This ends up being a really interesting question. For the record, I am an adult who took child IQ tests (most likely, just two separate administrations of the same Stanford-Binet L-M IQ test) by teacher referral at elementary school age. Much later, I had occasion, for online discussion of education policy, to begin reading about the enterprise of IQ testing in general. I eventually discovered some of the practitioner handbooks about IQ testing (note: NOT the actual test manuals) used then for current adult IQ testing.
Then, still later, I had occasion to take an adult IQ test. (The current test in that year was, if I remember correctly, the WAIS-R, but it might have been the WAIS-III.) As the test session began, I was asked by the test-giver, "Sometimes people have taken IQ tests before, or they've read about the tests so that they are familiar with some of the test items. Please let me know if any of the items we work on seem familiar to you, okay?" I told her up front that I did some reading about IQ testing and it might be possible that I'd find some items familiar. (The practitioners' guides to IQ testing are quite interesting in what they attempt to keep secret about test item content even as they introduce concepts related to correct administration and interpretation of tests.)
As I took the test, which I surely had never taken before, I did indeed encounter some items that were unfamiliar to me, and yet easy to figure out because of things I had read. I let the test-giver know as that happened during the testing. The test-giver, on her part, made sure to give me additional items from other, much more obscure tests. (It was years later before I figured out what test published before my dad was born was used to fill in for the Wechsler items in testing my vocabulary. That subtest is really cool.) If I remember correctly, the IQ test and accompanying memory test (the Wechsler memory test is actually designed to be given during an IQ test administration, so that the delay in recall is enforced by being busy with something else) demonstrated that I didn't have a memory issue to worry about--which had prompted the testing. The test-giver noted that there were some real difficulties in determining a valid WAIS IQ score for me, because of my recreational reading, but ventured a characterization of my general level of verbal ability based on the supplemental vocabulary items I was administered. Any responsible test-giver MUST note when a test-taker's performance is far outside the norm. To do otherwise, and especially when there is any doubt about standardized conditions of not having access to the test item content in advance, is a violation of professional ethics.
http://www.nytimes.com/2002/03/02/u...e-faked-tests.html?pagewanted=all&src=pm 
