (radiotherapy, an enormous source of energy) as
well as major disasters (the atomic bomb). It is up to societies to decide what
they do with scientific knowledge. In our view, understanding the biological
mechanisms that control human behavior should have more positive than neg-
ative effects on the happiness of the human species.
THE ATTITUDE OF SOCIETY TOWARD HOMOSEXUALS

SHOULD (STILL) CHANGE
Throughout history, homosexuals have been persecuted for many reasons and
homosexuality was considered a sin, a disease, or a perversion. This type of
opinion has changed during the 20th century, at least for a large number of
people in Europe and the United States. In these countries many homosexuals
are now free to display their sexual orientation, even to marry and adopt chil-
dren. However, the fact remains that this sexual orientation is often difficult to
assume and is associated in the mind of many people, even those with a high
degree of education, with notions of deviance, perversion, and guilt. Many par-
ents are also affected by a deep sense of guilt because they mistakenly think they
somehow should have provided different conditions for a “normal” develop-
ment of their child.
Biological studies presented in this work, even though they do not identify
the mechanisms that lead to homosexuality in a formal manner, nonethe-
less clearly show that homosexuality is not, in general, a free choice of life
and that any role of parents in its appearance can be only very marginal and
may not even exist at all. The fact that homosexuality is statistically associated
with physical (the D2:D4 ratio of finger sizes) or functional (oto-acoustic emis-
sions) differences shows that this is not a choice. Homosexuality should there-
fore be regarded as a spontaneous variation of a biological character. This
character (sexual orientation) and its determinism are complex, but there is no
reason to believe a priori that the status of this character is different from that
of any other physical/functional trait such as height or hair color. Just as one
does not choose to be left-handed or right-handed, one cannot choose to be
homosexual or heterosexual.
Finally, we must recognize that there are homosexuals who claim their
lifestyle and say, contrary to the argument presented in this book, that their
sexual orientation reflects a free choice of life. I respect this attitude and give
everyone the right to do what he or she wants with the data presented here. The
level of biological determinism that has been identified to date remains impre-
cise and can leave a place for individual choice. Large interindividual variations
in the determinism of homosexuality probably exist and homosexuality is not,
in all probability, a uniform phenomenon. If some think that they are homo-
sexual by choice, there is no reason to try to dissuade them, and it is possible
that they are right. In many other cases, however, homosexuality is seen as
a “defect” or a perversion that is in conflict with deep religious or moral convic-
tions. For many homosexuals, homosexuality was never an option but a differ-
ence difficult to accept and a challenge for social integration. To all these people,
we should be able to say that the difference is essentially the same as the differ-
ence that affects people in relation to their stature, eye color, or hair color.
Determinism of sexual orientation is simply more complex than that of these
physical characteristics.
If we accept these premises, it becomes apparent that homosexuals and their
parents are not responsible for their sexual orientation and there is no objective
reason to reject them. Homosexuals are not perverse (at least not more than
straight people), they are not dangerous (homosexuality is not “contagious”),
and they are generally not responsible for their condition. They should be able
to live their lives according to their nature (biology) without worrying about
issues of personal guilt. In return, society should, as it is increasingly but not yet
uniformly doing, accept them without any form of discrimination.
REFERENCES
Abdelgadir, S.E., Roselli, C.E., Choate, J.V. & Resko, J.A. (1999) Androgen receptor
messenger ribonucleic acid in brains and pituitaries of male rhesus monkeys: studies
on distribution, hormonal control, and relationship to luteinizing hormone secre-
tion. Biol Reprod, 60, 1251–1256.
Adkins-Regan, E. (1999) Testosterone increases singing and aggression but not male-
typical sexual partner preference in early estrogen treated female zebra finches.
Horm. Behav., 35, 63–70.
Adkins-Regan, E. (2011) Neuroendocrine contributions to sexual partner preference in
birds. Front Neuroendocrinol, 32, 155–163.
Adkins-Regan, E. & Ascenzi, M. (1987) Social and sexual behaviour of male and female
zebra finches treated with oestradiol during the nestling period. Anim. Behav., 35,
1100–1112.
Adkins-Regan, E. & Wade, J. (2001) Masculinized sexual partner preference in female
zebra finches with sex-reversed gonads. Horm. Behav., 39, 22–28.
Adkins-Regan, E., Yang, S. & Mansukhani, V. (1996) Behavior of male and female zebra
finches treated with an estrogen synthesis inhibitor as nestlings. Behaviour, 133,
847–862.
Agmo, A. (2007) Functional and dysfunctional sexual behavior. A synthesis of neurosci-
ence and comparative psychology. Elsevier, Amsterdam.
Ahmed, E.I., Zehr, J.L., Schulz, K.M., Lorenz, B.H., DonCarlos, L.L. & Sisk, C.L. (2008)
Pubertal hormones modulate the addition of new cells to sexually dimorphic brain
regions. Nat. Neurosci., 11, 995–997.
Alexander, B.M., Rose, J.D., Stellflug, J.N., Fitzgerald, J.A. & Moss, G.E. (1999) Behavior
and endocrine changes in high-performing, low-performing, and male-oriented
domestic rams following exposure to rams and ewes in estrus when copulation is
precluded. J. Anim. Sci., 77, 1869–1874.
Alexander, G.M. & Hines, M. (2002) Sex diferences in response to children’s toys in non-
human primates (Cercopithecus aethios sabaeus). Evolution and Human Behavior,
23, 467–479.
Allen, L.S. & Gorski, R.A. (1992) Sexual orientation and the size of the anterior
commissure in the human brain. Proc. Natl. Acad. Sci. USA, 89, 7199–7202.
164 References
Allen, L.S., Hines, M., Shryne, J.E. & Gorski, R.A. (1989) Two sexually dimorphic
cell groups in the human brain. J. Neurosci, 9, 497–506.
Anderson, D.K., Rhees, R.W. & Fleming, D.E. (1985) Effects of prenatal stress on
differentiation of the sexually dimorphic nucleus of the preoptic area (SDN-POA) of
the rat brain. Brain Res., 332, 113–118.
Arnold, A.P. & Chen, X. (2009) What does the “four core genotypes” mouse model
tell us about sex differences in the brain and other tissues? Front Neuroendocrinol,
30, 1–9.
Arnold, A.P. & Gorski, R.A. (1984) Gonadal steroid induction of structural sex differ-
ences in the central nervous system. Ann. Rev. Neurosci., 7, 413–442.
Arnold, A.P., Xu, J., Grisham, W., Chen, X.Q., Kim, Y.H. & Itoh, Y. (2004) Minireview:
Sex chromosomes and brain sexual differentiation. Endocrinology, 145, 1057–1062.
Bagatell, C.J., Heiman, J.R., Rivier, J.E. & Bremner, W.J. (1994) Effects of endogenous
testosterone and estradiol on sexual behavior in normal young men. J. Clin. Endocrinol.
Metab., 78, 711–716.
Bagemihl, B. (1999) Biological exuberance. Animal homosexuality and natural diversity.
St. Martin’s Press, New York.
Bailey, J.M., Willerman, L. & Parks, C. (1991) A test of the maternal stress theory of
human male homosexuality. Arch Sex. Behav., 20, 277–293.
Bakker, J., Baum, M.J. & Slob, A.K. (1996a) Neonatal inhibition of brain estrogen
synthesis alters adult neural Fos responses to mating and pheromonal stimulation in
the male rat. Neuroscience, 74, 251–260.
Bakker, J., Brand, T., van Ophemert, J. & Slob, A.K. (1993a) Hormonal regulation of
adult partner preference behavior in neonatally ATD-treated male rats. Behav.
Neurosci., 107, 480–487.
Bakker, J. & Brock, O. (2010) Early oestrogens in shaping reproductive networks:
evidence for a potential organisational role of oestradiol in female brain develop-
ment. J. Neuroendocrinol, 22, 728–735.
Bakker, J., De Mees, C., Douhard, Q., Balthazart, J., Gabant, P., Szpirer, J. & Szpirer, C.
(2006) Alpha-fetoprotein protects the developing female mouse brain from mascu-
linization and defeminization by estrogens. Nat Neurosci., 9, 220–226.
Bakker, J., Honda, S., Harada, N. & Balthazart, J. (2002a) Sexual partner preference
requires a functional aromatase (cyp19) gene in male mice. Horm. Behav., 42,
158–171.
Bakker, J., Honda, S.I., Harada, N. & Balthazart, J. (2002b) The aromatase knock-out
mouse provides new evidence that estradiol is required during development in the
female for the expression of sociosexual behaviors in adulthood. J. Neurosci., 22,
9104–9112.
Bakker, J., Van Ophemert, J. & Slob, A.K. (1993b) Organization of partner preference
and sexual behavior and its nocturnal rhythmicity in male rats. Behav. Neurosci., 107,
1049–1058.
Bakker, J., Van Ophemert, J. & Slob, A.K. (1996b) Sexual differentiation of odor and
partner preference in the rat. Physiol. Behav., 60, 489–494.
Balthazart, J. & Ball, G.F. (2006) Is brain estradiol a hormone or a neurotransmitter?
Trends Neurosci., 29, 241–249.
References 165
Balthazart, J. & Ball, G.F. (2007) Topography in the preoptic region: differential regula-
tion of appetitive and consummatory male sexual behaviors. Front Neuroendocrinol,
28, 161–178.
Balthazart, J., Cornil, C.A., Charlier, T.D., Taziaux, M. & Ball, G.F. (2009) Estradiol,
a key endocrine signal in the sexual differentiation and activation of reproductive
behavior in quail. J Exp Zool Part A Ecol Genet Physiol. 311, 323–345.
Bancroft, J. (1995) The pharmacology of sexual function and dysfunction. Elsevier
Science, Amsterdam.
Baron-Cohen, S. (2004) The Essential Difference: Men, Women and the Extreme Male
Brain. Penguin Press Science, London.
Baron-Cohen, S. (2006) Prenatal testosterone in mind. MIT Press, Cambridge, MA.
Barres., B.A. (2010) Neuro Nonsense. Plos Biology, 8, e1001005.
Baum, M.J., Carroll, R.S., Erskine, M.S. & Tobet, S.A. (1985) Neuroendocrine response
to estrogen and sexual orientation. Science, 230, 960–961.
Becker, J.B., Berkley, K.J., Geary, N., Hampson, E., Herman, J.P. & Young, E.A. (2008)
Sex differences in the brain. From Genes to behavior. Oxford University Press,
Oxford.
Becker, J.B., Breedlove, S.M., Crews, D. & McCarthy, M.M. (2002) Behavioral
Endocrinology. MIT Press, Cambridge MA.
Bell, A.P., Weinberg, M.S. & Hammersmith, S.K. (1981) Sexual preference: It’s develop-
ment in men and women. Indiana University Press, Bloomington.
Benbow, C.P. (1993) Sex differences in mathematical reasoning ability in intellectually
talented preadolescents: Their nature, effects and possible causes. BBS 11:169–232.
Behav. Brain Sci., 16, 187–189.
Bénesteau, J. (2002) Mensonges Freudiens. Pierre Mardaga, Wavre (Belgique).
Berenbaum, S.A., Bryk, K.K., Nowak, N., Quigley, C.A. & Moffat, S. (2009) Fingers as
a marker of prenatal androgen exposure. Endocrinology, 150, 5119–5124.
Berenbaum, S.A., Duck, S.C. & Bryk, K. (2000) Behavioral effects of prenatal versus
postnatal androgen excess in children with 21-hydroxylase-deficient congenital
adrenal hyperplasia. J Clin Endocrinol Metab, 85, 727–733.
Berenbaum, S.A. & Hines, M. (1992) Early androgens are related to childhood sex-
typed toy preferences. Psychol. Sci., 3, 203–206.
Berenbaum, S.A. & Snyder, E. (1995) Early hormonal influences on childhood sex-
typed activity and playmate preferences: implications for the development of sexual
orientation. Developmental Psychology, 31, 31–42.
Berglund, H., Lindstrom, P. & Savic, I. (2006) Brain response to putative pheromones in
lesbian women. Proc. Natl. Acad. Sci. U S A, 103, 8269–8274.
Berta, P., Hawkins, J.R., Sinclair, A.H., Taylor, A., Griffiths, B.L., Goodfellow, P.N. &
Fellous, M. (1990) Genetic evidence equating SRY and the testis-determining factor.
Nature, 348, 448–450.
Berthold, A.A. (1849) Transplantation der Hoden. Arch. F. Anat. U. Physiol., 16, 42–46.
Blanchard, R. (1997) Birth order and sibling sex ration in homosexual versus hetero-
sexual males and females. Ann. Rev. Sex Res. 8, 27–67.
Blanchard, R. (2001) Fraternal birth order and the maternal immune hypothesis of
male homosexuality. Horm. Behav., 40, 105–114.
166 References
Blanchard, R. (2004) Quantitative and theoretical analyses of the relation between older
brothers and homosexuality in men. J. Theor. Biol. 230, 173–187.
Blanchard, R. & Bogaert, A.F. (1996) Homosexuality in men and number of older broth-
ers. Am. J. Psychiatry, 153, 27–31.
Blanchard, R., Cantor, J.M., Bogaert, A.F., Breedlove, S.M. & Ellis, L. (2006) Interaction
of fraternal birth order and handedness in the development of male homosexuality.
Horm. Behav., 49, 405–414.
Blanchard, R. & Klassen, P. (1997) H-Y antigen and homosexuality in men. J. Theor.
Biol., 185, 373–378.
Blanchard, R. & Lippa, R.A. (2007) Birth order, sibling sex ratio, handedness, and sex-
ual orientation of male and female participants in a BBC internet research project.
Arch. Sex. Behav., 36, 163–176.
Bloch, G.J. & Gorski, R.A. (1988a) Cytoarchitectonic analysis of the SDN-POA of
the intact and gonadectomized rat. J. Comp. Neurol., 275, 604–612.
Bloch, G.J. & Gorski, R.A. (1988b) Estrogen/progesterone treatment in adulthood affects
the size of several components of the medial preoptic area in the male rat. J. Comp.
Neurol., 275, 613–622.
Bocklandt, S., Horvath, S., Vilain, E. & Hamer, D.H. (2006) Extreme skewing of
X chromosome inactivation in mothers of homosexual men. Hum. Genet., 118,
691–694.
Bocklandt, S. & Vilain, E. (2007) Sex differences in brain and behavior: hormones versus
genes. Adv. Genet., 59, 245–266.
Bodo, C. & Rissman, E.F. (2007) Androgen receptor is essential for sexual differentia-
tion of responses to olfactory cues in mice. Eur. J. Neurosci., 25, 2182–2190.
Bodo, C. & Rissman, E.F. (2008) The androgen receptor is selectively involved in
organization of sexually dimorphic social behaviors in mice. Endocrinology, 149,
4142–4150.
Bogaert, A.F. (2006) Biological versus nonbiological older brothers and men’s sexual
orientation. Proc. Natl. Acad. Sci. U S A, 103, 10771–10774.
Bogaert, A.F. & Skorska, M. (2011) Sexual Orientation, Fraternal Birth Order, and the
Maternal Immune Hypothesis: A Review of Evidence. Front Neuroendocrinol, 32,
247–254.
Bradley, S.J., Oliver, G.D., Chernick, A.B. & Zucker, K.J. (1998) Experiment of nurture:
ablatio penis at 2 months, sex reassignment at 7 months, and a psychosexual fol-
low-up in young adulthood. Pediatrics, 102, 1–5.
Breedlove, S.M. (2010) Minireview: Organizational hypothesis: instances of the finger-
post. Endocrinology, 151, 4116–4122.
Brock, O. & Bakker, J. (2011) Potential contribution of prenatal estrogens to the sexual
differentiation of mate preferences in mice. Horm. Behav., 59, 83–89.
Brown, W.M., Finn, C.J., Cooke, B.M. & Breedlove, S.M. (2002a) Differences in finger
length ratios between self-identified “butch” and “femme” lesbians. Arch. Sex. Behav.,
31, 123–127.
Brown, W.M., Hines, M., Fane, B.A. & Breedlove, S.M. (2002b) Masculinized finger
length patterns in human males and females with congenital adrenal hyperplasia.
Horm. Behav., 42, 380–386.
Byne, W., Tobet, S., Mattiace, L.A., Lasco, M.S., Kemether, E., Edgar, M.A., Morgello, S.,
Buchsbaum, M.S. & Jones, L.B. (2001) The interstitial nuclei of the human anterior
References 167
hypothalamus: An investigation of variation with sex, sexual orientation, and HIV

