How to Turn a Philanderer Into a Father

My younger brother — middle-aged, heterosexual, and single — just moved to Atlanta. He’s a handsome guy, intelligent, funny, and keeps in great shape. Loves to flirt. The problem? He’s in the online dating pool trap, and the bulk of his real-world social interactions occur in the gym. My wife and I, helpfully we think, suggest that he needs to get a dog. But our reasons are different. 

My wife thinks if he gets a dog he’d have great companionship. This is true. I think that he should get a dog because it’s a great way to meet women: If you walk around with your dog, you are advertising that you are a responsible caregiver, a great future father. I think a dog could function as a surrogate baby.  

I’ve studied the biology and behavior of fish, rats, lemurs, multiple species of monkeys, and humans. I can tell you that fathers are rare. I’m not talking about male individuals who manage to fertilize an egg. They are plenty. They mate and then they take off to find another mate. I’m talking about males who stick around to care for their offspring in one way or another. Why do so few males across the animal kingdom want to take care of their own offspring, particularly among mammals? What must happen to change a species or an individual so that the male exhibits parental care? Let’s start at the very beginning, with fish.¹

For fish, the standard is to spawn — the female squirts out her eggs, the male squirts out the sperm — and then they move on. The baby fish don’t need anyone to care for them. It’s a numbers game, the same one sea turtles play: Lay a huge number of fertilized eggs on the beach and then hope that some of them survive a gauntlet of dangers.

The next most common strategy in fish, however, is male care of the eggs — they protect them from predators. This means that if we define parental care as any behavior that increases the likelihood that an offspring will survive, then the very first parent exhibiting care was likely a male fish, not the female of the species.² The scenario for how this came about is pretty simple. 

A male fish who is territorial, protecting a little piece of the ocean that is attractive to females, can, without much additional energy, protect the eggs that were spawned in that territory. He’s effectively not doing much additional work, and the presence of the eggs is not going to hamper his ability to attract additional females. It might even enhance it. Besides protection from being eaten by predators, fertilized fish eggs don’t need much else. 

Of course, both sexes are capable of guarding eggs, but there are always energetic trade-offs to consider. Most fish keep growing throughout their lives. As  females get bigger, the number of energetically costly eggs they produce increases. The typical female fish doesn’t defend a territory for reproductive reasons; they leave their eggs after spawning. This may be why, in fish, male-only care is common—it doesn’t cost them much as they are sticking around to defend the area anyway. The male gets a larger net benefit by guarding the eggs. Males, however, don’t slavishly protect eggs in their territories. They can evaluate their own costs and adjust their behavior accordingly. For example, experiments with the bluegill sunfish showed that their decision to guard or not reflects how many eggs are present.³

Eventually, a type of lobe-finned fish — one whose fins had bones in it — found itself exploiting the shallows, where there were new feeding opportunities and where larger predators couldn’t go. To make a very long story short, this lobe-finned fish is the ancestor of a cornucopia of terrestrial vertebrates that left the water entirely to exploit new opportunities. Being on land, however, meant exposure to air. Eggs and sperm would dry out if land animals spawned the way fish do. So many terrestrial animals evolved mechanisms to fertilize their eggs internally, and two different ways to protect the fertilized egg. 

One solution was to make hard-shelled eggs that could keep moisture in but still allow for some gas exchange. The other solution, viviparity, was to keep the developing embryo inside the mother’s body, obviating the need for a protective shell.  

These strategic changes in response to being on land massively impacted the potential for fatherhood. Internal fertilization meant that males could not actually observe their sperm fertilize a batch of eggs, as one could during spawning in water. Paternity thus became uncertain. 

The mammalian mother commits to the responsibility of growing the embryo, at great energetic cost, and  providing milk to that infant until it is weaned. 

Viviparity also creates a long temporal gap between fertilization and when the offspring actually appear in the outside world. A lot could happen in that interval — for example, females can mate with other males, even as a strategy to generate paternity uncertainty. What are mammalian males to do in the meantime? Wait around until gestation is complete to help care for offspring that might not even be theirs?

