Instead of trying, impossibly, to divorce ourselves from culture, claiming, somehow, that we scientists can create value-free knowledge, suppose we em­braced our cultural locations? Suppose we strove to create stories about ro­dent sexuality that look at everything from genes to culture (rat culture, that is) as part of an indivisible system that produces adult behavior. Such a narra­tive would resemble ‘‘Dungeons and Dragons’’ more than ‘‘Little Red Riding Hood.’’ The elements of such a narrative already exist in the scientific litera­ture. What remains is to draw them together.

In broad outline the O/A theory goes like this. During the pre – (guinea pigs) or peri- (rats) natal period, hormones (usually testosterone, but some believe estrogen will do the trick) permanently affect brain development. Somehow (although it is still not at all clear how),148 neural structures in the brain become dedicated to future behaviors such as mounting or lordosis (see figure 8 .y). Puberty hits, activates the previously organized neural pathways, and the behavior becomes visible. Beach, Young, and the many fine animal behaviorists who have followed in their footsteps have all known that this pic­ture is static and oversimplified, and fails to integrate the developing animal into its environment. Why haven’t they proposed more dynamic accounts of rodent sex?

The experiments exist. It is the will and the theory that are missing. As long as one insists that in the interaction between nature and nurture, at some early moment in development nature starts it all, while only later does nur­ture tinker, a resolution is impossible. Often scientists talk in terms of ‘‘pre­dispositions’’: natural inclinations that experience and social interactions can modify, but with greater or lesser difficulty. One thoughtful review of the

Understanding rat sex

figure 8.6: An enriched account of behavioral development in rodents.

interactions of social and hormonal influences on sex differences in rhesus monkeys concludes that nature needs nurture and that nurture needs na­ture.149 That’s almost right, but still the nature/nurture dualism persists. What I suggest is that we switch our vision (sort of like putting on 3-D glasses), so that we see nature and nurture as an indivisible, dynamic system. Such a systems approach to developmental psychology is not new, merely underre – ported.150

Animals develop in an environment. In utero, that environment includes the mother’s physiology. A mother’s body chemistry results from her behav­ior. What does she eat? Does she encounter stress? How do her hormones respond to such experiences?151 Life experience before birth may also depend on how many wombmates an animal has, and even on whether it is sandwiched in between opposite-sex siblings.152 Furthermore, the fetus’s own movements and spontaneous neural responses can affect its development.153 But that’s only the beginning. Rodents have lots of brothers and sisters, and the number and type of sibling affects their behavior after birth.154 So too does their inter­action with their mothers. The entire life cycle, from before birth through weaning and juvenile play into puberty and adulthood, provides opportunities for experiences key to the development of the rodent sexual response (see figure 8.6).

How might life experiences and hormones co-produce adult behavior?

Some concrete examples illustrate. In one of the classic O/A papers, Harris and Levine reported that hormone-treated females have smaller, rounder, and otherwise abnormal vaginal openings;155 others found that all the female rats exposed perinatally to androgen had closed vaginas, and most (91 percent) had enlarged clitorises.156 Furthermore, testosterone-exposed females were larger than unexposed females.157 Such physical differences could easily lead to different learning experiences. Larger females may learn to mount more often, and those with enlarged genitalia may find certain forms of sexual activ­ity to be exceptionally pleasurable. De Jonge, for example, provides evidence that progesterone affects a female rat’s mating interest only if a sexual reward is available. A closed vagina may make a female less receptive to mounting attempts, resulting in fewer juvenile learning experiences and lowered adult lordosis. Carefully timed chemical treatments, however, can produce an ani­mal with normal-looking external genitalia, but exhibiting altered behavior. Therefore, changes in behavior do not result only from altered genitalia.158

Beach, Young, and many others have provided abundant evidence for the importance of social interactions in the development of mating behaviors. Animals bred in isolation are sexually incompetent,159 and just having a room­mate is not enough. The kind of companion matters, too. What components of upbringing contribute to developing sexual behaviors? In one set of experi­ments, 15 percent of normal male rats raised in isolation exhibited lordosis; if raised with females of the same age, however, half of the males exhibited lor­dosis; and if raised with males, 30 percent exhibited lordosis.160 The reasons for such differences are unknown; but behaviors such as lordosis, the develop­ment of which involve perinatal hormones, also depend heavily on the circum­stances of upbringing.161

What about the five senses? Testosterone affects more than just the genita­lia and the brain. For example, male and female rat pups smell different. This testosterone-dependent difference induces mother rats to lick their male off­spring more frequently and vigorously, especially in the anogenital region. The licking, in turn, affects adult male sexual behavior. Males raised by moth­ers whose nasal passages had been blocked (and who thus licked them less) took longer to ejaculate and had a longer refractory period between ejacula­tions. The psychologist Celia Moore and her colleagues also report that males raised by mothers that licked less had fewer spinal motor neurons in a region of the spinal cord associated with ejaculatory behavior. In other words, the development of a part of the central nervous system (a specific region of the spinal cord) is influenced by maternal behavior. Here the effect of testosterone is only indirect (on pup odor that stimulates licking).162

Young male rats also spend more time grooming their genitalia than do females, and this additional stimulation speeds up the journey to puberty. Similarly, female mice mature more quickly if they stay in the vicinity of cer­tain puberty-enhancing odors.163 In other words, an individual rat’s own growth depends in part on his or her behavior. Nature and nurture are not separate here. Salt and water balance, pup leg extension, and urine release— each of which differs in male and female pups—all affect licking behavior. The brain, it seems, is one among several elements affected by early hormone exposure. Some elements are anatomical, some physiological, some behav­ioral, and some social. They all form part of a unitary system.164

