File this one under "Do We Really Spend Money Studying This?": A study from Masaki Yukia, William W. Maddux and Takahiko Masuda (from Hokkaido University, Northwestern University and the University of Alberta, respectively) sought to discover whether facial cues are read differently in different cultures. The most awesome part of the study was figuring out if emoticons, the sideways faces we type out in combinations of letters and punctuation marks, are interpreted the same globally...
Their emoticon study hinged on their theory that the eyes are more difficult to control than the mouth and that "individuals in cultures where emotional subduction is the norm (such as Japan) would focus more strongly on the eyes ... when interpreting others' emotions" and "people in cultures where overt emotional expression is the norm (such as the US) would tend to interpret emotions based on the position of the mouth."
They looked to previous research (really, who does this?) showing that the Japanese tend to use emoticons with expressive eyes and a neutral mouth (like ^_^ this one) while Americans did the opposite (like our good ol' standby :) happy face).
They then tested the response of students from both cultures to various emoticons and found that Americans didn't know how to interpret emoticon eyes and often badly misinterpreted the meaning assigned to popular emoticons from japanese culture.
This bit from Arstechnica pretty much sums the whole thing up:
Ironically, emoticons are generally used to attempt to convey emotions that may not be easy to express in pure text—these data suggest that they may not always be effective for their intended task. As with many of these situations, knowing your audience is probably more important than anything else.