Assuming your summary of the study is essentially accurate and complete,
then I'd say that the investigators probably erred in terms of the scope
of the experiment. My guess is, if they had also set up an experiment
that had the babies interacting with another human (or a human analogue),
they would have found just the opposite.
From what I've been able to read and observe, males seem to have a
congenital predisposition towards interacting with, and subsequently
learning the mechanics of, their immediate environment, with the (assumed)
goal of manipulating it to their advantage. Females, on the other hand,
seem to have a predisposition towards interacting with, and subsequently
learning the motivations of, their immediate peer group, with the (assumed)
goal of manipulating it to their advantage. As with all aspects of
human nature however, these descriptions merely represent the median of
a rather broad Gaussian distribution (bell-shaped curve), and therefore
have no descriptive or predictive power on an individual level, especially
when the amount of overlap of the curves is factored. That's why these
differences in traits only manifest themselves clearly in statistically
relevant group sizes.
Seen from this perspective, it may be argued that these dualistic
and complimentary (sound familiar?) behavioral predispositions and
cognitive abilities could, as a result of the dynamic equilibrium in the
natural tension resulting from their interaction (sound familiar?) in a
mated-pair, manifest an emergent property that consists of a unified set
of behaviors and abilities that, in prehistorical survival terms, is
greater than the sum of its parts (sound familiar?).
But we don't live in prehistory. We don't live in bands of a couple of
dozen hunter-gatherers, where the "natural world" is the environment we
primarily interact with. We live in highly interconnected groups of millions
or billions, and that in itself constitutes a large part of our environment
now. In this setting, those somewhat polarized traits manifest themselves at
the cultural level as the "battle of the sexes". Shit, I'm rambling again.
Anyway, in answer to your question, I would say that the females actually
did care less. Their programming is (on average) different than that of
the males, and results in making them perceive areas of interest that are
different from the males. It is in their own particular areas of perceived
import that the "stress/reward" dynamic would manifest itself. But like I
said, this is just a guess.
Dan