Hormone treatment also affects muscle and nerve development outside the brain. For example, male rats have a set of three muscles, needed for erection and ejaculation, attached to the penis. Nerve cells growing out of the lower spinal cord connect to these muscles. The muscles and nerves accumu­late androgen that the muscles require for sexual function. In female rats, one of these muscles degenerates shortly after birth unless it receives androgen during a particular period.165 We don’t know whether testosterone-mediated changes in female mounting behavior might be related to the presence of this muscle, but we do understand that how much sex a male rat has affects the size of the motor neurons that innervate these muscles. In this example, ‘‘differences in sexual behavior cause, rather than are caused by, differences in brain structure.’’166

And what about rat multiculturalism? Again, Beach, Young, and others showed years ago that different genetic strains exhibit different patterns of sexual activity.167 An adequate model of sexual behavior must include individ­ual genetic differences and incorporate the effects of an extensive period of maternal interaction, as well as experience gained from littermate, cagemate, and partner interaction. In recent years only the studies of de Jonge and her colleagues and of Moore have analyzed hormonal effects on behavior in this more complex framework, but even their work still takes place in a severely oversimplified environment: the laboratory. There is no guarantee that hor­monal effects on mating behavior, proven in restricted laboratory situations, have much explanatory power in natural populations.168

The O/A pretty much ignores possible hormone effects from shortly after birth until puberty. Whether or not hormones are important between birth and puberty varies with the species. In some, ovarian hormones may affect the development of sex-related behaviors more or less continuously until puberty. Measures of feminine mating behaviors were higher in both female and cas­trated male rats that had ovaries implanted at various times prior to puberty. Implanted animals also had smaller body weights at puberty, a result propor­tional to the length of time the ovary had been implanted.169 Furthermore, secretions during postnatal development can change the response of adult fe­male rats to estrogen.170

Although many mammals have an initial discrete testosterone-sensitive pe­riod, some do not. Pigs, for example, respond to testosterone from birth until puberty, and the effects of injected hormones on behavior progress with time. Since juvenile pigs frequently engage in sexual play in both male-male and male-female combinations, it seems especially possible that experience and hormone co-produce adult behaviors.171 In rats, both masculine copulatory responses and an increased orientation toward other females can result either from specific sexual experience during adulthood or from hormonal treat­ments during puberty or early adolescence.172 In short, the fact that varying levels of specific hormones circulate during the course of an individual’s life span affecting nervous system anatomy and function warrants a life-span ap­proach to understanding the role of hormones in the development of sex differences in neural structure. A life cycle, systems account of animal devel­opment does not ignore the weeks between birth and puberty, and a more complete theory opens new experimental vistas, ones less visible under the O/A regime.173

In an article on the sexual differentiation of the nervous system, the neuro­anatomist C. Dominique Toran-Allerand writes: ‘‘It is generally believed that testicular androgens exert an inductive, or organizational influence in the developing CNS [central nervous system] during restricted (critical), late fe­tal or early postnatal periods of neural differentiation, at which time the tissue is sufficiently plastic to respond permanently and irreversibly to these hor­mones’’174 In their 1959 paper, Young and colleagues concluded their experi­ments after testing treated guinea pigs twice, once at six to nine months and again at one year of age. Guinea pigs, however, can live as long as eight years. Yet there are no lifelong longitudinal studies of guinea pig mating behaviors under different hormonal and experiential situations. This is true as well for virtually all the other rodents studied in similar fashion, although the claim of permanence may be more accurate for animals such as mice, which nor­mally live for only one to two years.175

Behaviors that show up in the months immediately following puberty may change with subsequent life experience. For instance, perinatally andro – genized female rats, under certain circumstances, will show a lowered fre­quency or intensity of lordosis. Extensive testing, however, can overcome such changes.176 Similarly, testosterone can typically activate mounting in de­velopmentally normal female rats.177 As one reviewer states, ‘‘the essential ‘wiring’ for these behaviors persists. … In this sense Beach was correct in questioning the idea that perinatal steroids change the essential structure of the nervous system.’’178

The notion of permanence faces other troubles as well. Activating effects were originally thought to be transitory, lasting from a few hours to a few days. In contrast, permanent organization events are supposed to last a lifetime. In practice, this has meant several months to about a year. But how does one classify hormonal effects on the brain that last for weeks rather than days or months? A variety of such cases exists for both songbirds and mammals. In these examples, particular brain structures respond to hormone increases, even in adulthood, by growth and to hormone reduction by shrinkage.179 If the brain can respond to hormonal stimuli with anatomical changes that can endure for weeks or even months, then the door opens wide for theories in which experience can play a significant role. Even rodents engage in extensive periods of social play, activities that influence the development of the nervous system and future behaviors. It is at least plausible that play activities alter hormone levels and that the developing brain can respond to such changes.180 Hormonal systems, after all, respond exquisitely to experience, be it in the form of nutrition, stress, or sexual activity (to name but a few possibilities). Thus, not only does the distinction between organizational and activational effects blur, so too does the dividing line between so-called biologically and socially shaped behaviors.

Humans are learners, and proudly so. We are, arguably, the most mentally complex of all animals (no offense meant to the great apes, who might argue with us if they could speak). It seems ironic, therefore, that our most promi­nent and influential accounts of the development of sexual behaviors in ad­vanced mammals omit learning and experience. Because the control of hor­mone synthesis differs between primates and other species,181 a case can be made that studies on the hormonal basis of sexual behaviors in nonprimates tell us little, if anything, about primates, including humans.182 As I turn in the final chapter to theories of human sexuality, I make a broader claim: that the theories we have derived from rodent experimentation are inadequate even for rodents.

toward a theory of human sexuality