status. Horm. Behav., 40, 86–92.
Cameron, P. & Cameron, K. (1995) Does incest cause homosexuality? Psychol. Rep., 76,
611–621.
Campbell, B.C., Prossinger, H. & Mbzivo, M. (2005) Timing of Pubertal Maturation and
the Onset of Sexual Behavior among Zimbabwe School Boys. Arch. Sexual Behav., 34,
505–516.
Camperio-Ciani, A., Corna, F. & Capiluppi, C. (2004) Evidence for maternally inherited
factors favouring male homosexuality and promoting female fecundity. Proc. Biol. Sci,
271, 2217–2221.
Carani, C., Bancroft, J., Granata, A., Del Rio, G. & Marrama, P. (1992) Testosterone
and erectile function, nocturnal penile tumescence and rigidity, and erectile response
to visual erotic stimuli in hypogonadal and eugonadal men. Psychoneuroendocrinology,
17, 647–654.
Carani, C., Granata, A.R., Rochira, V., Caffagni, G., Aranda, C., Antunez, P. &
Maffei, L.E. (2005) Sex steroids and sexual desire in a man with a novel mutation
of aromatase gene and hypogonadism. Psychoneuroendocrinology, 30, 413–417.
Carani, C., Rochira, V., Faustini-Fustini, M., Balestrieri, A. & Granata, A.R.M. (1999)
Role of oestrogen in male sexual behaviour: insights from the natural model of aro-
matase deficiency. Clinical Endocrinology, 51, 517–524.
Chivers, M.L., Rieger, G., Latty, E. & Bailey, J.M. (2004) A sex difference in the specific-
ity of sexual arousal. Psychol. Sci., 15, 736–744.
Chung, W.C.J., De Vries, G.J. & Swaab, D.F. (2002) Sexual differentiation of the bed
nucleus of the stria terminalis in humans may extend into adulthood. J. Neurosci.,
22, 1027–1033.
Churchill, W. (1967) Homosexual behavior among males: A cross-cultural and cross-
species investigation. Hawthorn books, New York.
Colapinto, J. (2000) As nature made him: The boy that was raised as a girl. Harper
Collins, New York.
Collaer, M.L. & Hines, M. (1995) Human behavioral sex differences: a role for gonadal
hormones during early development? Psychol. Bull., 118, 55–107.
Connellan, J., Baron-Cohen, S., Wheelwright, S., Batk, A. & Ahluwalia, J. (2000) Sex
differences in human neonatal social perception. Infant Behav. Develop, 23, 113–118.
Cooke, B.M., Tabibnia, G. & Breedlove, S.M. (1999) A brain sexual dimorphism
controlled by adult circulating androgens. Proc. Natl. Acad. Sci. USA, 96, 7538–7540.
Cooper, J.R., Bloom, F.E. & Roth, R.H. (1996) The biochemical basis of neuropharmacol-
ogy. Oxford University Press, New York.
Cornil, C.A., Ball, G.F. & Balthazart, J. (2006) Functional significance of the rapid
regulation of brain estrogen action: Where do the estrogens come from? Brain Res.,
1126, 2–26.
Crews, D., Camazine, B., Diamond, M., Mason, R., Tokarz, R. & Gartska, W.R. (1984)
Hormonal independence of courtship behavior in the male garter snake. Horm.
Behav., 18, 29–41.
Crews, D. & Gartska, W.R. (1982) The ecological physiology of a garter snake. Sci. Am.,
247, 159–168.
Davidson, J.M., Camargo, C.A. & Smith, E.R. (1979) Effects of androgen on sexual
behavior in hypogonadal men. J. Clin. Endocrinol. Metab., 48, 955–958.
168 References
Davis, E.C., Shryne, J.E. & Gorski, R.A. (1995) A revised critical period for the sexual
differentiation of the sexually dimorphic nucleus of the preoptic area in the rat.
Neuroendocrinol., 62, 579–585.
de Lacoste-Utamsing, C. & Holloway, R.L. (1982) Sexual dimorphism in the human
corpus callosum. Science, 216, 1431–1432.
De Vries, G.J., Rissman, E.F., Simerly, R.B., Yang, L.Y., Scordalakes, E.M., Auger, C.J.,
Swain, A., Lovell-Badge, R., Burgoyne, P.S. & Arnold, A.P. (2002) A model system for
study of sex chromosome effects on sexually dimorphic neural and behavioral traits.
J. Neurosci., 22, 9005–9014.
Diamond, M. (1993) Some genetic considerations in the development of sexual orien-
tation. In Haug, M., Whalen, R.E., Aron, C., Olsen, K.L. (eds) The development of
sex differences and similarities in behavior. Kluwer Academic Publishers, Dordrecht,
pp. 291–309.
Diamond, M. & Sigmundson, H.K. (1997) Sex reassignment at birth: long-term review
and clinical implications. Arch. Pediatrics and Adolescent Medicine, 151, 298–304.
Dittmann, R.W., Kappes, M.E. & Kappes, M.H. (1992) Sexual behavior in adolescent
and adult females with congenital adrenal hyperplasia. Psychoneuroendocrinology,
17, 153–170.
Dörner, G. (1969) Zur Frage einer neuroendocrinen Pathogenese, Prophylaxe und
Therapie angeborenen Sexualdeviationen. Deutsche Medizinische Wochenschrift, 94,
390–396.
Dörner, G. (1972) Sexualhormonabhängige Gehirndifferenzierung und Sexualität.
Springer, Berlin, Heidelberg, New York.
Dörner, G. (1976) Hormones and Brain Differentiation. Elsevier, Amsterdam, Oxford,
New York.
Dörner, G. (1980) Sexual differentiation of the brain Vitamins and hormones. Academic
Press, Inc., pp. 325–381, New York.
Dörner, G., Geier, T., Ahrens, L., Krell, L., Munx, G., Sieler, H., Kittner, E. &
Muller, H. (1980) Prenatal stress as possible aetiogenetic factor of homosexuality in
human males. Endokrinologie, 75, 365–368.
Dörner, G., Schenk, B., Schmiedel, B. & Ahrens, L. (1983) Stressful events in prenatal
life of bi- and homosexual men. Exp Clin Endocrinol, 81, 83–87.
Dörner, G. & Staudt, J. (1969) Structural changes in the hypothalamic ventromedial
nucleus of the male rat, following neonatal castration and androgen treatment.
Dubb, A., Gur, R., Avants, B. & Gee, J. (2003) Characterization of sexual dimorphism
in the human corpus callosum. Neuroimage, 20, 512–519.
Dufresne, T. (2007) Against Freud. Critics talk back. Stanford University Press,
Stan ford, CA.
Ehrhardt, A.A., Meyer-Bahlburg, H.F., Rosen, L.R., Feldman, J.F., Veridiano, N.P.,
Zimmerman, I. & McEwen, B.S. (1985) Sexual orientation after prenatal exposure to
exogenous estrogen. Arch. Sex. Behav., 14, 57–77.
Elbert, T., Pantev, C., Wienbruch, C., Rockstroh, B. & Taub, E. (1995) Increased cor-
tical representation of the fingers of the left hand in string players. Science, 270,
305–307.
References 169
Ellis, L., Ames, M.A., Peckham, W. & Ahrens, L. (1988) Sexual orientation of human
offspring may be altered by severe maternal stress during pregnancy. J. Sex. Res., 25,
152–157.
Ellis, L., Hershberger, S., Field, E., Wersinger, S., Pelis, S., Geary, D., Palmer, C.,
Hoyenga, K., Hetsroni, A. & Karadi, K. (2008) Sex differences: summarizing more than
a century of scientific research. Psychology Press, New York.
Ellis, L., Hoffman, H. & Burke, D.M. (1990) Sex, sexual orientation and criminal and
violent behavior. Personality and Individual Differences, 11, 1207–1211.
Everitt, B.J. (1990) Sexual motivation: a neural and behavioural analysis of the mecha-
nisms underlying appetitive and copulatory responses in male rats. Neurosci. Biobehav.
Rev., 14, 217–232.
Everitt, B.J. (1995) Neuroendocrine mechanisms underlying appetitive and consum-
matory elements of masculine sexual behavior. In Bancroft, J. (ed) The pharmacology
of sexual function and dysfunction. Elsevier, Amsterdam, pp. 15–31.
Fernández-Guasti, A., Kruijver, F.P., Fodor, M. & Swaab, D.F. (2000) Sex differences in
the distribution of androgen receptors in the human hypothalamus. J. Comp. Neurol.,
425, 422–435.
Fine, C. (2010) Delusions of gender: how our minds, society and neurosexism create
difference. W.W. Norton and Company, New York.
Foidart, A., Legros, J.J. & Balthazart, J. (1994) Les phéromones humaines: vestige animal
ou réalité non reconnue. Revue Médicale de Liège, 49, 662–680.
Fraga, M.F., Ballestar, E., Paz, M.F., Ropero, S., Setien, F., Ballestar, M.L., Heine-
Suner, D., Cigudosa, J.C., Urioste, M., Benitez, J., Boix-Chornet, M., Sanchez-
Aguilera, A., Ling, C., Carlsson, E., Poulsen, P., Vaag, A., Stephan, Z., Spector, T.D.,
Wu, Y.Z., Plass, C. & Esteller, M. (2005) Epigenetic differences arise during the life-
time of monozygotic twins. Proc. Natl. Acad. Sci. U S A, 102, 10604–10609.
Freund, K. & Blanchard, R. (1983) Is the distant relationship of fathers and homo-
sexual sons related to the son’s erotic preference for male partners, or to the sons’
atypical gender identity, or both? Journal of Homosexuality, 9, 7–25.
Freund, K.W. (1974) Male homosexuality: an analysis of the pattern. In Lorraine, J.A.
(ed) Understanding homosexuality: its biological and psychological bases. Elsevier,
Amsterdam, pp. 25–81.
Garcia-Falgueras, A. & Swaab, D.F. (2008) A sex difference in the hypothalamic unci-
nate nucleus: relationship to gender identity. Brain, 131, 3132–3146.
Gladue, B.A. (1985) Neuroendocrine response to estrogen and sexual orientation.
Science, 230, 961.
Gladue, B.A. & Bailey, J.M. (1995) Aggressiveness, competitiveness, and human sexual
orientation. Psychoneuroendocrinol., 20, 475.
Gladue, B.A., Beatty, W.W., Larson, J. & Staton, R.D. (1990) Sexual orientation and
spatial ability in men and women. Psychobiol., 18, 101–108.
Gladue, B.A., Green, R. & Hellman, R.E. (1984) Neuroendocrine response to estrogen
and sexual orientation. Science, 225, 1496–1499.
Gobrogge, K.L., Breedlove, S.M. & Klump, K.L. (2008) Genetic and environmental
influences on 2D:4D finger length ratios: a study of monozygotic and dizygotic male
and female twins. Arch. Sex. Behav., 37, 112–118.
170 References
Gorski, R.A. (1984) Critical role of the medial preoptic area in the sexual differentiation
of the brain. In De Vries, G.J., De Bruin, J.P.C., Uylings, H.B.M., Corner, M.A. (eds)
Sex differences in the brain. Elsevier, Amsterdam, pp. 129–146.
Gorski, R.A., Gordon, J.H., Shryne, J.E. & Southam, A.M. (1978) Evidence for a mor-
phological sex difference within the medial preoptic area of the rat brain. Brain Res.,
148, 333–346.
Goy, R.W. & McEwen, B.S. (1980) Sexual differentiation of the brain. The MIT Press,
Cambridge, MA.
Green, R. (1978) Sexual identity of 37 children raised by homosexual or transsexual
parents. Am. J. Psychiatry, 135, 692–697.
Grimbos, T., Dawood, K., Burriss, R.P., Zucker, K.J. & Puts, D.A. (2010) Sexual orienta-
tion and the second to fourth finger length ratio: a meta-analysis in men and women.
Behav. Neurosci., 124, 278–287.
Guiso, L., Monte, F., Sapienza, P. & Zingales, L. (2008) Diversity. Culture, gender, and
math. Science, 320, 1164–1165.
Hajjar, R.R., Kaiser, F.E. & Morley, J.E. (1997) Outcomes of long-term testosterone
replacement in older hypogonadal males: a retrospective analysis. J. Clin. Endocrinol
Metab., 82, 3793–3796.
Hall, L.S. & Kimura, D. (1995) Sexual orientation and performance on sexually dimor-
phic motor tasks. Arch. Sex. Behav., 24, 395–407.
Hall, L.S. & Love, C.T. (2003) Finger-length ratios in female monozygotic twins discor-
dant for sexual orientation. Arch. Sex. Behav., 32, 23–28.
Halperin, D.M. (1990) One hundred years of homosexuality and other essays on Greek
love. Routledge, New York.
Halpern, C.T., Udry, J.R. & Suchindran, C. (1998) Monthly measures of salivary testos-
terone predict sexual activity in adolescent males. Archives of Sexual Behavior, 27,
445–465.
Halpern, C.T., Udry, R. & Suchindran, C. (1997) Testosterone Predicts Initiation of
Coitus in Adolescent Females. Psychosomatic Medicine, 59, 161–171.
Halpern, M. & Martinez-Marcos, A. (2003) Structure and function of the vomeronasal
system: an update. Prog Neurobiol, 70, 245–318.
Hamer, D.H., Hu, S., Magnuson, V.L., Hu, N. & Pattatucci, A.M.L. (1993) A linkage
between DNA markers on the X chromosome and male sexual orientation. Science,
261, 321–327.
Hassett, J.M., Siebert, E.R. & Wallen, K. (2008) Sex differences in rhesus monkey toy
preferences parallel those of children. Horm. Behav., 54, 359–364.
Hefez, S. (2007) Dans le coeur des hommes. Hachette, Paris.
Henley, C.L., Nunez, A.A. & Clemens, L.G. (2009) Estrogen treatment during develop-
ment alters adult partner preference and reproductive behavior in female laboratory
rats. Horm. Behav., 55, 68–75.
Henley, C.L., Nunez, A.A. & Clemens, L.G. (2011) Hormones of choice: the neuroendo-
crinology of sexual orientation in animals. Front Neuroendocrinol, 32, 146–154.
Hepper, P.G., Shahidullah, S. & White, R. (1991) Handedness in the human fetus. Neuro-
psychologia, 29, 1107–1111.
Hines, M. (2003) Sex steroids and human behavior: prenatal androgen exposure and
sex-typical play behavior in children. Ann. N. Y. Acad. Sci., 1007, 272–282.
References 171
Hines, M. (2004) Brain gender. Oxford University Press, Oxford.

