I have a hypothesis that humans are better able to identify the largest (or smallest) in a set of objects, compared to their ability to identify an average object, i.e. the one closest to the mean.

Has anyone tested response times or accuracy of people given a collection of objects or shapes, when the subjects are asked to identify the most extreme object, as opposed to the object that best represents the "average" object? Or for a set of numbers, are people faster and/or more accurate at identifying the largest number as opposed to the one closest to the mean?

  • $\begingroup$ Interesting question - I think it's going to be hard to test because the task is so different, you might not be measuring what you think you are and instead measure something trivial. $\endgroup$
    – Bryan Krause
    Mar 8 '18 at 17:13
  • $\begingroup$ I would say that it depends on the person. For someone who likes accuracy, it might take them longer to estimate than others who are less fussy, especially with mean average. Also, you are talking about mean what when referring to objects? Mean height, width, general size? Interesting question but could be broad. $\endgroup$ Mar 9 '18 at 16:58
  • $\begingroup$ Good comments, thanks. I was thinking about physical objects or shapes. For example, if you have a set of 7 circles of different diameters, one will be largest, one smallest, and one closest to the mean size. I suspect people are better at identifying the largest/smallest, and less good at identifying the one closest to the mean. In contrast, if I write a computer program to do that task it is equally easy to identify the largest or the average. I probably don't know what key words to use, but I can't find any papers with data on this kind of comparison. $\endgroup$
    – Brian I
    Mar 10 '18 at 21:06
  • $\begingroup$ yes, good question. i think this question has an answer that is part perceptual and part cognitive psych. regarding circles analogy: perhaps our visual perception is biased towards focusing on the "odd one out". or perhaps detecting averageness requires additional cognitive processes, such as performing some mental test of "most similar to everything else". $\endgroup$
    – faustus
    Mar 11 '18 at 7:10

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.