sequences as IQ tests

[JEREMY GARDINER]

Perhaps this posting by David Hamill on the Yahoo Robitron group may be of passing interest to some seqfans:
  ------------------------------------------------------------
Intelligence also includes an ability to work around problems instead of running headlong into them [...] This is the kind of intelligence that's not tested by IQ tests.
  With tests the boundaries are firmly set and there's almost no scope for creative thinking. What's more, with multiple-choice questions you don't get a chance to explain your thinking:
   
  Q: What is the next number in this sequence? 1, 2, 3, 5, 8, ...
  A: 9.
  Explanation: The numbers increase in size from left to right and are all single digits; the only single-digit number that's greater than 8 is 9, so this must be the answer.
   
  or, A: 1.
  The digit 1 is formed with one stroke of the pen, 2 with two, 3 with two strokes, 5 with three strokes, and 8 with two, so we have: 1, 2, 3, 3, 2. By symmetry, the next number of strokes is 1. There's only one digit formed from one stroke, 1, so this must be the answer.
   
  Of course the "correct" answer is 13, so highly creative answers like this are penalised as if they're random guesses.
------------------------------------------------------------
  Indeed, a search for 1,2,3,5,8  in the OEIS returns 278 sequences.
   
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://list.seqfan.eu/pipermail/seqfan/attachments/20060210/f0b3091b/attachment-0001.htm>