It seems to make more sense to philander: to mate and take off, repeatedly, and hope that maybe you left some of your own offspring behind. It’s an economic decision where the currencies are time and energy. For the vast majority of mammals, the balance is such that they exhibit female-only care of offspring. So, what skews the decision-making of those 4-5% of mammalian males that actually do care for their offspring?

Your typical male house mouse hates kids, even his own. If he happened upon his own pups, he would try to kill them. There’s nothing that he can discern about them that would tell him that they’re his, so he assumes that they are not. Getting rid of them means that their mom is no longer expending energy nursing and caring for those pups; that energy shifts back to being fertile again. 

But here’s what’s funny: If experimenters take a typically infanticidal male house mouse and force him to live with the female after mating, the male will care for the pups born around 18 days later. What seemed like a fixed behavioral program — aggression towards pups — disappears, turns into something altogether different just by a simple change in context. 

Moreover, this effect can be titrated. The longer the males were forced to stay with their pregnant partners, the more care (and less aggression) they showed towards their pups when they were born. It seems that the experience of proximity modulates the male’s biological system to somehow discern paternity certainty and to thus invest in the care of their offspring. 

Evolution essentially conducted the same experiment in some species with the advent of the pair bond. Neural systems for mating were tuned in such a way that following the already very rewarding act of copulation, the particular individual with whom you were copulating also became very rewarding to be around. Other potential partners lost their appeal. Thus, the “pair bond” was formed. And where there are pair bonds, there is greater confidence in paternity. In essence, mammals that form pair bonds typically also have fathers that care for their offspring.

What causes the pair bond to be formed? Consider two species of voles — the prairie vole and the montane vole — that to the average person look like identical  dark grey hamsters. However, they live in different habitats (reflected in their names) that have different pressures, and as a result, they have totally different reproductive strategies. The montane vole is like the typical mammal. Males mate and then take off; females care for offspring exclusively. Prairie voles, on the other hand, pair bond. They stick around; the males are fathers. They keep their offspring warm by huddling with them, they groom them and protect them from predators. The prairie vole strategy has a lot to do with the fact that they live in a much harsher and less food-dense environment than the montane voles. Under these conditions, it's more adaptive to pair bond, have fewer offspring and have both parents care for those offspring.

There are still two puzzles to be addressed. First, what is the neural circuit that turns philandering males into fathers? Is it a novel circuit present only in those few species that show paternal care or a latent one present in most if not all mammalian males, as the mouse experiments suggest? 

The second puzzle is: Whether the circuit is novel or latent, how does it get activated? 

It turns out that it’s easier for evolution to generate similar brains and neural circuits across males and females but activate them differently as a function of context. One key node in the parenting circuit is a part of the hypothalamus — located in the middle of all mammalian brains, including yours — called the “medial preoptic area” (or MPOA). When this region is experimentally lesioned in rodent mothers of any species, parenting behavior is drastically reduced. Similarly, in rodents, like the California mouse, that naturally have fathers, lesioning the father’s MPOA reduces their parental behaviors.⁴

This circuit is very similar in both sexes, but obviously females and males of any species don’t go around caring for any infants they come across at any time. Something has to activate the circuit around the arrival of their offspring. In mothers, we know that the trigger is the huge fluctuation of hormones associated with pregnancy and birth, and that the placenta is a major source of those hormones. Obviously, males don’t generate a placenta after they fertilize an egg, but it seems that a fluctuation of hormones, like an increase in estrogen, prolactin and/or a decrease in testosterone, may also tune the male brain towards parenting behaviors.⁵ For example, Mongolian gerbils and Campbell’s dwarf hamsters (formerly known as Djungarian hamsters) both have fathers that care for their young, and that caring is linked to a drop in testosterone after the birth of their pups. 

There are some species wherein the hormone flux needed to activate parenting in males is caused by being near the female, maybe via some pheromonal signals. Typically, for pheromones to work, the individual receiving the signal must have a specialized organ, called the vomeronasal organ, located in the nasal cavity. Urine, for example, contains pheromonal signals. The amniotic fluid may as well. While there may not be an all-or-nothing “you-should-turn-into-a-father-now” signal in the urine or amniotic fluid, their effect on the male’s vomeronasal organ may push them towards care behaviors. 

