I don't think correlation/causation confusions apply here. It's a question of whether these evolutionary explanations for actual observed behaviours are falsifiable. What is correlated but not causated (is that even a word?) in this situation?