On the Wall Street Journal’s “Total Return” blog, Jason  Zweig highlights research that indicates human beings have an “optimism bias” — that is, we learn more from our successes than we do from our failures. And that, he says, has major implications for investors.

Zweig says a new study, performed by a team of neuroscientists in London and Berlin found that “in short, humans don’t learn equally well from upside and downside mistakes.” Instead, “we pay more attention when the future turns out to be better than we expected.”

“If you bought Apple at $60 a share thinking maybe it would double, you’ve probably spent a fair amount of time wondering why you underestimated its potential and trying to apply those lessons to find other great stocks,” Zweig explains. “On the other hand, if you bought Netflix at $200 a share, never dreaming it would go down by more than half, you’re probably not doing much self-reflection at all; you’re looking for somebody to blame.”

Another example of how this plays out, Zweig says, involves selling decisions. Most investors, he says, stop following a stock after they’ve sold it. But, he says, unless you continue to follow it — and compare the performance to the performance of  any stocks you replace it  with — you don’t really know whether the sell decision was a good one.

The bottom line, according to Zweig, is that investors have to “force themselves to study their mistakes, or they will never learn from them.  Otherwise your automatically optimistic brain will keep you from confronting the truth.”