Oh dear! That's a tall order, so I shall be brief. Sunny and I did most of this work in the summer of 08 when she was 14 and I was Old. Although I am most familiar with neuropsych test protocols, we both realized that there are obvious interspecies discrepanceies that are not easily reconciled and, at Sunny's suggestion (she had just read
The Hitchhiker's Guide), we used Brett Watson's "Magic Monkey Formula":
http://www.nutters.org/docs/more-monkeys (note that you will save time as you work the formulae if you attribute the same value to e and the natural log, n)That was the easy part. To keep our standard error within manageable limits, we used 100,000 monkeys, goldfish, mantids and mantis flies (building a typewriter suitable for even one mantis fly presents significant problems, and over half were left clawed), and the experiment was carried out in a time capsule, so that we could complete100,000 years of exps before Sunny had to return to school. The results were remarkably consistent with the raw IQ scores, once we corrected for some practical problems, like the three monkeys that ran away to join the circus and 27 goldfish that were accidentally eaten when some of the mantids got loose.
At the end of testing:
27 monkeys produced a copy of Shakespeares's
Hamlet 13 in the 2nd folio version and 14 in the Furness variorum edition.
5 guppies produced
Hamlet in a variety of editions (variorum editions were graded only on the play, not the citations).
0 mantids produced
Hamlet, but three produced
Two Gentlemen of Verona (and when did you last see a professional production of that?)
0 mantis flies produced
Hamlet, but three produced
Fun with ###### and Jane, with a few minor typos.
Obviously, with such a small n, the experiment requires duplication to establish its validity, but I trust that these results will inspire you to try some related experiments of your own. If you do, I can put you in touch with an inexpensive source for --just a moment -- 999,270 monkeys.
