As I’ve discussed before on this blog, I love doing psychology questionnaires online an have reached the point where I can recognise several of the standard instruments used, if not by name then at least by purpose/intent. The OCEAN/Five Factor Model test is pretty standard, testing for Openness, Conscientiousness, Extraversion, Agreeableness and Neuroticism. There’s the Dark Triad test (if I recall correctly, that’s Psychopathy, Narcissism, Machiavellianism). There’s a variety of standard memory tests (e.g. “have you seen this word in the previous section?”) and so on.
One of the tests that occurs frequently is the “Reading the Mind in the Eyes” test, created by the Cambridge psychologist Simon Baron-Cohen. It is usually used as a test of “empathy” or “theory of mind, although there are some clear issues with the assumptions behind that assessment (my search on DuckDuckGo found plenty of commentary, as well as places to try the test myself). As a test of reading emotions in others’ facial expressions it seems to have some merit, although again, there are objections to that.
I tend to doubt my ability to read others’ emotions, especially in social situations. At times, I doubt others’ ability to read my emotions also. So I feel challenged and struggle when I encounter this test. Normally, when doing an online questionnaire for someone’s research project, you don’t get told what your score is, or what it supposedly means. I felt convinced that I do not do well on it, but when I decided to write about it and did that DuckDuckGo search, I found sites that let you know your score.
The format of the test is that you are shown a sequence of photographs cropped to show the eyes, with four options of the emotion or state of mind each pair of eyes might be indicating. Usually, the options are arranged with one option at each corner of the photo. You are told there is no time limit, but to try to answer as quickly as possible. Your job is to choose which of the four options is “correct”.
The ways in which I have struggled are:
- The emotion I believe is being shown is not one of the options
- I find it hard to choose between two or more of the options
In both cases, it takes me much longer to choose an answer. In the first case, I have to reason which of the options is closest to the emotion I believed was being shown, or if the emotion I immediately identified is opposite to the range suggested, then I have to reason which of them is least unlike the emotion I saw. With the second case, I have to reason by a process of elimination (“I am certain it isn’t those, what else do I think it isn’t?”)
Not all questions were hard: sometimes I had an immediate response, and the answer was there, or I had no immediate response but when i saw the four options I had an immediate conviction that one answer was clearly “right”.
With the experience of having difficulty and doubts about my answers, I believed I must be doing badly at it. So I was surprised when, two weeks ago, I used an online test and scored slightly higher than the average for a neurotypical male (although well within the Standard Deviation).
One conclusion would be that, when given a choice of four options instead of the full range, and the time to reason out the answer from whatever principles I have available rather than having to make an instant read in a social situation, then I can do quite well. This would argue against the test being valid as a predictor of performance in “live” situations with real people.
An alternative conclusion is that I am better at reading other people’s faces than I believe I am, and that my difficulties in “live” situations come from my self-doubt rather than from an underlying inability. By this interpretation, the difficulties I reported are the consequence of self-doubt more often than they are a consequence of not knowing. Of course, it is harder to explain the first case listed above by that logic.
As it happens, I had forgotten when I finally sat down to write this post what the average was supposed to be, and if it was the same as that reported in the .pdf of the paper outlining the new version (it turns out, it was). So I decided today to take the test again, to get to that information.
This time, my result was significantly lower, and about average for a person with Asperger’s syndrome or high-functioning autism (and just outside the Standard Deviation for a neurotypical male). That would tend to support the first conclusion. I took less time over my answers instead of trying to reason them out carefully; if self-doubt had been the primary cause of my difficulties then this ought if anything to have improved my score as I relied more on “gut feeling”. It occurred to me that my mood might have also affected my result: last time, I was excited and enthusiastic about taking the test and finding out how I really did. This time, I have had a tiring weekend so far and felt quite grumpy. I am left wondering if it is possible that feeling tired or grumpy makes me less able to read emotions, or more inclined to read “positive” emotions wrongly as negative emotions?
I believe that there are probably elements of all these theories or explanations of what I experience in social situations, and the results from the two attempts at the “Eyes” test. I genuinely do read emotions wrongly or confusedly in a lot of situations. Probably my mood also affects how well I can interpret other people’s moods. Because I am introverted, a lot of social situations leave me feeling tired, stressed or antagonistic. Sometimes the difficulty I experience in reading others’ emotions and intentions also make this worse. But perhaps the stress I feel is making it harder for me to read other people correctly, too?
I am also convinced that sometimes I have read the situation accurately, but lack the confidence in my assessment to believe that I have done so. In a few situations where the company has been close enough and I trust them enough that I can check in with them and it won’t seem too weird, I have been able to ask if my read is correct and I generally am told that yes, I have got it right (once or twice I was wrong, but very rarely). This gives me a basis from which to say that self-doubt is probably also a factor in my difficulty. Curing the self-doubt is harder, because there are still too many occasions where I haven’t been able to check in but have clearly misread something – I genuinely don’t get it right often enough, or don’t have any clear answer (as opposed to an answer that I then doubt).
The test itself may or may not be any good. For me, the value is to prompt once again thoughts about how I relate to the social world and what I can do to deal with the issues I find in doing so. Simply understanding what my difficulties are (especially that self-doubt one) can be helpful in terms of interpreting what I feel when faced with confusing situations.