First, most of our hardware is not designed to tight specifications when we compare ourselves with one another. Many of us approximate average values for most of our equipment; most of us, however, are significantly "off" on one trait or another. This variability is perhaps "noise," it also has the important property of generating individual variability that encompasses environmental variability. In other words, if there's a change in survival conditions, some of us well be pre-wired to adjust more quickly to the changes.
Bonding and attachment -- Oxytocin, the same hormone sometimes used to induce labor, is strongly implicated in our bonding circuits. So are certain combinations of odors. So is early experience with a particular type of parent (geese) or particular odors (lemon odor and rat pups) as a factor in later mate selection. Our attachment systems are very "old" and shared with some very surprising creatures.
Highly reactive rhesus (and I think kids and adult males and some adult females) have particularly strong needs for an "anchor" in order to be daring and commanding. Stephen Suomi comments that highly reactive infant rhesus become troop leaders if given unusual degrees of nurturance but are likely to be injured during environmental distress without the extra mothering. Highly reactive females become poor mothers unless given help with the first kid or two by an "aunt" female.
Thus, we vary in our amount of "glue" and in our ability to be detached. Useful for the group, can be painful or puzzling for individuals.