Another possible source for care induction are sensory cues from the newborn. If you are a male mammal, and fairly certain that the kid is yours, then maybe your infanticidal tendencies are in check just enough for you to appreciate the smell, vocalizations and/or other signals coming from pups or infants. 

Finally, the placenta itself can be a source of hormones for males. In those California mice, males help induce the birth of pups by licking the anogenital region of the females. After birth, the male eats the placenta and immediately starts helping to care for the pups. When those signals from the placenta are depleted through digestion, California mice males maintain their parental behaviors through exposure to their mate’s urine. This would be a shift to the pheromonal system. It has to be the mate’s urine, no one else’s: One experiment showed that males exhibited more paternal care when exposed to the urine of their mate versus urine from either a virgin female, a lactating female (i.e., another mom) or a female that he was familiar with (but not a mate). The female doesn’t even have to be there.⁶ If fathers were given a dose of his mate’s urine a couple of times a day, then paternal behavior towards their pups was maintained. In Campbell’s dwarf hamster, males even act as midwives, helping with the delivery of pups. When they do, they consume the amniotic fluid and a good chunk of the placenta, both potent sources of hormones.⁷

It seems like fathering should be a fixed behavior. A lot is riding on which reproductive strategy a male picks: finding a bunch of mates or sticking with one and caring for offspring. But the flipside could also be argued: Because it is so important to pick the correct strategy, one should be flexible and able to adapt. This brings us to the African striped mouse.⁸

Males of this species can show care towards their offspring but also towards offspring that are not even their own. It gets weirder: They can show parental care without the mating/pair bond requirement, in contrast to the typical mammalian father. Even stranger: Whether or not males — sexually naive or not — exhibit parental care depends on how they grew up.⁹  

After being weaned, male striped mice can live in very different social environments. They can live in all-male bachelor groups,  never leave their birthplace and family, establish a different territory or just wander the environment alone. It all depends on their population density and what resources are available. The strategy they pick influences their tendency to exhibit parental care. Males that grew up as solitary individuals exhibit parental care (licking and grooming pups, keeping them warm) and those that grew up in bachelor groups show no care for pups and can often be infanticidal. 

These conditions were recreated in the lab. By manipulating housing conditions, the experimenters could investigate the neural circuit for infant care versus infant-directed aggression in males that are sexually naive. These mice had no pair bond with a mate, and no need to activate the care circuit with urine, amniotic fluid or the placenta. What did they find? Sexually naïve striped mice that grew up isolated, and exhibit parental care towards pups that were not their own, had far greater neural activity in the MPOA than group-housed males that were either indifferent to pups or infanticidal. 

How does “lifestyle” (living in isolation versus bachelor groups) contribute to this neural difference? As in all things biological, there are many factors and no singular causes. That said, in this case, one major influence on parental behaviors in the striped mice is the expression of a gene called Agouti. When expressed in the MPOA, it produces a protein that reduces MPOA activity. In fact, its influence on infant care behavior is so strong that researchers can experimentally express more Agouti in males and cause them to switch their parenting style. If they were indifferent towards pups before, they become infanticidal. If they showed care before, they became indifferent. 

In the wild and in the lab, something about social density influences the expression of Agouti which has these (and other) behavioral consequences. These behaviors are not fixed — either you parent or you don’t — for the rest of their lives. If you take a group-housed, non-infant caring male and keep it socially isolated for a month, then it will become more caring towards pups (and its Agouti expression goes down in the MPOA). This shows that the infant care behavior in striped mice males can change as the social environment changes, that neural circuits can be modulated back and forth as needed.

From fish to the striped mice, what the data collectively tell us is that fatherhood is the product of a system. The system includes many things at many levels. It includes gene expression in certain places at certain times, neural circuits, hormones that modulate both gene expression and neural activity, social context and the environment. No component of the system is privileged over another. Fatherhood can emerge from shifts in hormones, changes in social context, the balance of energetic resources — all of which can change neural activity to express a range of behavior from infanticidal to infant care. Now imagine how complicated it’s going to get when it comes to us, a species with a big brain whose culture and biology evolve together.

Human societies vary in the expression of fatherhood, from complete absence or aloofness to great intimacy and direct care. 

