All my life, my brother and I accused our parents of taking each other's side. "You always side with him. Of course you do — you like him more." It was our final argument, that basic emotional blackmail, every time we lost a dispute mediated by our parents. The answer was always the same: "There is no favorite child. That's impossible."

The same script plays out with my own children. The oldest says I spoil the youngest; the youngest says I compare him to the oldest; and the middle one, squeezed between the two, says he is relegated to his own "irrelevance" (his word).

Any mother or father will insist that the only possible answer is: you cannot love one child more than another.

But science says it isn't quite that simple.

A study published last year by the American Psychological Association concluded that certain characteristics can lead to more favorable treatment of one child. Gender, birth order and temperament can all influence this favoritism.

Faced with the evidence, and as someone who champions science, I have no choice: I confess that I do have a favorite child. I just can't say which one.

I have my favorite for talking about soccer. My favorite for walking around the Liberdade neighborhood. My favorite for riding bikes. One is the best for conversation; another, for a particular kind of silence. With one I prefer to talk about manga; with another, about Machado de Assis. They all make me laugh, but some days my mood is more in tune with one of them than the others. It doesn't depend only on who they are — it depends even more on how I am.

There are days, in fact, when my deepest preference is for all of them to sleep over at a friend's house.

A smile, a joke, a gesture of affection, a generous act, an unexpected beauty can send one of them shooting up the rankings in seconds — the same one who, the day before, was in a relegation zone for treating his brother badly.

Is there any instrument that measures affection? A lie detector? I cannot imagine how the repetition of certain patterns could determine the intensity of something that exists only on the inside. Love is a more mysterious variable.

I may prefer one child's attitude without preferring that child. I may identify more with one child's temperament and worry more about another precisely because we are so different.

Children don't ask the same things of their parents either. One needs to be held when another wants distance. One needs boundaries; another needs autonomy. One likes to share everything; the other is reserved, and I make an effort not to intrude — that is how I love him best. Treating them equally would mean treating them with less truth.

There is a physics of motherhood that escapes any science. The feeling expands without taking up more space; the matter grows without taking matter from anywhere else. The birth of a second child does not divide by two what we felt for the first.

Biology, curiously, offers a better image than any theory. After birth, some of the mother's cells remain in the child. And for many years, some of the child's cells can continue living in the mother's body. The scientific name is not as beautiful as the idea itself: microchimerism.

Rather than trying to explain to my children that none of them is my favorite, I try to make them understand that I literally carry pieces of each of them inside me.

Now I just need science to tell me: which cell should I prefer?