Hines, M. (2006) Prenatal testosterone and gender-related behaviour. Eur. J. Endocrinol.,
155 Suppl 1, S115–121.
Hines, M. & Alexander, G.M. (2008) Monkeys, girls, boys and toys: a confirmation.
Letter regarding “Sex differences in toy preferences: striking parallels between mon-
keys and humans.” Horm. Behav., 54, 478–479; author reply 80–71.
Hines, M., Brook, C. & Conway, G.S. (2004) Androgen and psychosexual development:
core gender identity, sexual orientation and recalled childhood gender role behav-
ior in women and men with congenital adrenal hyperplasia (CAH). J. Sex. Res., 41,
75–81.
Hines, M., Fane, B.A., Pasterski, V.L., Mathews, G.A., Conway, G.S. & Brook, C.
(2003) Spatial abilities following prenatal androgen abnormality: targeting and
mental rotations performance in individuals with congenital adrenal hyperplasia.
Psychoneuroendocrinology, 28, 1010–1026.
Hines, M. & Kaufman, F.R. (1994) Androgen and the development of human sex-
typical behavior: rough-and-tumble play and sex of preferred playmates in children
with congenital adrenal hyperplasia (CAH). Child Dev., 65, 1042–1053.
Honda, S., Harada, N., Ito, S., Takagi, Y. & Maeda, S. (1998) Disruption of sexual
behavior in male aromatase-deficient mice lacking exons 1 and 2 of the cyp19 gene.
Biochem. Biophys. Res. Commun., 252, 445–449.
Houtsmuller, E.J., Brand, T., de Jonge, F.H., Joosten, R.N., van de Poll, N.E. &
Slob, A.K. (1994) SDN-POA volume, sexual behavior, and partner preference of
male rats affected by perinatal treatment with ATD. Physiol. Behav., 56, 535–541.
Hu, S., Pattatucci, A.M., Patterson, C., Li, L., Fulker, D.W., Cherny, S.S., Kruglyak, L.
& Hamer, D.H. (1995) Linkage between sexual orientation and chromosome Xq28
in males but not in females. Nat. Genet., 11, 248–256.
Hyde, J.S., Lindberg, S.M., Linn, M.C., Ellis, A.B. & Williams, C.C. (2008) Diversity.
Gender similarities characterize math performance. Science, 321, 494–495.
Iijima, M., Arisaka, O., Minamoto, F. & Arai, Y. (2001) Sex differences in children’s free
drawings: a study on girls with congenital adrenal hyperplasia. Horm. Behav., 40,
99–104.
Imperato-McGinley, J. (1994) 5 alpha-reductase deficiency: human and animal models.
Eur. Eurol., 25 Suppl 1, 20–23.
Imperato-McGinley, J., Miller, M., Wilson, J.D., Peterson, R.E., Shackleton, C. &
Gajdusek, D.C. (1991) A cluster of male pseudohermaphrodites with 5 alpha-
reductase deficiency in Papua New Guinea. Clin. Endocrinol. (Oxf), 34, 293–298.
Imperato-McGinley, J. & Zhu, Y.S. (2002) Androgens and male physiology the syn-
drome of 5alpha-reductase-2 deficiency. Mol. Cell. Endocrinol., 198, 51–59.
Imwalle, D.B., Bateman, H.L., Wills, A., Honda, S., Harada, N. & Rissman, E.F. (2006)
Impairment of spatial learning by estradiol treatment in female mice is attenuated
by estradiol exposure during development. Horm. Behav., 50, 693–698.
Jacobson, C.D., Csernus, V.J., Shryne, J.E. & Gorski, R.A. (1981) The influence of gon-
adectomy, androgen exposure, or a gonadal graft in the neonatal rat on the volume
of the sexually dimorphic nucleus of the preoptic area. J. Neurosci., 1, 1142–1147.
Jacobson, C.D., Shryne, J.E., Shapiro, F. & Gorski, R.A. (1980) Ontogeny of the sexually
dimorphic nucleus of the preoptic area. J. Comp Neurol., 193, 541–548.
172 References
Jimbo, M., Okubo, K., Toma, Y., Shimizu, Y., Saito, H. & Yanaihara, T. (1998) Inhibitory
effects of catecholamines and maternal stress on aromatase activity in the fetal rat
brain. J. Obstet. Gynaecol. Res., 24, 291–297.
Katz, J.N. (1976) Gay American History: Lesbians and Gay Men in the U.S.A. T.Y. Crowell,
New York.
Kelley, D.B. & Pfaff, D.W. (1978) Generalizations from comparative studies on neuroana-
tomical and endocrine mechanisms of sexual behaviour. In Hutchison, J.B. (ed) Biological
determinants of sexual behaviour. John Wiley & Sons, Chichester, pp. 225–254.
Kerchner, M. & Ward, I.L. (1992) SDN-MPOA volume in male rats is decreased by
prenatal stress, but is not related to ejaculatory behavior. Brain Res., 581, 244–251.
Kessler, S.J. & McKenna, W. (1978) Gender: an ethno-methodological approach. Wiley,
New York.
Keverne, E.B. (1999) The vomeronasal organ. Science, 286, 716–720.
Kinsey, A.C., Pomeroy, W.R. & Martin, C.E. (1948) Sexual behavior in the human male.
W.B. Saunders Company, Philadelphia.
Kinsey, A.C., Pomeroy, W.R., Martin, C.E. & Gebhard, P.H. (1953) Sexual behavior in
the human female. Saunders, Philadelphia.
Kiumura, D. (1999) Sex and cognition. MIT Press, Cambridge.
Kourany, R.F. (1987) Suicide among homosexual adolescents. J. Homosex., 13, 111–117.
Kraemer, B., Noll, T., Delsignore, A., Milos, G., Schnyder, U. & Hepp, U. (2006) Finger
length ratio (2D:4D) and dimensions of sexual orientation. Neuropsychobiology, 53,
210–214.
Kruijver, F.P.M., Balesar, R., Espila, A.M., Unmehopa, U.A. & Swaab, D.F. (2002)
Estrogen receptor-alpha distribution in the human hypothalamus in relation to sex
and endocrine status. J. Comp. Neurol., 454, 115–139.
Kruijver, F.P.M., Balesar, R., Espila, A.M., Unmehopa, U.A. & Swaab, D.F. (2003)
Estrogen-receptor-beta distribution in the human hypothalamus: Similarities and
differences with ERalpha distribution. J. Comp. Neurol., 466, 251–277.
Kruijver, F.P.M., de Jonge, F.H., van den Broek, W.T., van der Woude, T., Endert, E.
& Swaab, D.F. (1993) Lesions of the suprachiasmatic nucleus do not disturb sexual
orientation of the adult male rat. Brain Res., 624, 342–346.
Kruijver, F.P.M., Fernández-Guasti, A., Fodor, M., Kraan, E.M. & Swaab, D.F. (2001)
Sex differences in androgen receptors of the human mamillary bodies are related to
endocrine status rather than to sexual orientation or transsexuality. J. Clin. Endocrinol.
Metab., 86, 818–827.
Lalumiere, M.L., Blanchard, R. & Zucker, K.J. (2000) Sexual orientation and handed-
ness in men and women: a meta-analysis. Psychol. Bull, 126, 575–592.
Lebson, M. (2002) Suicide among homosexual youth. J. Homosex., 42, 107–117.
Legros, J.J., Van Cauwenberge, H., Lambotte, R., Bauduin, A., Franchimont, P. &
Legros, J. (1974) Problèmes psycho-endocriniens posés par un cas d’hyperplasie sur-
rénalienne congénitale reconnu tardivement. Revue Médicale de Liège, 29, 73–80.
LeVay, S. (1991) A difference in hypothalamic structure between heterosexual and
homosexual men. Science, 253, 1034–1037.
LeVay, S. (1993) The sexual brain. MIT Press, Cambridge, MA.
LeVay, S. (1996) Queer science. MIT Press, Cambridge, MA.
References 173
LeVay, S. (2010) Gay, straight, and the reason why. The science of sexual orientation.
Oxford University Press, New York.
LeVay, S. & Hamer, D.H. (1994) Evidence for a biological influence in male homosexu-
ality. Sci. Am., 270, 44–49.
LeVay, S. & Valente, S.M. (2006) Human Sexuallity. Sinauer Associates Inc,
Sunderland, MA.
Lippa, R.A. (2003a) Are 2D:4D finger-length ratios related to sexual orientation? Yes
for men, no for women. J. Pers. Soc. Psychol., 85, 179–188.
Lippa, R.A. (2003b) Handedness, sexual orientation, and gender-related personality
traits in men and women. Arch. Sex. Behav., 32, 103–114.
Lorenz, K. (1937) Uber die Bildung des Instinktbergiffes (The establishment of the
instinct concept). Die Naturwissenschaften, 25, 280–300, 307–318, 325–331.
Lorenz, K. (1950) The comparative method in studying innate behavior patterns. Symp.
Soc. Exp. Biol., 4, 221–268.
Lutchmaya, S., Baron-Cohen, S. & Raggatt, P. (2002) Foetal testosterone and eye contact
in 12-month-old human infants. Infant Behav. Develop, 25, 327–335.
Lutchmaya, S., Baron-Cohen, S., Raggatt, P., Knickmeyer, R. & Manning, J.T. (2004) 2nd
to 4th digit ratios, fetal testosterone and estradiol. Early Hum. Dev., 77, 23–28.
MacLaughlin, D.T. & Donahoe, P.K. (2004) Sex determination and differentiation.
N. Engl. J. Med., 350, 367–378.
Manning, J.T., Fink, B., Neave, N. & Szwed, A. (2006) The second to fourth digit ratio
and asymmetry. Ann. Hum. Biol., 33, 480–492.
Mansukhani, V., Adkins-Regan, E. & Yang, S. (1996) Sexual partner preference in
female zebra finches: The role of early hormones and social environment. Horm.
Behav., 30, 506–513.
Marler, P. & Hamilton, W.J.I. (1966) Mechanisms of animal behavior. Wiley and Sons,
New York.
Martin, J.T. & Nguyen, D.H. (2004) Anthropometric analysis of homosexuals and
heterosexuals: implications for early hormone exposure. Horm. Behav., 45, 31–39.
McCormick, C.M. & Witelson, S.F. (1991) A cognitive profile of homosexual men
compared to heterosexual men and women. Psychoneuroendocrinology, 16, 459–73.
McEwen, B.S. & Alves, S.E. (1999) Estrogen actions in the central nervous system.
Endocr. Rev., 20, 279–307.
McFadden, D. (2002) Masculinization effects in the auditory system. Arch. Sex. Behav.,
31, 99–111.
McFadden, D. (2008) What do sex, twins, spotted hyenas, ADHD, and sexual orienta-
tion have in common? Perspectives Psychol. Sci, 3, 309–323.
McFadden, D. (2011) Sexual orientation and the auditory system. Front. Neuroendocrinol.,
32, 201–213.
McFadden, D. & Champlin, C.A. (2000) Comparison of auditory evoked potentials in
heterosexual, homosexual, and bisexual males and females. J. Assoc. Res. Otolaryngol.,
1, 89–99.
McFadden, D. & Pasanen, E.G. (1998) Comparison of the auditory systems of hetero-
sexuals and homosexuals: click-evoked otoacoustic emissions. Proc. Natl. Acad. Sci.
U S A, 95, 2709–2713.
174 References
McFadden, D. & Pasanen, E.G. (1999) Spontaneous otoacoustic emissions in hetero-

