Let's think first about what mouse and human fathers actually do. Human fathers have an enormous range of behaviors toward their children, depending on their culture and environmental demands. Some fathers go away from the home, hunting or working for many months at a time; others are in hourly contact with their children, feeding, cleaning, soothing, carrying, and teaching them. Most human fathers are kind and nurturing toward their children, while a small number attack and injure or kill the offspring. At least in captivity, mouse fathers do jobs similar to those of mouse mothers, and are especially likely to retrieve pups that climb out of the nest and have to be fetched back before they get chilled or are caught by predators. Like the mothers, mouse fathers may also act to chase away intruders from the nest. But mouse fathers and mothers alike may kill and eat pups, and the likelihood that they will do this is determined in part by genetic differences. Different species of non-mouse rodents also behave differently in terms of infanticidal tendencies.

Human and mouse parents have different jobs to do because human and mouse young are born at different levels of development and need to be cared for at different levels for different lengths of time. Mouse pups can make their own way to the mother's nipples and latch on to nurse, as soon as they are born, whereas human babies need a good deal of assistance on these tasks for quite a while. Human parents continue to provide care and protection right up to the offspring's puberty and often into adulthood, but mice do not.

The lifetime "bond" mentioned in the ScienceNOW paragraph often exists for humans, and is a mutual attachment. However, what the ScienceNOW author interpreted as a "bond" in mice is the fact that the fathers who had spent time with their pups recognized those pups when they were adults and did not behave as if they were strangers by sniffing and examining them. Above all, the experienced fathers generally did not attack their now-adult pups. The extent to which these fathers "protected" their young was that they, the fathers, did not kill them-- that is, the young were protected FROM the fathers by the fathers' recognition and related behavior change. While this was an important adaptive change in the fathers' behavior toward the young, it is a far cry from protecting the mature young from intruders, as the fathers would have done when the young were pups. We would probably not describe a powerful bond between a human father and child as being demonstrated just by the father's failure to murder the child.

Lifetime-long relationships between parents and children are common among human beings, although those relationships change gradually until adult offspring often nurture and protect their elderly parents. The mouse research in fact did not show that the fathers' recognition would continue through the roughly two years of a lab mouse's normal life-- nor did it test whether the offspring were wise enough to know their own fathers.

I have no wish to downplay Mak and Weiss's interesting work, which focused on the interacting roles of genetics, hormones, and experience in determining paternal behavior. I just want to apply the brakes to a discussion (not by those original authors) that rather precipitately jumps from mouse fathers to proposals about human fathers. Please, let's not have any one claim that human fathers must nuzzle their offspring, or they won't recognize or have a "bond" with them. Human fathers don't need to do other mouse things, either, like licking the young or keeping them from getting out of a nest.

Human parental behavior and emotions are much more complex than those of mice, and we can't use one as a direct basis for understanding the other. Keep in mind how soldier-fathers may long to be with children they have never seen. There is far more to human parenting than hormone levels, important and interesting as those may be.