Far apart from the brain structure involved, what amazed me the most is the severability of fear from other emotions. For someone untrained in neuroscience and coming purely from an experiential point of view, it seems impossible that one emotion could be completely absent without having any effects on the functioning of others. For instance, even with an emotion like happiness, it seems like there's some component of the feeling that relates to the enjoyment of being free of fearful things. Perhaps the emotions don't actually function the same, but her perception of them is similar to unaffected patients?
Many of the other emotions are restricted to apes, specially humans. I'd say a dog can't be happy or sad, just excited or not excited. If those human emotions evolved thousands of years apart there is a good chance that they are at different parts of the brain.
>After this point, neither the concrete definition as to the extent of the amygdala is not clear, nor is the exact function of each of its subgroups. In the amygdala region alone, there is much controversy surrounding the nuclear subgroups, resulting in classifications that range between 5 and 22 different groups within the amygdala itself.
"The Role of the Amygdala in Fear and Panic". http://serendip.brynmawr.edu/exchange/node/1749