A while back there was some ABC news show about how unreliable eyewitnesses can be, particularly when they are involved with another activity.
They showed a video tape of three people tossing around different colored balls. They ask you to count how many times a certain colored ball was tossed. At the end of the test they didn't ask so much about the number of tosses, but about the gorilla. What gorilla? I thought it was a trick, but a replay showed that after a few seconds of ball tossing a guy in a gorilla suit came out and stayed on camera for several seconds.
I was so intent on my count that I never saw it. I was wondering about my lack of observation when they mentioned that over half who view the tape with those instructions don't see the gorilla.
Inaccurate perception plus inaccurate recall can make a number of people poor eyewitnesses, so who to believe?
ABB