Proof that talking about one thing can lead to more interesting work