The purpose of the faces/trucks data point is just to illustrate that these differences start from birth. The larger argument is that the people-vs-things gap, and later, interest in computers specifically, basically remains stable throughout childhood, adolescence, and early adulthood:
https://slatestarcodex.com/2017/08/07/contra-grant-on-exagge...
I don't even know if the data is ready for us to draw any conclusions, but at the very least it seems to me we should find this to be an extremely interesting pattern, and use it to motivate hypotheses.
The core activity of software engineering, programming, writing code, is a solitary activity. It's what people think of when they think of programming. It's what we spend hours being trained for, getting good at. Yes, in between writing software, we need to coordinate with others, so we have meetings. You'd really describe this overall process as being more people-oriented than thing-oriented?
I've heard the argument that somehow "actual coding is a relatively small part of being a software engineer," but unless you're a manager (of which there are many more women), the thing you're being trained for, the thing you spend most of your time doing, and the basis of how people perceive the profession, is sitting in front of a computer coding. You can describe any profession as people-oriented on the basis that one needs to work with others, but the key question is whether the basic activity of the job is a social one.
Regardless of whether the people-vs-thing distinction is significant or not, it seems inaccurate in a big-picture way to describe programming as people-oriented. Like, that's not what people mean when they draw that distinction.