You are probably right, but it is not clear which category this finding of a fungus-Alzheimer's correlation would be in. Most theories could retroactively be p

You are probably right, but it is not clear which category this finding of a fungus-Alzheimer's correlation would be in. Most theories could retroactively be put in to either category (new facts or changed facts), so I am not sure whether this framework gets you anywhere. In any case, the relative difficulty of updating previous beliefs does not excuse the failures of doctors which cause a great deal of death and suffering. When engineers fail to update their beliefs, and people die, it is called negligence.