The facial feedback hypothesis: Methodological concerns and new supporting evidence