sexuals, homosexuals, and bisexuals. J. Acoust. Soc. Am., 105, 2403–2413.
McFadden, D., Pasanen, E.G., Valero, M.D., Roberts, E.K. & Lee, T.M. (2009) Effect of
prenatal androgens on click-evoked otoacoustic emissions in male and female sheep
(Ovis aries). Horm. Behav., 55, 98–105.
McFadden, D. & Shubel, E. (2002) Relative lengths of fingers and toes in human males
and females. Horm. Behav., 42, 492–500.
Meredith, M. (2001) Human vomeronasal organ function: a critical review of best and
worst cases. Chem. Senses, 26, 433–445.
Meyer-Bahlburg, H.F. (1984) Psychoendocrine research on sexual orientation. Current
status and future options. Prog. Brain Res., 61, 375–398.
Meyer-Bahlburg, H.F. (2005) Gender identity outcome in female-raised 46, XY persons
with penile agenesis, cloacal exstrophy of the bladder, or penile ablation. Arch. Sex.
Behav., 34, 423–438.
Meyer-Bahlburg, H.F. (2009) Male Gender Identity in an XX Individual with Congenital
Adrenal Hyperplasia. J. Sex. Med. 5, 298–299.
Meyer-Bahlburg, H.F., Dolezal, C., Baker, S.W. & New, M.I. (2008) Sexual orientation
in women with classical or non-classical congenital adrenal hyperplasia as a function
of degree of prenatal androgen excess. Arch. Sex. Behav., 37, 85–99.
Meyer-Bahlburg, H.F., Ehrhardt, A.A. & Rosen, L.R. (1995) Prenatal estrogens and the
development of homosexual orientation. Dev. Psychol., s31, 12–21.
Micevych, P. & Dominguez, R. (2009) Membrane estradiol signaling in the brain. Front.
Miller, G. (2008) Neuroimaging. Growing pains for fMRI. Science, 320, 1412–1414.
Miller, G., Tybur, J. & Jordan, B.D. (2007) Ovulatory cycle effects on tip earnings by
lap dancers: economic evidence for human estrus? Evolution and Human Behavior,
28, 375–381.
Money, J. & Ehrhardt, A.A. (1972) Man & Woman, Boy & Girl. Johns Hopkins University
Press, Baltimore.
Money, J., Schwartz, M. & Lewis, V.G. (1984) Adult erotosexual status and fetal hor-
monal masculinization and demasculinization: 46, XX congenital virilizing adrenal
hyperplasia and 46, XY androgen-insensitivity syndrome compared. Psychoneuroen-
docrinology, 9, 405–414.
Morrell, J.I. & Pfaff, D.W. (1978) A neuroendocrine approach to brain function:
localization of sex steroid concentrating cells in vertebrate brains. Amer. Zool., 18,
447–460.
Mosher, W.D., Chandra, A. & Jones, J. (2005) Sexual behavior and selected health mea-
sures: men and women 15–44 years of age, United States 2002. http://www.cdc/gov/
nchs/data/ad/ad362.pdf.
Mustanski, B.S., Dupree, M.G., Nievergelt, C.M., Bocklandt, S., Schork, N.J. &
Hamer, D.H. (2005) A genomewide scan of male sexual orientation. Hum. Genet.,
116, 272–278.
Nardi, P. (1992) Men’s friendships. Sage Publications, Newbury Park, CA.
Neave, N., Menaged, M. & Weightman, D.R. (1999) Sex differences in cognition: the
role of testosterone and sexual orientation. Brain Cogn., 41, 245–262.
Nelson, R.J. (2005) An introduction to behavioral endocrinology. Sinauer Associates,
Inc., Sunderland, Massachussets.
References 175
Ngun, T.C., Ghahramani, N., Sanchez, F.J., Bocklandt, S. & Vilain, E. (2011) The genet-
ics of sex differences in brain and behavior. Front. Neuroendocrinol., 32, 227–246.
Nordenstrom, A., Servin, A., Bohlin, G., Larsson, A. & Wedell, A. (2002) Sex-typed
toy play behavior correlates with the degree of prenatal androgen exposure assessed
by CYP21 genotype in girls with congenital adrenal hyperplasia. J. Clin. Endocrinol.
Metab., 87, 5119–5124.
Nottebohm, F. & Arnold, A.P. (1976) Sexual dimorphism in vocal control areas of the
songbird brain. Science, 194, 211–213.
Onfray, M. (2010) Le crépuscule d’une idole. L’affabulation freudienne. Bernard Grasset,
Paris.
Op de Beeck, H.P., Haushofer, J. & Kanwisher, N.G. (2008) Interpreting fMRI data:
maps, modules and dimensions. Nat. Rev. Neurosci., 9, 123–135.
Ozdemir, S.T., Ercan, I., Sevinc, O., Guney, I., Ocakoglu, G., Aslan, E. & Barut, C. (2007)
Statistical shape analysis of differences in the shape of the corpus callosum between
genders. Anat. Rec. (Hoboken), 290, 825–830.
Paredes, R.G. & Baum, M.J. (1995) Altered sexual partner preference in male ferrets
given excitotoxic lesions of the preoptic area anterior hypothalamus. J. Neurosci., 15,
6619–6630.
Paredes, R.G., Tzschentke, T. & Nakach, N. (1998) Lesions of the medial preoptic
area anterior hypothalamus (MPOA/AH) modify partner preference in male rats.
Brain Res., 813, 1–8.
Pelletier, G. (2000) Localization of androgen and estrogen receptors in rat and primate
tissues. Histol. Histopathol., 15, 1261–1270.
Perkins, A. & Fitzgerald, J.A. (1992) Luteinizing hormone, testosterone, and behavioral
response of male-oriented rams to estrous ewes and rams. J. Anim. Sci., 70, 1787–1794.
Perkins, A., Fitzgerald, J.A. & Price, E.O. (1992) Luteinizing hormone and testosterone
response of sexually active and inactive rams. J. Anim. Sci., 70, 2086–2093.
Perkins, A. & Roselli, C.E. (2007) The ram as a model for behavioral neuroendocrinol-
ogy. Horm. Behav., 52, 70–77.
Pinckard, K.L., Stellflug, J., Resko, J.A., Roselli, C.E. & Stormshak, F. (2000) Review:
brain aromatization and other factors affecting male reproductive behavior with
emphasis on the sexual orientation of rams. Domes. Anim. Endocrinol., 18, 83–96.
Poianni, A. (2010) Animal homosexuality. A biological perspective. Cambridge University
Press, Cambridge UK.
Rahman, Q., Abrahams, S. & Wilson, G.D. (2003a) Sexual-orientation-related differ-
ences in verbal fluency. Neuropsychology, 17, 240–246.
Rahman, Q., Kumari, V. & Wilson, G.D. (2003b) Sexual orientation-related differ-
ences in prepulse inhibition of the human startle response. Behav. Neurosci., 117,
1096–1102.
Rahman, Q. & Wilson, G.D. (2003a) Born gay? The psychobiology of human sexual
orientation. Personality and Individual Differences, 34, 1337–1382.
Rahman, Q. & Wilson, G.D. (2003b) Large sexual-orientation-related differences in per-
formance on mental rotation and judgment of line orientation tasks. Neuropsychology,
17, 25–31.
Rahman, Q. & Wilson, G.D. (2003c) Sexual orientation and the 2nd to 4th finger length
ratio: evidence for organising effects of sex hormones or developmental instability?
Psychoneuroendocrinology, 28, 288–303.
176 References
Raisman, G. & Field, P.M. (1971) Sexual dimorphism in the preoptic area of the rat.
Science, 173, 731–733.
Reiner, W.G. & Gearhart, J.P. (2004) Discordant sexual identity in some genetic males
with cloacal exstrophy assigned to female sex at birth. N. Engl. J. Med., 350, 333–341.
Resko, J.A., Perkins, A., Roselli, C.E., Fitzgerald, J.A., Choate, J.V. & Stormshak, F.
(1996) Endocrine correlates of partner preference behavior in rams. Biol. Reprod.,
55, 120–126.
Rhees, R.W., Shryne, J.E. & Gorski, R.A. (1990) Termination of the hormone-sensitive
period for differentiation of the sexually dimorphic nucleus of the preoptic area in
male and female rats. Dev. Brain Res., 52, 17–23.
Rice, G., Anderson, C., Risch, N. & Ebers, G. (1999) Male homosexuality: absence of
linkage to microsatellite markers at Xq28. Science, 284, 665–667.
Richelle, M. (1966) Le conditionnement operant. Delachaux et Niestlé, Neuchatel, Suisse.
Rieger, G., Chivers, M.L. & Bailey, J.M. (2005) Sexual arousal patterns of bisexual men.
Psychol. Sci, 16, 579–584.
Robarts, D.W & Baum, M.J. (2007) Ventromedial hypothalamic nucleus lesions dis-
rupt olfactory mate recognition and receptivity in female ferrets. Horm. Behav., 51,
104–113.
Romano, M., Rubolini, D., Martinelli, R., Bonisoli Alquati, A. & Saino, N. (2005)
Experimental manipulation of yolk testosterone affects digit length ratios in the ring-
necked pheasant (Phasianus colchicus). Horm. Behav., 48, 342–346.
Roselli, C.E., Larkin, K., Resko, J.A., Stellflug, J.N. & Stormshak, F. (2004a) The volume
of a sexually dimorphic nucleus in the ovine medial preoptic area/anterior hypo-
thalamus varies with sexual partner preference. Endocrinology, 145, 478–483.
Roselli, C.E., Larkin, K., Schrunk, J.M. & Stormshak, F. (2004b) Sexual partner preference,
hypothalamic morphology and aromatase in rams. Physiol. Behav., 83, 233–245.
Roselli, C.E., Reddy, R. & Kaufman, K. (2011) The development of male-oriented behav-
ior in rams. Front. Neuroendocrinol. 32, 164–169
Roselli, C.E., Stadelman, H., Reeve, R., Bishop, C.V. & Stormshak, F. (2007) The ovine
sexually dimorphic nucleus of the medial preoptic area is organized prenatally by
testosterone. Endocrinology, 148, 4450–4457.
Rosenzweig, M.R., Breedlove, S.M. & Watson, N.V. (2004) Biological psychology: An intro-
duction to behavioral and cognitive neuroscience. Sinauer Associates, Sunderland MA.
Rosler, A. & Witztum, E. (1998) Treatment of men with paraphilia with a long-acting
analogue of gonadotropin-releasing hormone. N. Engl. J. Med., 338, 416–422.
Rubin, R.T., Reinisch, J.M. & Haskett, R.F. (1981) Postnatal gonadal steroid effects on
human behavior. Science, 211, 1318–1324.
Sanders, A.R. & Dawood, K. (2003) Nature encyclopedia of life sciences. Nature Publishing
Group, London.
Savic, I., Berglund, H., Gulyas, B. & Roland, P. (2001) Smelling of odorous sex hormone-
like compounds causes sex-differentiated hypothalamic activations in humans. Neuron,
31, 661–668.
Savic, I., Berglund, H. & Lindstrom, P. (2005) Brain response to putative pheromones in
homosexual men. Proc. Natl. Acad. Sci. U S A, 102, 7356–7361.
Schmidt, G. & Clement, U. (1990) Does peace prevent homosexuality? Arch. Sex. Behav.,
19, 183–187.
References 177
Schober, J.M., Carmichael, P.A., Hines, M. & Ransley, P.G. (2002) The ultimate chal-
lenge of cloacal exstrophy. J. Urol., 167, 300–304.
Sherwin, B.B. & Gelfand, M.M. (1987) The role of androgen in the maintenance of
sexual functioning in oophorectomized women. Psychosom. Med., 49, 397–409.
Sinclair, A.H., Berta, P., Palmer, M.S., Hawkins, J.R., Griffiths, B.L., Smith, M.J., Foster,
J.W., Frischauf, A.M., Lovell-Badge, R. & Goodfellow, P.N. (1990) A gene from the
human sex-determining region encodes a protein with homology to a conserved
DNA-binding motif. Nature, 346, 240–244.
Sisk, C.L. & Zehr, J.L. (2005) Pubertal hormones organize the adolescent brain and
behavior. Front. Neuroendocrinol., 26, 163–174.
Skinner, B.F. (1965) L’analyse expérimentale du comportement. Pierre Mardaga, Wavre.
Skinner, B.F. (1971) L’analyse expérimentale du comportement (Traduit de l’Amércain par
A.M et M. Richelle). Charles Dessart, Bruxelles.
Snyder, P.J., Peachey, H., Berlin, J.A., Hannoush, P., Haddad, G., Dlewati, A., Santanna, J.,
Loh, L., Lenrow, D.A., Holmes, J.H., Kapoor, S.C., Atkinson, L.E. & Strom, B.L. (2000)
Effects of testosterone replacement in hypogonadal men. J. Clin. Endocrinol. Metab.,
85, 2670–2677.
Sommer, V. & Vasey, P.L. (2006) Homosexual behaviour in animals. An evolutionary
perspective. Cambridge University Press, Cambridge.
Stacey, J. & Biblarz, T.J. (2001) (How) does the sexual orientation of parents matter?
American Sociological Review, 66, 159–183.
Strong, B. & DeVault, C. (1997) Human sexuality. Mayfield Publishing Co, Mountain
View, CA.
Swaab, D.F. (2007) Sexual differentiation of the brain and behavior. Best Pract. Res. Clin.
Endocrinol. Metab., 21, 431–444.
Swaab, D.F. & Fliers, E. (1985) A sexually dimorphic nucleus in the human brain.
Science, 228, 1112–1115.
Swaab, D.F. & Hofman, M.A. (1988) Sexual differentiation of the human hypothalamus:
Ontogeny of the sexually dimorphic nucleus of the preoptic area. Dev. Brain Res., 44,
314–318.
Swaab, D.F. & Hofman, M.A. (1990) An enlarged suprachiasmatic nucleus in homo-
sexual men. Brain Res., 537, 141–148.
Taziaux, M., Keller, M., Bakker, J. & Balthazart, J. (2007) Sexual behavior activity tracks
rapid changes in brain estrogen concentrations. J. Neurosci., 27, 6563–6572.
Titus-Ernstoff, L., Perez, K., Hatch, E.E., Troisi, R., Palmer, J.R., Hartge, P., Hyer, M.,
Kaufman, R., Adam, E., Strohsnitter, W., Noller, K., Pickett, K.E. & Hoover, R. (2003)
Psychosexual characteristics of men and women exposed prenatally to diethylstilbe-
strol. Epidemiology, 14, 155–160.
Tuncer, M.C., Hatipoglu, E.S. & Ozates, M. (2005) Sexual dimorphism and handedness
in the human corpus callosum based on magnetic resonance imaging. Surg. Radiol.
Anat., 27, 254–259.
Tutle, G.E. & Pillard, R.C. (1991) Sexual orientation and cognitive abilities. Arch. Sex.
Behav., 20, 307–318.
Udry, J.R., Billy, J.O., Morris, N.M., Groff, T.R. & Raj, M.H. (1985) Serum androgenic
hormones motivate sexual behavior in adolescent boys. Fertil. Steril., 43, 90–94.
Van Rillaer, J. (1980) Les illusions de la psychanalyse. Pierre Mardaga, Sprimont (Belgique).
178 References
Vasudevan, N. & Pfaff, D.W. (2008) Non-genomic actions of estrogens and their inter-
action with genomic actions in the brain. Front. Neuroendocrinol., 29, 238–257.
Vermeulen, A., Rubens, R. & Verdonck, L. (1972) Testosterone secretion and metabo-
lism in male senescence. J. Clin. Endocrinol. Metab., 34, 730–735.
Vidal, C. (ed) (2000) Féminin Masculin Mythes et idéologie. Belin, Paris.
Vidal, C. (2007) Hommes, femmes avons-nous le même cerveau? Le Pommier, Paris.
Wallen, K. (2009) Does finger fat produce sex differences in second to fourth digit
ratios? Endocrinology, 150, 4819–4822.
Wang, C., Swerdloff, R.S., Iranmanesh, A., Dobs, A., Snyder, P.J., Cunningham, G.,
Matsumoto, A.M., Weber, T., Berman, N. & Grp, T.G.S. (2000) Transdermal testosterone
gel improves sexual function, mood, muscle strength, and body composition parameters
in hypogonadal men. J. Clin. Endocrinol. Metab., 85, 2839–2853.
Ward, I.L. (1972) Prenatal stress feminizes and demasculinizes the behavior of males.
Science, 175, 82–84.
Ward, I.L. (1984) The prenatal stress syndrome: current status. Psychoneuroendocrinology,
9, 3–11.
Ward, I.L. & Ward, O.B. (1985) Sexual behavior differentiation: effects of prenatal
manipulations in rats Handbook of Behavioral Neurobiology, Reproduction. 7, 77–98.
Weisz, J. (1983) Influence of maternal stress on the developmental pattern of the ste-
roidogenic function in Leydig cells and steroid aromatase activity in the brain of rat
fetuses. Monogr. Neural. Sci., 9, 184–193.
Wellings, K., Field, J., Johnson, A.M. & Wadsworth, J. (1994) Sexual behavior in Britain:
The National Survey of Sexual Attitudes and Lifestyles. Penguin Books, London.
Williams, C.L. & Pleil, K.E. (2008) Toy story: why do monkey and human males prefer
trucks? Comment on “Sex differences in rhesus monkey toy preferences parallel
those of children” by Hassett, Siebert and Wallen. Horm. Behav., 54, 355–358.
Wisniewski, A.B., Migeon, C.J., Meyer-Bahlburg, H.F., Gearhart, J.P., Berkovitz, G.D.,
Brown, T.R. & Money, J. (2000) Complete androgen insensitivity syndrome: long-
term medical, surgical, and psychosexual outcome. J. Clin. Endocrinol. Metab., 85,
2664–2669.
Witelson, S.F., Kigar, D.L., Scamvougeras, A., Kideckel, D.M., Buck, B., Stanchev, P.L.,
Bronskill, M. & Black, S. (2008) Corpus Callosum Anatomy in Right-Handed
Homosexual and Heterosexual Men. Arch. Sex. Behav. 37, 857–863
Wysocki, C.J. & Preti, G. (2004) Facts, fallacies, fears, and frustrations with human
pheromones. Anat. Rec. A Discov. Mol. Cell. Evol. Biol., 281, 1201–1211.
Yang, X., Schadt, E.E., Wang, S., Wang, H., Arnold, A.P., Ingram-Drake, L., Drake, T.A.
& Lusis, A.J. (2006) Tissue-specific expression and regulation of sexually dimorphic
genes in mice. Genome Res., 16, 995–1004.
Zhou, J.N., Hofman, M.A., Gooren, L.J.G. & Swaab, D.F. (1995) A sex difference in the
human brain and its relation to transsexuality. Nature, 378, 68–70.
Zucker, K.J., Bradley, S.J., Oliver, G., Blake, J., Fleming, S. & Hood, J. (1996) Psycho-
sexual development of women with congenital adrenal hyperplasia. Horm. Behav.,
30, 300–318.
Index
adolescent males, homosexuality homosexual preference expression,

