The model suggests, if I read it correctly, that media would portray Contrarians (who dislike crime but don’t think jail is the solution) as Troglodytes (who op

The model suggests, if I read it correctly, that media would portray Contrarians (who dislike crime but don’t think jail is the solution) as Troglodytes (who oppose jail because they favor crime.) The “Dupe” label would then apply to those who accept the jail policy because they dislike crime. Perhaps the model tries to disparage people who make this leap, without themselves considering if the policy is a good way to combat crime. I guess I see your point; there might be people who dislike crime and have done their homework on the policy, and the model unfairly labels them as dupes. Perhaps the scope of the model is implicitly limited to reason about those who blindly accept the media’s points. The article seems to focus elsewhere: on the case where media pushes the top left quadrant toward the diagonal. Equipped with Hanlon’s razor, one may alternatively posit a dimensionality reduction which, albeit lossy, improves the performance of media.