Hi:

Very interesting story...and what an ending too.

I've got one about a pigeon.

There was a trainable pigeon in a cage. The cage had viewing slits around its sides so that the observers could glipse the action of the pigeon and the cage's "pigeon-training" mechanism.

The pigeon was trained to pull a lever and receive a treat if he correctly interpreted the stimulus he received (through an internal mechanism in the cage, which was a changing color coded bar).

The team of observers were tasked to understand the logic the pigeon used to correctly interpret the stimulus and receive the treat. Of course they all viewed the pigeon through the same viewing slit.

The observers saw that the pigeon was rewarded with a treat an equal number of times when the color seen was "Red" as well as when it was "Green". They therefore concluded that the process was random with no way to tell what allowed the pigeon to win the treat.

What the observers did not know was that there were actually 4 colors and not 2. They could only see the "don't care" randomly distributed (it seems) red and green colors but they could not see the actual ones the pigeon was trained to respond to.

Interestingly enough Had They sent one team member around to another viewing port the training rule would have been obvious....

In addition, a mirror showing the active mechanism at work was also visible to the team but they failed to detect the clue....

After completion of the experiment, they decided to have the pigeon for dinner ...since there was nothing more to be learned....

:)