among, 13 51–52, 54
adrenal glands, 29 hormonal influences on, 47–48,
aggression, gender differences in, 65f, 106 50–55
AIS. See androgen insensitivity syndrome hypothalamic control of, 48–50
5α-dihydrotestosterone (DHT), 27, 30–31, lordosis behavior, 53
35, 62 measurement methodology in, 49f
in clinical cases, deficiency of, 136–37 during ontogenesis, 55
deficiencies, in young males, 77–79, 78f ovariectomies, 49
testosterone converted into, 78 prenatal stress theory and, 131
alpha-fetoprotein expression, 52–53 preoptic area lesions, 50
Alzheimer’s disease, 62 in rats, 55–56
amygdala, lesions of, 36 in SDN, 55–56
AND. See 1.4-andostradien-3-1 sex steroids as influence on, 54–55
androgens, 29–31. See also testosterone in sheep, 57–60
female sexual behavior and, social context during rearing, 54
combination treatment with TFM in, 52
estrogen, 95–96, 96f anterior commissure, 120–21, 120f
insensitivity, 62–63, 76–77 anti-Müllerian hormones, 39
receptor mutations, 76 appetitive component, of sexual behavior.
androgen insensitivity syndrome (AIS), See sexual motivation
76–77, 82 Arnold, Art, 44
homosexuality, 138 aromatase inhibitors, 50–51
masculinization of embryos, 77 As Nature Made Him: The Boy That Was
in women, 76, 77f Raised as a Girl (Colapinto), 25
1.4-andostradien-3-1 (AND), 117 autism, 62
animal studies, sexual orientation in,
154–55 Bakker, Julie, 53
alpha-fetoprotein expression, 52–53 Baron-Cohen, Simon, 86–87, 103
aromatase inhibitors, 50–51 Baum, Michael, 49
endocrine controls for, 53–54 behavioral endocrinology
estradiol production, 50 in animal studies, for sexual orientation,
estrogen treatment, 51, 51f, 54 53–54
180 Index
behavioral endocrinology (Cont’d) Centers for Disease Control and

