Katherine E. Wynne-Edwards is a professor of biology at Queen's University in Kingston, Ontario, who studies hormonal changes in expectant fathers and hormone-behavior interactions in other animal models. She offers the following explanation:

When pregnancy symptoms such as nausea, weight gain, mood swings and bloating occur in men, the condition is called couvade, or sympathetic pregnancy. Depending on the human culture, couvade can also encompass ritualized behavior by the father during the labor and delivery of his child. Couvade has a long anecdotal history and is named from the French verb couver, which translates as ¿to hatch¿ or "to brood." The phenomenon has received attention from biologists only quite recently, however.

Estimates of the frequency of couvade are hard to obtain because of the low rate of reporting symptoms. For example, a research team led by Anne Storey of Memorial University in Newfoundland found that when wives were asked about their husband's experiences a higher incidence of couvade was reported than when the husbands answered the same questions at the same time. Across a wide range of studies--and an equally wide range of definitions of what constitutes couvade--estimates of the frequency in modern Western populations range from under 20 percent to more than 80 percent of expectant fathers.

Society and health professionals all show a lot of interest in a pregnant woman. She is encouraged to talk about any symptoms of her pregnancy, even common ones she is not experiencing. At home, the conversation can range from frustrated incapacitation as a result of her symptoms to boundless joy in anticipation of a child. Thus, it is not surprising that a large number of mental health professionals have considered a range of hypotheses--from jealousy about a man's inability to carry a child to guilt over having caused this transformation in his partner to selfish attention seeking--as the root causes of couvade.

Of course, there are other obvious origins for at least some of the symptoms. For example, if the pregnant wife does much of the shopping and cooking, her cravings, as well as the increasing food intake she needs during pregnancy, are quite likely to result in weight gain for her husband as well as associated symptoms of heartburn and indigestion. There are also studies suggesting that men who have deep empathy toward their pregnant partner and are prone to couvade symptoms end up with strong attachments to their child. If this is the case, then the symptoms might either stimulate, or result from, underlying biological processes that are involved in social attachment.

In recent years animal models of social monogamy (defined as a strong social preference for a single partner), such as prairie voles, have contributed a great deal to our understanding of the neurobiology of love and attachment. Neuropeptides, including oxytocin and vasopressin, are now known to play important roles in the formation and maintenance of strong pair bonds. Mammalian parental behavior involves the formation of a strong social bond to the infant and also engages hormonally stimulated neural circuits. Parental behavior, however, currently appears to depend on sex steroid hormones and prolactin more than on oxytocin and vasopressin. In addition, there is hormonal data from nonhuman primates and naturally paternal rodents--including California mice and dwarf hamsters--that indicates a positive association between the expression of paternal behavior and increases or decreases in prolactin, estradiol, testosterone, progesterone and cortisol concentrations. For example, male mice that lack a gene for the progesterone receptor are not infanticidal toward unrelated pups and, instead, retrieve and huddle over them. Unfortunately, few experiments have established causal relationships between hormonal changes and the behavior.

The situation is similar with respect to our understanding of the hormonal experiences associated with fatherhood in men. Since 2000, several studies have reported hormonal differences between expectant fathers, men in committed relationships and men who are single. There is no doubt that testosterone concentration is lower in the men in relationships, but it is unclear whether men have a decrease in testosterone after the relationship begins or whether men with lower testosterone are more likely to enter into stable relationships. Similarly, there are hormone changes associated with fatherhood. Prolactin is highest in men in the weeks just before the birth, testosterone is lowest in the days immediately after the birth, estradiol levels increase from before to after the birth, and cortisol peaks during the labor and delivery (although it remains an order of magnitude below the hormonal experience of the laboring mother). Alison Fleming of the University of Toronto and her colleagues have shown that maternal cortisol is linked to social bonding with the infant and to postpartum depression, whereas Storey has shown that paternal prolactin is positively associated with the self-reporting of couvade symptoms and powerful emotional responses to infant stimuli. Hormone changes in expectant fathers therefore involve the same hormones that are changing in an expectant mother. They are also the same hormones that are implicated in animal models of parental behavior. In addition, it is known that men are not reflecting a "muted" version of the hormonal experiences of their partner: Within couples, day-to-day hormone status is not correlated during pregnancy or after the birth.

Unfortunately, these data remain correlations, and the exact role of hormones in facilitating paternal behavior or causing couvade symptoms in expectant fathers remains unknown. It is certainly tempting to look to hormones for the biological root of couvade symptoms, but caution is needed. Other events also happen around a pregnancy, especially the birth of a first child, and could independently affect hormone concentrations. Changes in sexual activity, shifts in the social priorities of the couple, time off work, or the arrival of a mother-in-law for a potentially stressful extended visit are obvious candidates. Of course, even if the stimuli causing the hormonal changes are not the result of an approaching birth the hormone changes might produce couvade symptoms and/or facilitate a father's social bond with his child. Either way, this kind of research has quietly expanded the horizons for research on hormones in men--testosterone alone is clearly no longer the sum of the man.