The first article uses research from Ariely, who is famously not believable.

Three years ago: https://www.newyorker.com/magazine/2023/10/09/they-studied-d...

Today: https://datacolada.org/139