sex determination and, in human Prevention (CDC), 8–9
embryos, 75–81 chromosomes, 36–38. See also sex
sexual behavior influence by, 27–29 chromosomes, anomalies of
behavioral theories, for sexual orientation, female sexual differentiation, 36–39
15–16 male sexual differentiation, 36–37
binding sites sex determination and, 72
in hormones, 28 SRY gene, 37–38
sex steroid activity, 32–33, 32f XQ28 region of X chromosome, 145–48
biological development, 19–21. See also classical conditioning, 15
brain development clinical cases, sexual orientation in,
DNA, 21 130–40
genetic encoding, 20–21 AIS, 138
RNA, 21 CAH, in females, 132–35
of sexual orientation, in animals, 47–60 cloacal exstrophy, 137–38, 159–60
bisexuality, 10–11 DHT deficiency, 136–37
among females, 10 invoked experiences, 130
among males, 10 for prenatal stress, 131–32
Blanchard, Ray, 148 cloacal exstrophy, 137–38, 159–60
Bodo, Christian, 52 cognitive abilities
brain development, 20. See also social gender differences and, in humans, 65,
learning 105–7
amygdala, lesions of, 36 homosexuality and, 104
hormonal imprinting, 19 sexual behavior and, 92
medial preoptic area, 36 visual-spatial tasks, by gender, 105–6
neurons, 21–23 Colapinto, John, 25
pheromones and, activation response coming out, 7
by, 116–17 congenital adrenal hyperplasia (CAH),
plasticity in, 18 79–82
sexual behavior controlled by, 22 in clinical cases, for females, 132–35,
sexual differentiation and, critical 133f
period for, 41, 89–90 genital formation and, in females, 134
brain structure, gender differences in, genital structure masculinization, in
118–28 young females, 80–81, 80f
anterior commissure, 120–21, 120f conscious learning, sexual orientation
corpus callosum, 121–22 from, 16
SCN, 119–20 corpus callosum, 121–22
Broca, Pierre Paul, 21 critical period, of sexual differentiation,
Byne, Bill, 123–24, 126 41, 129
CAH. See congenital adrenal hyperplasia Davidson, Julian, 98

castration, 70 defeminization. See masculinization, of
male sexual differentiation and, 41 sexual differentiation
SDN of preoptic area after, 55 deoxyribonucleic acid (DNA), 21
CDC. See Centers for Disease Control and depression, 62
Prevention DES. See diethylstilbestrol treatment
Index 181
DHT. See 5α-dihydrotestosterone females, homosexuality in, 4

