According to a paper by Hsee and Zhang, the answer is: not well.
Traditional economists and decision theorists assume that people know their preferences and that what they choose reveals what is best for them, given the information they have at the time of choice. In reality, this is not the case. In a series of seminal articles, Kahneman and his coauthors have argued that what people predict will make them happy (i.e., predicted utility) and what people choose (i.e., decision utility) can be systematically different from what actually makes them happy (i.e., experienced utility). In other words, people may mispredict and mischoose.
Why can’t people predict their own happiness? One reason is that people often neglect to take into account their own ability to adapt or subsequently rationalize prior behavior. Many people also overpredict the importance of external rewards such as income and status and underestimate the importance of activities with intrinsic values such as hobbies and socializing with friends.
For instance, how much value would you gain from a change in your income?
“…most people would find gaining money good and losing money bad, but they would be relatively insensitive to the size of the gain and the size of the loss. Thus, the resulting evaluation function…will be close to a step function: steep around zero or a reference point and flat elsewhere.”
- Christopher K. Hsee and Jiao Zhang. Distinction Bias: Misprediction and Mischoice Due to Joint Evaluation. Journal of Personality and Social Psychology 2004, Vol. 86, No. 5, 680–695.