Quite Interesting IQ test

Why have I bothered to reproduce this test? - This IQ test is fascinating to me because there is no time limit. Kevin Langdon created a beautiful set of questions that are obviously 'quite hard' - hard enough to need time to answer. His test is a thorough test of logical problem-solving ability, rather than any of factual memory, speed-thinking, general knowledge or literary learning.

More info

Standard IQ tests, such as the famous Stanford-Binet scale, are designed to measure scores clustering around 100, the average in the human population. It's hence unsurprising that there is considerable interest in high IQ tests - see Darryl Miyaguchi's compilation at Uncommonly Difficult IQ tests.

 The test reproduced here was designed in 1979 by San Fransisco systems analyst Kevin Langdon to measure IQs in the range 125 to 180, and was claimed at that time to be the world's most difficult IQ test. It is unsupervised and untimed. The only rule was that you were honour-bound to take it on your own, without help, but you could spend as long as you wanted on it - hours, days or months. What mattered was not the time taken, nor any special expertise or knowledge you may have, but just your power of attention and ability to find the only logical solution by thinking deeply.

When this test was first published, those brave enough to try it sent their completed answer sheet to Omni for marking, and received an IQ report in return. The correct answers were not revealed. This raises an interesting question as to who decides when an answer is correct.  The designer would presumably aim to design each question so that the correct solution, when properly understood, is self-evidently the only possible solution: simple, unambiguous, elegant and satisfying. Then the quiz-taker can also be confident that they have found the right answer.

Way back in 1979, after a month of concentrated deliberation, I sent in my score sheet and received a reply saying my IQ was 169. That sounds impressive, but only corresponds to being 1 in 100,000 persons. At the time I wasn't sure of the answers to about 6 of the 56 test questions so I had to make an educated guess, eliminating the obviously wrong suggested answers. Looking again in 2019 after 40 years of subconscious deliberations, ha ha, I have a satisfying answer for all but one of the questions (Q47), each with a high probability of being the same as Langdon's secret answer. Interestingly, I feel smarter these days, albeit older and slower, in that I can now easily see the answers to a couple of the questions that vexed me way back then.

This test has not been marked by Langdon for many years. Out of respect for the spirit of the original, I do not wish to reveal my set of answers. It has been pointed out to me that some of the answers have been circulating on the web for decades. However, if you still want to test yourself for fun (yeah right - fun!), you can always send me an email with all your final answers at once, but only when you are sure you are done!
I will be able to reply with a single number: the number of your answers with which I agree. If our number of disagreements was low, say less than 5 out of 56, we could discuss the reasons.

In the original scoring method, in order to counteract the possibility of guessing (with a 20% chance of being right for each multiple choice question) a 'raw score' was computed equal to the number of correct answers minus one fourth of the wrong answers. Barring statistical flukes, this means a set of all-guessed answers would score zero.

Langdon provided a norming document with his original marking sheet, to show how his scores were derived and how they related to standard tests, and also provided some supplementary information later in the May 1980 edition of his Sigma Four magazine.

 

During my academic life as a research physicist, I often witnessed races to offer quick solutions to problems arising in our laboratory brain-storming sessions. I suspect that the usual 'winners' succeeded via a strategy (perhaps unconscious) of aiming to be right around only 90% of the time, and not fretting too much over any confounding doubts or alternative theories that might delay an answer.  However, to do better, to be right 99% of the time, those doubts and alternatives need to be examined, and that takes longer. How often we ended a brain-storming session all bamboozled into a hasty plan, only to realise after the meeting had been disbanded, and after further thought and effort in the lab, that plan A was inherently flawed. Thank God for plan B.  

Still - horses for courses; a good team contains people with a variety of skill sets. What's nice about Langdon's 'Omni' test is that it gives deep thinkers a chance to challenge themselves. I hope you have fun with it.