Diamond, Milton, 14, 25 bisexuality and, 11
diethylstilbestrol (DES) treatment OAEs expression, 112
homosexuality and, 135–36, 135f stability of, 10
during pregnancy, 81 XQ28 region of X chromosome, 145
dizygotic twins, sexual orientation among, females, sexual differentiation in
143–45, 144f, 158 chromosomal influences on, 36–39
DNA. See deoxyribonucleic acid gonadal development, 39
Dörner, Günter, 43, 102, 131 masculinization of, 42
drawing types, influenced by sex phenotype development, 40
steroids, 85 testosterone injections and, 41
dyslexia, 62 female embryos, sexual determination
in, 74
early sexual experiences, sexual CAH and, 79–81
orientation influenced by, 12–14 female sexual behavior
impregnation concept, 13 activity peaks, 96–97
education. See social learning androgen/estrogen combination
emotional intimacy, sexual attraction treatment for, 95–96, 96f
and, 11 during estrus, 33
endocrinology. See behavioral menstrual cycles as influence on,
endocrinology 96–97, 98f
erections. See nocturnal erections, increase sexual attraction and, 97
of after testosterone treatment testosterone and, 95–96
estra-1,3,5 (10), 16-tetraine-3-ol (EST), 117 transsexuality, 67
estradiol fMRI. See functional magnetic resonance
in animal studies, for sexual imaging
orientation, 50 Freud, Sigmund, 14–15, 157
17β-estradiol, 27, 35 functional magnetic resonance imaging
female secondary sexual characteristics (fMRI), 22, 117
and, 31
sexual behavior influenced by, 33 gender differences, in humans, 61–68.
testosterone transformed into, 29 See also females, sexual
estrogen differentiation in; males, sexual
in animal studies, as treatment for differentiation in
sexual orientation, 51, 51f, 54 for aggressive behavior, 65f, 106
female puberty induced by, 25, 92 Alzheimer’s disease, 62
female sexual behavior and, androgen insensitivity, 62–63
combination treatment with for autism, 62
androgens, 95–96, 96f for brain structure, 118–28
human sexual behavior influenced in cognitive abilities, 65, 105–7
by, 29 for depression, 62
olfactory stimuli modulated by, 28 DHT transformation, 62
production of, 29 for dyslexia, 62
estrus, female sexual behavior during, 33 effect size in, 64f, 65
ethically deviant societies, homosexuality for height, 65f
in, 160 homosexuality and, 8–11
182 Index
gender differences (Cont’d) Great Britain, homosexuality in, 14

index/ring finger length ratio, 107–9, group theory, of evolution, 152–53
108f
intercultural comparisons of, 64–65 Hefez, Serge, 19
manual lateralization and, 106–7 height, gender differences in, 65f
in math skills, 64, 65f, 66 Holloway, Ralph, 121
metabolic aspects of, 61–62 homosexuality. See also females,
for OAEs, 110, 111f, 112 homosexuality in; males,
origin identification in, 63–66 homosexuality in
SDN in, 65 among adolescent boys, 13
secondary sexual characteristics, 61–62 AIS and, 138
for sexual behavior, 63–66 in animal studies, as preference
sexual identity and, 66–68 expression, 51–52, 54
sexual orientation and, 66–68 anterior commissure development and,
for startle response expression, 112–13 120–21, 120f
for toy choice, 65f aspects of, 6
variable amplitudes for, 63–66, 65f brain structure and, 118–28
for verbal fluency, 65f, 66, 106 cloacal exstrophy and, 137–38, 159–60
visual-spatial tasks and, 105–6 cognitive abilities and, 104
gender identity corpus callosum development and,
sexual orientation and, 17 121–22
transsexuality, 5 critical period of sexual differentiation
gender-nonconformism, 15 and, 41, 129
gender roles, 5 cures for, 16
genetic encoding, in biological definition, 3
development, 20–21 DES treatment, during pregnancy,
genitals, formation of 135–36, 135f
CAH and, in females, 134 deterministic principle and, 8
growth of, testosterone as influence DHT deficiency and, 136–37
on, 31 among dizygotic twins, 143–45,
intersex, 23 144f, 158
Müllerian ducts, 37, 39 in ethically deviant societies, 160
sex steroids and, 19, 74, 155 evolutionary significance of, 151–53
Wolffian ducts, 37 expression of, 6–7
genomic action, 31 false theories for, 157–58
Gladue, Brian, 114 gender and, 8–11
GnRH. See gonadotropin releasing in group theory of evolution,
hormone 152–53
gonadotropin releasing hormone (GnRH), hormonal inducement of, 7
33, 99–100 hormonal levels, 101–2
gonads, 29 as immunological mechanism, 141–53
development of, 72 INAH development and, 123–28, 123f
female sexual differentiation and, 39 index/ring finger length ratio, 107–9,
ovarian development from, 72 108f
testis development from, 72 individual classifications of,
Gorski, Roger, 44–45, 122–23 methodology for, 128
Index 183
Kinsey scale for, 9–10 human sexuality. See also bisexuality;

latency of, 103 homosexuality; transsexuality
LH release and, 113–15, 115f clinical case studies, 6
long bone length, 109–10 dimensions of, 3–8
among monozygotic twins, 143–45, hypogonadism, 99f
144f, 158 hypospadias, 79
OAEs and, 110, 111f, 112 hypothalamus gland, 29, 43
older brothers effect, 148–51, 149f in animal studies, for sexual orientation,
open display of, in society, 7 48–50
parental attitudes toward, 15
percentages of, in all societies, 9, 13t identity. See gender identity; sexual
pheromones and, 117–18, 118f identity
prenatal hormonal determinism, 102–4 impregnation concept, for sexual
prenatal stress theory for, 131–32 orientation, 13
right-handedness and, 106–7 INAH. See interstitial nuclei of the
SDN of preoptic area and, 122–28 anterior hypothalamus
sexual attraction and, 7 index/ring finger length ratio
societal attitudes toward, 161–62 among monozygotic twins, 109
startle response expression and, sex steroids as influence on, 107–9, 108f
112–13 intersex genital formation, 23. See also
testosterone concentrations and, 102, Money, John; Reimer, Bruce
129, 141–42, 142f interstitial nuclei of the anterior
unavailability of opposite sex partners hypothalamus (INAH)
and, 6–7 brain location for, 126–28
XQ28 region of X chromosome and, homosexuality and, 123–28, 123f
145–48 invoked experiences, 130
hormonal embryonic imprinting, 102–3
hormones. See also testosterone Katz, Jonathan, 16
adrenal gland, 29 Kinsey, Alfred, 8
in animal studies, sexual orientation Kinsey scale, for homosexuality, 9–10
influenced by, 47–48, 50–55
anti-Müllerian, 39 de Lacoste-Utamsing, Christine, 121
behavioral influences on, 19–20 left-handedness. See manual lateralization,
binding sites in, 28 gender differences and
in brain development, 19 lesbians. See females, homosexuality in
effector organ development, 28 Leydig cells, 74
embryonic imprinting and, 23–24 LH. See luteinizing hormone
homosexuality induced by, 7 long bone length, 109–10
hypothalamus gland, 29 lordosis, 42
pituitary gland, 29 in animal studies, for sexual
sexual behavior influenced by, orientation, 53
28–29, 34f luteinizing hormone (LH), 113–15, 115f
sexual motivation and, 35–36
human embryos, sex determination in, males, homosexuality in
75–81 in adolescence, 13
androgen insensitivity, 76–77 methodology variation, 4
184 Index
males, homosexuality in (Cont’d) sexual orientation in, 143–45,

older brothers effect, 148–51, 149f 144f, 158
in populations, variable percentages motivation. See sexual motivation
of, 4 Müllerian ducts, 37, 39, 74
social favoring structures for, 13
stability of, 10 National Health Statistics Center (NHC),
testosterone levels, 102 8–9
XQ28 region of X chromosome, 145–46, the Netherlands
146f homosexuality in, 14
males, sexual differentiation in transsexuality in, 67
castration and, 41 NHC. See National Health Statistics
chromosomal influences, 36–37 Center
male embryos, sexual determination in, 74 nocturnal erections, increase of after
DHT deficiencies and, 77–79, 78f testosterone treatment, 70
male-oriented females (MOFs), 58 Nottebohm, Fernando, 43–44
male-oriented males (MOMs), 57–59
male sexual behavior OAEs. See oto-acoustic emissions
GnRH treatment, 99–100 older brothers effect, 148–51, 149f
hormonal mechanisms, 34f birth weight factors, 150
transsexuality, 67 trisomy 21 and, 150
manual lateralization, gender differences ontogenesis, in animal studies, for sexual
and, 106–7 orientation, 55
masculinization, of sexual differentiation operant conditioning, 15
AIS as influence on, 77 oSDN. See ovine sexually dimorphic
of females, 42 nucleus of preoptic area
of males, 42–43 oto-acoustic emissions (OAEs), 110,
from SDN of preoptic area, 45 111f, 112
masturbation, increase of after in lesbians, 112
testosterone treatment, 70 ovariectomies, 33
math skills, gender differences in, 64, animal studies of, for sexual
65f, 66 orientation, 49
medial preoptic area, 36 ovaries, development of, 39
in animal studies, lesions as influence from gonads, 72
on, 50 ovine sexually dimorphic nucleus (oSDN)
sexual differentiation and, 43–46 of preoptic area, 58–60, 59f
Men and Women: Do We Have the Same
Brain (Vidal), 92 parents, 15–16
menstrual cycles, female sexual behavior attitudes toward homosexuality, 15
influenced by, 96–97, 98f sexuality of, as influence on sexual
missionary position, 4 orientation, 16
MOFs. See male-oriented females Parkinson’s disease, 21–22
MOMs. See male-oriented males penile ablation, 26
Money, John, 24–26, 133 PET. See positron emission tomography
monozygotic twins pheromones, 116–18
index/ring finger length ratio AND, 117
among, 109 brain activation and, 116–17
Index 185
EST, 117 secondary sexual characteristics

