There is no shortage of comment on JetBlue flight attendant Steve Smith's "emotional meltdown", including Andrew Bernstein's PT blog (http://www.psychologytoday.com/blog/the-myth-stress). What interests me is the mostly positive reaction to his escapade, despite his cavalier attitude to occupational safety. (At this point the Yahoo news headline features the phrase "folk hero"). It contrasts with negative reactions - at least on the part of some of the stuffier print media commentators (e.g. http://www.guardian.co.uk/commentisfree/andrewbrown/2010/aug/10/r... ) 
- towards another unusual emotional display. I am referring to the Nao robot developed by Lola Cañamero and her team at the University of Hertfordshire. It is said to mimic the emotional skills of a one-year-old child. Simulations of emotional expression aren't new, but what is interesting about Nao is that it learns when to display emotions through interactions with humans. It's cool stuff, in my opinion, but the idea of a machine faking emotion is disturbing to some.
It is easy to think of numerous further examples of how people are conflicted about emotional displays. I think there may be multiple levels of ‘emotional intelligence' at play here. The base level is recognition of the source of emotion. You would have to be distinctively emotionally challenged to fail to empathize with the flight attendant's frustration with willfully obstreperous passengers, or to believe the robot was experiencing some real emotion. (Although we do seem to be wired to attribute emotion to artifacts such as cuddly toys.).
Beyond such essentially rational reactions, another level is the appreciation of authenticity. Smith's anger was heartfelt, unvarnished, and unmotivated by personal gain. Manipulative use of emotional displays is common enough to make such spontaneity refreshing. By contrast, the robot Nao seems phony, even creepy. (You may also recall the Sega "Japanese robot girlfriend" which amps up the creepiness by an order of magnitude). Movies such as Blade Runner have explored an essential unwholesomeness in robots that simulate humans.
But, there is one more factor here, which is the capacity to appreciate emotional displays in context, with a degree of detachment. I have a sneaking admiration for Steve Smith's exit from the JetBlue plane - the two beers were a nice touch - but I don't think I would want to fly with him. Likewise, the point of the robot is not to simulate human emotion but to facilitate human-robot interaction. As the University of Hertfordshire group have explained, robots that act friendly may be reassuring to patients in hospitals. You can be perfectly aware that the robot is just a mechanism while still enjoying the interacting with it. So, while I am normally in favor of emotional authenticity, there are times when inauthentic emotion has its place. This goes for human-human interactions also. I actually prefer civility rather than surliness in a customer service agent, even though both parties know that it is just an act of emotional labor. True emotional dishonesty is another matter.
A final thought is that there are personality factors at work in these meta-emotions (i.e., emotions about the emotions of others). Research on individual differences in cognitive style, has uncovered a number of constructs that have to do with preferences for certainty over uncertainty and ambiguity in information (e.g., intolerance of ambiguity, need for closure, need for cognition). Perhaps we need something analogous for emotion and personality, contrasting the person who prefers straightforward, direct emotional displays with the individual comfortable with multiple layers of emotional expression and ambiguity. Or perhaps we just need robot flight attendants.
