As you mention, it may be quite difficult to replicate Broadcast/Seed and Common Knowledge/No Common Knowledge ideal experimental conditions in a policy application setting. I imagine a particular challenge is that Seed to Broadcast and NCK to CK may be one way streets. I.e. once we have common knowledge, it is hard to make it "uncommon" and, likewise, once information is known by many (Broadcast), it is hard to make it Seed only. This may make it difficult to get to the more optimal Broadcast + NCK and Seed + CK quadrants. Given such challenges, what should be the guiding strategy for a government trying to disseminate reliable information well?
As you mention, it may be quite difficult to replicate Broadcast/Seed and Common Knowledge/No Common Knowledge ideal experimental conditions in a policy application setting. I imagine a particular challenge is that Seed to Broadcast and NCK to CK may be one way streets. I.e. once we have common knowledge, it is hard to make it "uncommon" and, likewise, once information is known by many (Broadcast), it is hard to make it Seed only. This may make it difficult to get to the more optimal Broadcast + NCK and Seed + CK quadrants. Given such challenges, what should be the guiding strategy for a government trying to disseminate reliable information well?