homosexuality and, 117–18, 118f in females, 31
Philippines, homosexuality in, 14 gender differences and, 61–62
phimosis, 24 serotonin, 22
pituitary glands, 29 Sertoli cells, 38, 74
positron emission tomography (PET), 22, sex chromosomes, anomalies of, 72–73
117 sex determination, 36–39
pregnancy additional sex chromosomes, 72
DES treatment during, 81 in female embryos, 74
homosexuality expression and, with in humans, 72–74, 76–81
DES treatment, 135–36, 135f in male embryos, 74
Premarin, 114 SRY gene, 37–38, 73
prenatal stress theory, homosexuality and, sex-determining region of Y chromosome
131–32 (SRY) gene, 37–38, 73
in animal studies, 131 sex differences, animal studies on, 6
preoptic area. See medial preoptic area sex steroids, 29–32, 155. See also
progesterone, 31 androgens; estrogen; testosterone
sexual behavior influenced by, 33 activating effects, 91–100
progestogens, 29 androgens, 29–31
psychoanalytic theories, for sexual in animal studies, for sexual orientation,
orientation, 14–15 54–55
psychosexual development, 14–15 behavioral differentiation, 81–87
puberty behavioral traits influenced by, in
in females, sex steroids as influence on, humans, 75
91–92 binding site activity, 32–33, 32f
induced by estrogen, in females, CAH, in female embryos, 79–81
25, 92 DES during pregnancy, 81
DHT deficiency, in young males,
rats, in animal studies, for sexual 77–79, 78f
orientation, 55–56 drawing types influenced by, 85
Reimer, Brian, 24–25 in experimental human subjects, 69–72
Reimer, Bruce, 24–25 during female puberty, 91–92
relativism, 16–17. See also social forms of, 91–92
constructivist theories, for sexual functions of, 91–92
orientation genital formation, 19, 74, 155
ribonucleic acid (RNA), 21 index/ring finger length ratio
right-handedness, homosexuality and, influenced by, 107–9, 108f
106–7. See also manual lateralization, individual correlations for, in humans,
gender differences and 94–98
Rissman, Emilie, 52 irreversible effects of, 91
RNA. See ribonucleic acid long bone length, 109–10
metabolic pathways for, 30f
salivary testosterone, 94, 95f multiple peaks of, during embryonic
SCN. See suprachiasmatic nucleus development, 87–89, 88f
SDN. See sexually dimorphic nucleus of organizing effects, 69–90
preoptic area organizing effects of, 82
186 Index
sex steroids (Cont’d) chromosomal influences on, 36–38

peripheral physical differences, 107–10 critical period of, 41
physical traits influenced by, in humans, embryonic development of, 38f
73–74 for internal sexual organs, 37f
play behaviors, 83 neural structures for, 43–46
sexual behavior influenced by, 39–43, during ontogenesis, 40f
81–87, 92–93 SDN of preoptic area, 43–46, 55, 89–90
sexual identity influenced by, 5 stages of, 39
sexual orientation and, 17 testosterone and, 39
temporal correlations for, in humans, sexual disorders, 70
94–98 sexual identity, 5
toy choice influenced by, 83–85, 83f gender differences and, 66–68
visual stimuli influenced by, 86–87, 87f sex steroids as influence on, 5
sexual attraction social learning as influence on, 23–26
distribution of, in US, 9f transsexuality, 5
emotional intimacy and, 11 sexuality. See human sexuality
female sexual behavior and, 97 sexually dimorphic nucleus (SDN) of
homosexuality and, 7 preoptic area
sexual behavior. See also female sexual in animal studies, 44f, 55–56
behavior; male sexual behavior castration and, as influence on sexual
behavioral endocrinology as influence differentiation, 55
on, 27–29 gender differences influenced by, in
biological evolution of, 93–94 humans, 65
brain development as influence on, 22 homosexuality and, 122–28
clinical case analysis, 98–100 INAH in, 123–28, 123f
cognitive aspects of, 92 masculinization from, 45
consummatory component of, 35–36 oSDN, 58–60, 59f
estradiol as influence on, 33 sexual differentiation and, 43–46, 55,
estrogen as influence on, 29 89–90
gender differences in, for humans, sexual motivation, 4
63–66 hormones and, 35–36
hormonal influences on, 28–29 sexual orientation, 4–5. See also clinical
limits of, 4 cases, sexual orientation in
masculinization of, 41 in animal studies, 47–60
missionary position, 4 behavioral theories for, 15–16
progesterone as influence on, 33 from conscious learning, 16
sex steroids as influence on, 39–43, among dizygotic twins, 143–45,
81–87, 92–93 144f, 158
sexual motivation in, 35–36 early sexual experiences as influence on,
social learning as influence on, 63–64 12–14
toy choice and, 83–85, 83f gender differences and, 66–68
in women, testosterone levels and, gender identity and, 17
95–96 gender-nonconformism, 15
sexual differentiation, 36–39. See also in monozygotic twins, 143–45,
masculinization 144f, 158
brain development and, 89–90 older brothers effect, 148–51, 149f
Index 187
parental sexual orientation as influence testosterone

on, 16 conversion into DHT, 78
psychoanalytic theories for, 14–15 effector organs influenced by, 28
sex steroids and, 17 during embryonic development, 81
in sheep, 57–60 estradiol and, transformation into, 29
social constructivist theories for, 16–17 female sexual differentiation and, after
social learning as influence on, 17–20, injection, 41
23–26 homosexuality and, 102, 129, 141–42,
trisomy 21 and, 150 142f
from unconscious learning, 16 insufficient development of, 70
XQ28 region of X chromosome, 145–48, in limbic system, 33
146f male genital growth influenced by, 31
sexual preference. See sexual orientation masculinization of females from, 42
sheep, sexual orientation in, 57–60 masturbation episodes, increase of after
MOFs, 58 treatment, 70
MOMs, 57–59 multiple peaks of, 87–89, 88f
oSDN and, 58–60, 59f nocturnal erections, increase of after
testosterone as influence on, 60 treatment, 70
Sigmundson, Keith, 25 salivary, 94, 95f
Sinclair, Andrew, 73 sexual differentiation and, 39
single-sex institutions, homosexuality in, in sheep, sexual orientation influenced
6–7, 14 by, 60
Skinnerian conditioning, 15 in women, sexual behavior and,
social constructivist theories, for sexual 95–96
orientation, 16–17 TFM. See testicular feminizing mutation
social learning Thailand, homosexuality in, 14
sexual behavior influenced by, 63–64 toy choice
sexual identity influenced by, 23–26 gender differences in, 65f
sexual orientation influenced by, 17–20, sexual steroids and, 83–85, 83f
23–26 transsexuality
societal attitudes, toward homosexuality, definition, 3
161–62 among females, 67
SRY gene. See sex-determining region of Y gender identity, 5
chromosome gene among males, 67
startle response expression, 112–13 in the Netherlands, 67
steroids, 29. See also hormones; sex occurrence of, 67
steroids transvestitism, 3
cellular mechanisms of, 33–34 trisomy 21, 150
intracellular metabolism of, 35 twins. See monozygotic twins
organizing effects of, 41–42
suprachiasmatic nucleus (SCN), 119–20 unconscious learning, sexual orientation
Swaab, Dick, 122 from, 16
United States (US)
tabula rasa, 19, 23 homosexuality in, percentage of, 14
testes, gonadal development into, 72 sexual attraction in, distribution
testicular feminizing mutation (TFM), 52 patterns in, 9f
188 Index
verbal fluency, gender differences in, 65f, XQ28 region, of X chromosome

66, 106 candidate genes, 146–48
Vidal, Catherine, 92 female homosexuality and, 145
visual-spatial tasks, by gender, 105–6 male homosexuality and, 145–46,
146f
Ward, Byron, 131 maternal contribution of, 147
Wernicke, Carl, 21 paternal contribution of, 147
Wolffian ducts, 37, 74

The Biology of Homosexuality

Uploaded by

Copyright:

Available Formats

The Biology of Homosexuality

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

The Biology of Homosexuality

Uploaded by

Copyright:

Available Formats

The Biology of Homosexuality

Oxford Series in Behavioral Neuroendocrinology

Luis Miguel Garcia-Segura, Hormones and Brain Plasticity

Copyright © Jacques Balthazart 2012

Library of Congress Cataloging-in-Publication Data

Balthazart, J. (Jacques), 1949-

Typeset in Minion Pro

I would like to thank my wife, Claire, and my daughter-in-law, Nathalie, not

Introduction: Why This Book? ix

1. Sexuality 101: The Basics 3

2. Sexuality and Sexual Orientation: Learning or Biology? 12

3. The Hormonal Control of Sexual Behavior 27

4. Biological Determinism of Sexual Orientation in Animals 47

5. Gender Differences in Humans 61

6. The Effects of Sex Steroids in Humans: Organizing Effects 69

7. The Effects of Sex Steroids in Humans: Activating Effects 91

8. Sex Differences Suggest Homosexuality Is an

9. Sexual Orientation in Clinical Cases 130

10. A Genetic or Immunological Mechanism

11. General Conclusions 154

The attitude of Western societies toward homosexuals has changed consider-

THE FOUR DIMENSIONS OF HUMAN SEXUALITY

The first aspect of human sexuality is linked to the performance of sexual

attraction, or vice versa. This orientation becomes difficult to classify because it

in adulthood. But though many popular books claim to explain homosexuality

MALE AND FEMALE HOMOSEXUALITY: HOMOSEXUALITY

100 Men Women

2. Mostly opposite sex

Sexuality and Sexual Orientation

EDUCATION AND ENVIRONMENT CANNOT SUFFICE

EFFECT OF EARLY SEXUAL EXPERIENCES

Table 2-1. Percentages of Homosexual Men

Great Britain 5.0–9.0

THEORIES BASED ON LEARNING DERIVED FROM BEHAVIORISM

THEORIES BASED ON “SOCIAL CONSTRUCTIVISM”

truth itself in scientific knowledge would be related to a particular historical

BIOLOGICAL CONSTRAINTS ON LEARNING

will probably always remain impossible to prove. Learning therefore enhances

learning) to the control of human behavior. I mainly focus in this context on

THE UNITY OF LIFE

Genetic information is encoded in all living creatures on earth by special mole-

NEURONS GOVERN ALL OUR ACTIONS

specialized in the treatment of spatial memory) that is larger than in control

EDUCATION DOES NOT DETERMINE SEXUAL

The Hormonal Control of Sexual Behavior

specific neural circuits. This effect can be achieved in particular by changing

WHICH HORMONES ARE INVOLVED IN THE CONTROL

As already mentioned, testosterone, or similar androgenic compounds, rep-

MECHANISMS OF ACTION OF SEX STEROIDS

CH3 CH3 C=O

an enzyme called aromatase. This metabolic conversion plays a crucial role in

transformation plays a key role in the action of androgens, particularly in the

BINDING SITES AND ACTIVITY OF SEX STEROIDS IN THE BRAIN

In females, estradiol and progesterone induce behaviors associated with sexual

CELLULAR MECHANISMS OF ACTION OF STEROID HORMONES

activation of sexual behavior. Based on the continuity of evolutionary and brain

INTRACELLULAR METABOLISM OF STEROIDS

HORMONES AND SEXUAL MOTIVATION

an aversive stimulus (electrified grid) or the activity in an operant conditioning

SEX DETERMINATION AND SEXUAL DIFFERENTIATION