|The X-rite online color challenge|
So, you drag and drop the colored tiles to arrange them in graded order between the blocks on each end of the four rows. Then you get your score, which can range from 0 (perfect) to, well, very high (e.g., some 20 - 29 year old male has scored 444445389, whatever that means). Maybe you've seen this test (ok, advertisement). Maybe you've taken it. Maybe you've taken it twice, or three times. It's kind of mesmerizing. But does it tell you anything about your ability to discriminate color?
I got a 20 the first time I took this test, and then, shamed by my co-test takers, took it again and got an 8, resulting in fewer but different problem spectrum areas. A friend, an artist, first got an 8. then a 4, and still not satisfied, took it again and finally got a perfect score. My daughter got a perfect score the first time she took it. As did my friend's partner, another artist who, incidentally, has been reading a lot lately about color. Could that explain his score? Yes, they bragged.
But what is this test actually testing? Anything about how we really see color? Well, it would have to either be a sloppy version of color acuity (whatever that is) if scores can bounce around as they do, and if our problem spectrum areas can change. Or is it measuring trainability for taking tests of visual acuity? Or ... nothing?
Should we make an actual serious point or two here? Meaningless tests like this one aren't the only kinds of tests that can claim to be measuring one thing but actually measure another ... or nothing. The easy target is psychological testing, from IQ tests to fMRI scans, but hearing tests in noisy doctor's offices are another one, and. hell, let's even throw in background checks for gun purchases.
Or is the real problem the test's claim to precision? We have neural wiring and chemical (and mechano-chemical) processes that provide input to our wiring from the outside world. Even assuming no change in the wiring (no brain cells die, are produced, or change their synapses) or the detectors (no retinal or auditory or olfactory cells die or are produced), there will be probabilistic variation in strength of signals detected and sent to the wiring and among the responding signals sent among the wiring. In this sense, no one has a level of visual acuity, or other kinds of detection of the world. Cold feels colder sometimes than others, smells are more intense at times, and so on.
We vary in our ability to do a task, and only someone very rash would suggest that we understand all the sources of that variation. Reducing the phenomenon to individual causal elements--specific neural connections, specific genetic variants, may work for some strong-effect instances, but is illusionary, as so much research shows, most of the time. We may get approximations, but we usually don't know how approximate.
This is the challenge of complexity. It is one of the central challenges to science today. It's a serious challenge. Our natural tendency to wish it away or claim that we understand it, often obscures how deeply challenging it is. Maybe science doesn't even know how to frame the questions that need to be asked about it. If every acuity test, or every case of diabetes, is complexly different, then the 'it' we want to explain isn't really, well, 'it'.