As an example of one type of paternal care that may be wholly unfamiliar, consider the Aché, an indigenous people who traditionally inhabit the tropical forests of Paraguay. Among the Aché, both men and women may have multiple sexual partners. Aché women have autonomy in their choice of mates and how long a relationship endures. The Aché believe that any man who has had sex with a woman during a period starting several months before she was discovered to be pregnant, and lasting up until the day of the child’s birth, contributes to the paternity of the child. There are primary fathers, who were in long-term mating relationships with the mother, and secondary fathers. Children with primary and secondary fathers are more likely to survive than those that did not.

What do these fathers do? They may protect them from homicidal males and provide additional food or other resources. Beyond that, Aché fathers don’t spend much time holding, playing or carrying their kids as we in Western cultures typically expect “good” fathers to do. This whole arrangement goes against what we typically think of fatherhood, that the amount of paternal care should correlate with the certainty that the father knows that the kid is theirs.¹⁰

Just as in other animals, differences in human paternal care can be related to the environment. Household composition, availability of mating opportunities and resources, mortality risks, and pathogen prevalence are all potential factors that influence paternal care in humans. Even within a single society, there exists a lot of variation in paternal care. This could be differences within a generation as well as across generations. The way I interact with my own kids, for example, is very different from the way my father interacted with me and my siblings. 

Because our behaviors are part of an even more complex system than fish or other animals, understanding the biological mechanisms of fatherhood in primates is much more difficult to decipher than it is in rodents. Nevertheless, we are learning that at least some of the mechanisms that work in rodent fathers may also work in primates — including us. Though it is not the only factor, of course, and though causality is hard to establish, changing testosterone levels seems to be important for fatherhood in primates, too.

Marmoset monkeys are one of the few primate species that are cooperative breeders like us — both parents, older siblings, and even unrelated group members help take care of infants. By taking blood samples at strategic points in time, marmoset monkeys provided early evidence that, like some rodent species, testosterone levels are linked to offspring care. Marmoset fathers who carried infants the most had the lowest testosterone levels. Thus, testosterone seemed to be antagonistic to infant care. Its levels are high during mating, so its drop in these caring males suggest that there is a tradeoff.

When mating, you need testosterone. Post-mating, it seems testosterone tends to go away in association with caring for infants.  For sure, there are other hormones involved in fatherhood just as there are in motherhood — prolactin, estrogen, progesterone, cortisol — but testosterone is what we know most about.¹¹ This is true even for humans, a species in which controlled biological experiments have significantly steeper limits, comparatively speaking — experiments that would potentially manipulate one's relationships with adults and children are, obviously, not permissible.   

Here’s what we do know: The ascent to fatherhood in an individual man is linked to his decline in testosterone. Starting in the 1990s, a set of studies measured hormones in blood samples or saliva in adult male humans.¹² They showed that married men had lower testosterone than single men, and that divorce increases testosterone. These studies linked testosterone to the status of a man’s pair bond. Other studies showed that married (i.e., pair-bonded) fathers had lower testosterone than married non-fathers, linking testosterone to parenting. Most of these studies, however, were cross-sectional, comparing one group of individuals (e.g., married) with another (e.g., unmarried).  Thus, they cannot show a direct relationship between pair bond or parenthood and levels of testosterone. It could be that already having lower versus higher testosterone just increases the likelihood of getting married and having kids. One 10-year longitudinal study showed that being married was associated with lower testosterone levels and getting divorced was associated with higher testosterone. This study thus linked testosterone levels to the status of the pair-bond.¹³

In 2011, another longitudinal study came out, directly testing the hypothesis that levels of testosterone reflect a trade-off between mating effort and offspring care in human males.¹⁴ The study followed a cohort of hundreds of adult males in Cebu City in the Philippines, measuring salivary testosterone levels as a function of their relationship status, whether or not they had kids, and how much time they spent directly caring for those kids. 

