I once heard of an experiment in economics that offers insight into this.

Say you have 100 people. You give each of them one of two choices:

A : you get $40 unconditionally B: you get $70 - n, where n is the number of people who choose B

You end up getting, on average across experiments, n = 30.

If you move the numbers around (i.e, the $40 and the $70), you keep getting, on average, a number of people choosing B so that B pays out the same as A.

I think the interpretation is that people can be categorized by the amount of risk they’re willing to take. If you make B less risky, you’ll get a new category of people. If you make it more risky, you’ll lose categories.

Applied to traffic, opening up a new lane brings in new categories of people who are willing to risk the traffic.

Or something. Sorry I don’t remember it better and am too lazy to look it up. Pretty pretty cool though.