I'm not entirely sold on the methodological soundness. It seems like they went into it looking to confirm something, and designed and experiment that probably would. Maybe it's more rigorous than I got from reading the first part of the article, it doesn't come across that way.

I do think the conclusion makes sense, I'm just not sure about how well this demonstrates it.