We think a broader version of the hypothesis - that people avoid scaling down outrage from relevant reference points - is a big part of it. We are currently working on follow-ups that explore a preference for escalation from *others’* judgments and finding evidence for that prediction!