People are quick to make certain judgements. Our minds have evolved to excel at drawing conclusions from minimal data and thinking quickly using heuristics and mental models.
But some judgements are more reliable than others. When it comes to judging people, first impressions are quick, easy, and compelling — but are they accurate?
Alex Todorov found that we can form impressions of people in 1–10th of a second by relying on certain cues that aren’t always reflective of what we think they are. We stereotype people.
“One-tenth of a second of viewing provided ample face information for our participants to make up their minds. The effect of additional time was to simply increase confidence in their judgments.”
Todorov has conducted research that shows that people judged the competence and likability of faces. What the participants didn’t know was that the faces were of political candidates.
The faces judged more competent were also the people voted into office (likability didn’t say very much). What facial features reflect competence? A strong chin and slight confident smile.
Most of us probably think that strong chins and slight smiles are absurd factors to judge ability on. Surely there is more informative data available, why didn’t voters recognise this?
I assume we’re looking for a better answer than to call them lazy. Perhaps another possibility is that first impressions and stereotypes tend to overpower other information, even when we have our thinking caps on.
Daniel Kahneman and Amos Tversky ran several studies showcasing how our stubborn attachment to impressions and stereotypes tend to overrule or obscure more rational approaches, the effect of which has been dubbed the representativeness heuristic.
Let’s meet some people.
Linda is 31 years old, single, outspoken, very bright, and majored in philosophy. As a student she was deeply concerned with discrimination and social justice, and participated in anti-nuclear demonstrations.
Which of these statements is more likely to be true of Linda:
- She is a bank teller
- She is a bank teller and is active in the feminist movement
While Linda strikes most people as being a feminist, the second option is not the correct choice. It breaks the conjunction rule — the probability of A+B cannot be more likely than either of them alone.
If the odds of her being a bank teller are 20%, and the odds of her being a feminist are 99%, the odds of her being a feminist bank teller are 19.8%.
When Kahneman and Tversky asked the original question, rather than 2 they had 8 statements to rank, including options like an elementary school teacher and insurance salesperson.
In that setup most people erroneously had the conjunction ranked as more likely than a solo bank teller, but this could be because people didn’t compare every option against every other.
So they reduced the options to the two we see here, hopefully making the comparison more obvious and reducing the error. But the error survived, in fact 85% of people ranked the conjunction as more probable.
Now Meet Chris. His profile has been taken randomly from a pool of 100 people, 70 of which are engineers, and the other 30 are lawyers. Here’s his description:
Chris is a 45-year-old man. He is married and has four children. He is generally conservative, careful, and ambitious. He shows no interest in political and social issues and spends most of his free time on his many hobbies which include home carpentry, sailing, and mathematical puzzles.
What are the odds that Chris is one of the engineers? Has to be an engineer right? He doesn’t sound like a lawyer, and perhaps even more importantly, the initial 70:30 ratio between engineers and lawyers lends support to your assessment.
The funny thing is Kahneman found that the proportions of lawyers and engineers made little difference. If there are 70 lawyers and 30 engineers, Chris is still judged an engineer.
When people aren’t given a description and just asked to estimate what a random person would be, they don’t make the error. All they have at that point is the numbers, and so their judgements accurately reflect the odds.
Once people meet Chris the numbers are left out in the cold. This mistake is called base-rate neglect.
You put Chris’ description to the side and pull another profile from the same group of 70 engineers and 30 lawyers. You get this description of Larry:
Larry is a 30-year-old man. He is married with no children. A man of high ability and high motivation, he promises to be quite successful in his field. He is well-liked by his colleagues.
Do you think Larry is more likely to be an engineer or a lawyer?
He doesn’t really sound any more like an engineer than a lawyer, he isn’t as easy to stereotype as Chris, but if we’re going on the odds it makes sense to guess he’s an engineer — there are over twice as many of them.
That’s not what people do — they tend towards 50/50. They completely ignore the proportions and assume that because the description doesn’t suggest anything it must be equal odds.
That’s rather strange. People respect the numbers only if there’s no description. The moment we paint a picture of this person, even a picture that doesn’t point in any direction, the numbers become completely irrelevant.
Daniel Kahneman sums it up:
“Evidently, people respond differently when given no evidence and when given worthless evidence.”
The Final Judgement
Descriptions and stories are much more successful at capturing the attention of the mind than facts and statistics. This is rather widely known and clearly has an impact on stereotypes.
It doesn’t help that in real life we often don’t have reliable statistics on hand to refine these judgements — not that we’d use them if we did. In most daily situations, we make decisions based on limited information and limited time.
But before we go ruing stereotypes and wishing them away, we should consider their purpose and benefits — it may be that our typical judgements can be rather accurate.
Psychologist Paul Bloom notes:
“You know, you don’t ask a toddler for directions, you don’t ask a very old person to help you move a sofa and that’s because you stereotype.”
“You know my name, you know certain facts about me and you could make certain judgments. You could make guesses about my ethnicity, my political affiliation, my religious beliefs and the thing is these judgments tend to be accurate. We’re very good at this sort of thing and we’re very good at this sort of thing because our ability to stereotype people is not some sort of arbitrary quirk of the mind, but rather it’s a specific instance of a more general process, which is that we have experience with things and people in the world that fall into categories and we could use our experience to make generalizations of novel instances of these categories.”
Stereotypes are formed haphazardly by experience, they represent an intuitively held probability spectrum, they are educated guesses based on our observations of a population of people with similar characteristics, and those guesses can happen in fractions of a second.
Even if that guess is correct regarding the average person from that population, they might still be wrong about any individual case, and sometimes being wrong can be hurtful and damaging.
We should think about what stereotypes we’re using and how much confidence to place in them. And as difficult as these studies make it seem, we must ensure that we are not overlooking other crucial information. Remember that next time you meet a Linda, Chris, or Larry.