Do you know yourself?
There are three things extremely hard: steel, a diamond, and to know one’s self.
Benjamin Franklin, Poor Richard's Almanac, 1750.
Franklin was writing over 250 years ago. Surely we humans have learned strategies since then to aid self-insight – and avoid well-known pitfalls? Most of us are familiar, for example, with the better-than-average effect, the finding that most of us rank ourselves above average at everything from driving ability to desirable personality traits (even though of course we can't all be right). So armed with this kind of knowledge, are we better placed now to view ourselves accurately? And if not, how can we get better at this – and what benefits can we expect? The following studies provide some illuminating answers…
I feel like I’m smarter than other people
The better-than-average effect may be old news, but results from a systematic survey of Americans' beliefs about their own intelligence (the first to be conducted in 50 years) which we reported in 2018 found that about 70 per cent of men and 60 per cent of women agreed with the statement: ‘I am more intelligent than the average person’. The team was forced to conclude that Americans’ ‘self-flattering beliefs about intelligence are alive and well several decades after their discovery was first reported’. Other work has found that our over-estimates of our intelligence can be staggeringly huge – around 30 IQ points, on average, according to a study by Gilles Gignac and Marcin Zajenkowski published last year (which also found that we tend to over-estimate our romantic partner’s intelligence ever more than our own). The sobering lesson is that you're probably a lot less smart than you think are.
OK, but I at least know the limits of my knowledge
The ‘Dunning-Kruger effect’ relates specifically to the tendency of people who are poor at a task to over-estimate their ability at it. As David Dunning has written: ‘The scope of people's ignorance is often invisible to them’. (Although see Gignac and Zajenkowski again, with a paper from this year arguing that the effect is (mostly) a statistical artefact).
This over-confidence can be dangerous both to the individual, and to others. For example, a US study of student pilots led by Samuel Pavel found that those who’d scored lower on a pilot knowledge test ‘grossly overestimated their ability’ while higher-scoring students tended in fact to under-estimate theirs. The same effect has been noted among other groups, for example among chemistry students. In the 2018 study by Jeffrey Webb and Andrew Karatjas, students who’d scored less than 50 per cent on one exam had predicted that they’d get an average of 69 per cent, while their actual average mark was just under 37 per cent. That’s a massive discrepancy. If you don't know that you’re under-prepared for a test, this is clearly a problem.
Fine. But you can’t deny that I know who I am!
How well do you know your own personality as it fluctuates from moment to moment? This was explored in a recent study, which found that participants had self-insight into their momentary levels of extraversion and conscientiousness, but weren’t great at rating how agreeable they were being at any given time. As Jessie Sun and Samine Vazire write: ‘This apparent self-ignorance may be partly responsible for interpersonal problems’.
However, when it comes to personality in general – your fairly stable, trait levels of extraversion, agreeableness, and so on – there’s some rare good news in the field of self-insight. According to a large-scale review of data on self-reports of personality vs personality ratings from others, published by Hyunji Kim and others in Psychological Science in 2018, we're actually pretty good at judging ourselves in this way. In fact, the work revealed that if anything, we’re harsher judges of our personality than other people are. This was a surprise to the researchers, who'd assumed, based on other work in this field, to find a positive self-bias.
But I know what I like, right?
If you love salt and vinegar crisps, say, but hate cheese and onion, then fair enough.. No one's going to argue that you're mistaken. But if you tell me you love coffee, I might be less accepting. It turns out that it’s not that easy for us to tell the difference between liking something and wanting it. A clear example of this comes from a study I reported on earlier this year (tinyurl.com/y45c63wa), which found that ‘heavy’ coffee drinkers (people who drank three or more cups a day) actually want coffee a lot more than they like it. The implication is that they drink it mostly or entirely to feed their addiction, rather than for pleasure.
How can I get better at knowing myself?
Well, the people around you could be more honest…In general, other people don't help us to correct our biases. Too often, feedback from employers, family and friends is vague, and overly positive, according to research led by Zlatan Krizan at Iowa State University. ‘As a society, we make the wrong trade-off by thinking that boosting self-esteem is going to boost performance, and that rarely happens,’ Krizan says. ‘That empty praise of telling someone they're great, or pretending there are not skill differences when there are, can really become a problem.’
As well as seeking honest feedback, and bearing in mind the better-than-average effect, it could also be worth practising humility. Research published in the journal Self and Identity suggests that people who are more modest about their degree of self-knowledge actually know themselves better.
But is it important to have accurate self-knowledge?
No-one likes a narcissistt. But there could be some benefits to thinking you're better than you are. They relate to the optimism bias – the fact that we tend to over-estimate the likelihood of positive events in our lives, like getting a top promotion, and under-estimate our chances of suffering everything from a divorce to a car crash. Optimism is important for mental and physical health. So perhaps having inflated self-views is important for our wellbeing, too. In fact, according to the results of a study by Joyce He and Stéphane Côté published last year in Nature Human Behaviour, this may well be the case. The participants completed various tests of cognitive and emotional abilities and then reported how well they thought they'd done on all these tests. They then spent a week completing daily diaries, in which they reflected on their levels of satisfaction with their career, relationships and life in general. In terms of levels of satisfaction in all of these areas, the researchers found no support for benefits of self-insight.
The researchers found that having accurate self-insight was not related to higher levels of satisfaction in any of these areas. The data even suggested that people who most over-estimated their abilities had the highest levels of life satisfaction. Still, it’s worth noting that feeling great about your life and performing to your highest level are two different things. People with accurate self-knowledge may be more driven to improve, and achieve more.
A word of caution
Our understanding of the potential pluses or minuses of erroneous vs accurate self-insight is, according to a recent review in Advances in Experimental Psychology, ‘clouded’ by all kinds of issues, including differences between studies in the way self-insight is measured, and also fundamental differences in types of self-deception. (Self-deception may sometimes stem from an individual’s desire to defend their self-esteem, but other times exist simply because a person hasn't really engaged in close self-assessment.) Jennifer Beer and Michelle Harris write that the currently available research ‘does not allow us to confidently conclude that self-insight has advantages over some types of self-insight failure (or vice versa).’ They conclude by calling for ‘more systematic investigation of why, when, where and for whom self-insight is costly or beneficial.’
Find our Research Digest at www.bps.org.uk/digest
BPS Members can discuss this article
Already a member? Or Create an account
Not a member? Find out about becoming a member or subscriber