Single men with the highest testosterone levels were later more likely to be in a relationship and more likely to have fathered a child. This supports the idea that higher testosterone is beneficial to mating effort. Then, being newly partnered decreased testosterone levels, as did having a child. This decline in testosterone could not be attributed to increased age or stress levels. The largest decrease in testosterone was observed in men with newborns versus men with infants or toddlers. Finally, the fathers who spent the most time caring for their children had the lowest testosterone of the studied cohorts. A follow-up study showed that, as fathers changed the amount of time they spend with their kids, their testosterone levels adjust accordingly — more time, less testosterone.¹⁵ Thus, when mating is complete, a decline in testosterone facilitates fatherhood. All of this is consistent with what we’ve learned from rodents and monkeys. 

Where things get complicated, however, is when we incorporate culture. The study of Filipino men involved a culture where male care of children is common. Across the world, however, there is considerable variation in who becomes a father and how much care they provide. There is probably more variation within the human species than across all other primate species. This is because, with increasingly large brains, primates (including humans) can adopt different types of social organizations (which are often influenced by available resources). These social organizations change the mating game and therefore the existence, and nature of, male care. And of course in humans, where we exhibit many different types of cultures, and where it is extremely difficult for a mother to care for offspring alone, the variation in male care is high.

When a human male does become a father, what else actually correlates with that hormonal shift, that decline in testosterone (among other hormonal changes) and what brain areas are activated? Human males don’t typically eat the placenta or sample the urine or amniotic fluid expelled from mothers. And humans can’t detect pheromones like other animals (we don’t have a vomeronasal organ). So, it could be olfactory, auditory, visual and/or tactile cues coming from the mate (pregnant or not) and/or from the baby.

Cause and effect are hard to disentangle. This is true even when we can use animal models. For example, male rats are infanticidal, but if you castrate them around puberty, then they lose that killer instinct (so to speak).¹⁶ However, they don’t at the same time become caregivers. Like any system, multiple factors are always at play, and their interactions change over time.

In humans, some studies show that hearing or seeing a baby cry causes a drop in testosterone, particularly if the next action was to soothe the baby.¹⁷ Understanding the neural circuitry underlying human fatherhood is complicated by the fact that the MPOA, a brain region implicated in parental care in most, if not all, vertebrate species, is actually quite small. It’s accessible for analysis if you can kill the animal and dissect its brain. With humans, we are limited to MRI studies which cannot yet easily or reliably resolve activity in the MPOA in fathers versus non-fathers. But we can image the hypothalamus — the bigger structure which contains the MPOA — as well as the larger network that is connected to it and that may be activated during the care of offspring. 

These studies show that mothers and fathers share a similar pattern of activation of a caregiving brain network. One study compared the response of this network to videos of the subjects interacting with their infants versus the subjects alone versus a stranger interacting with the stranger’s own infant.¹⁸ The network was most activated by the first condition.

In a rare break from the heteronormativity typical in parental care research,¹⁹ this study also showed that homosexual fathers (one biologically related to the offspring, the other an adoptive father) also engaged this caregiving brain network. And for both heterosexual and homosexual fathers, the strength of connections within this network increased as a function of how much time they spent caring for their child alone. To put a ribbon on it, in heterosexual men at least, there is a relationship between this caregiving brain network and testosterone. A study measured testes volume and circulating testosterone in men as a function of infant care and activation of a “reward” area in the brain when viewing pictures of their kid.²⁰ They found an inverse relationship: the bigger the testes, the less fathering behavior and the less activation of one of the brain’s reward areas.²¹

Experience seems to play an outsized role in the expression of human fatherhood, both through culture and the vagaries of individual lives. My brother and I are similar in many ways but differ in the trajectory of our reproductive lives. While I would certainly not describe him as a philanderer, he does not have kids. I, on the other hand, have been with my partner for over 25 years and have two daughters with her.

I’m not sure what accounts for our differences. We grew up in the same small town in Idaho. We had the same parents, but not the same kind of relationships with them. It could be that the initial conditions of our early lives were different enough to modulate our neural circuits and hormones differently. That might account for why I’m bald and he’s not (he also claims to be slightly taller). It could also be that an early experience I had as an adolescent really made me want to have children. When I was 12 and he was eight, we spent the summer with our grandmothers and extended family just outside of Lahore, Pakistan. There was no TV or other entertainment, but there was my one-year-old cousin. He was so cute and interesting; I fell in love with this kid. Maybe all those infant cues did a number on my adolescent circuits.