Interesting quote on The Edge about simulation and happiness.
Daniel Gilbert says:
"For the last decade I've been obsessed with one problem: How well can the human brain predict the sources of its own future satisfaction? If the answer were "Very well, thank you," then I'd be out of a job. Research suggests that I will be employed for a long time to come.
We are often quite poor at predicting what will make us happy in the future for two reasons. First, we have been given a lot of disinformation about happiness by two sources: Genes and culture. Both genes and cultures are self-perpetuating entities that need us to do things for them so that they can survive. Because we are interested in our own happiness and not theirs, both entities fool us into believing that's what is good for them is also good for us. We believe that having children will make us happy, that consuming goods and services will make us happy. But the data show that money has minor and rapidly diminishing effects on happiness, and that parents are generally happier watching TV or doing housework than interacting with their children.
So what happens if we try to disregard the genetic and cultural imperatives and just figure it all out for ourselves? What happens if we just close our eyes, imagine different possible futures, and try to decide which one would make us happiest?
My research with Tim Wilson shows that when people try to simulate future events — and to simulate their emotional reactions to those events — they make systematic errors. Modern people take the ability to imagine the future for granted, but it turns out that this is one of our species' most recently acquired abilities — no more than three million years old. The part of our brain that enables us to simulate the future is one of nature's newest inventions, so it isn't surprising that when we try to use this new ability to imagine our futures, we make some rookie errors. The main error, of course, is that we vastly overestimate the hedonic consequences of any event. Neither positive nor negative events hit us as hard or for as long as we anticipate. This "impact bias" has proved quite robust in both field and laboratory settings.
Oddly, we don't seem to learn all that much from our own experience. To learn from experience requires that we be able to remember it, and research shows that people are about as bad at remembering their past emotions as they are predicting their future emotions. That's why we make the same errors again and again. For example, in one of our studies, Democrats predicted they'd be devastated if Bush won the 2004 presidential election, and as we always find, they were not nearly as devastated as they predicted. But several months after the election, they remembered being just as devastated as they had expected to be. It turns out that this is a very common pattern of memory errors. Retrospection and prospection share many of the same biases and hence reinforce each other."