The Halo Effect
Decision Hygiene, part four: Matt Damon, handsome candidates, and jumping answers
In 2006, Alex Todorov briefly showed study participants the pictures of two frontrunners in a political election. “Which of these two looks more competent?” he asked. The participants didn’t know the candidates. They didn’t even know that they *were* political candidates. Nevertheless, viewing each pair of photos for less than one second ”was sufficient to predict the winner in about 70 percent of the races for U.S. senator and state governor in the 2006 elections”.
This is an example of the Halo Effect: one element of an experience is so powerful that it produces a sort of halo that blinds you to other relevant elements in the experience. Advertisers rely on it when they use celebrities in their commercials. Just to give one example…FTX was a very risky product, a cryptocurrency platform that ultimately lost billions of dollars. Its CEO is now in jail. But when Tom Brady, Matt Damon, and other famous people promoted FTX, they made it seem more mainstream and desirable, even bold:
This halo of excitement enticed some investors to make a substitution error, similar to those described in the previous post. Fully researching a cryptocurrency platform is hard. Following the lead of a celebrity is easy. So instead of using System Two to answer a hard question (“Should I buy this product?”), we use System One to answer an easy question (“Do I recognize and admire this person?”).
The Halo Effect isn’t limited to ads. You experience it every day, when you get a good feeling about your friend's new boyfriend, or decide to try out a restaurant as you walk by it for the first time. These gut reactions can be very useful – sometimes you have to make quick decisions about people and restaurants! – but they can also be wrong. Maybe you just liked the guy because he laughed at your joke. Maybe the restaurant hired a great logo-designer, but a poor chef.
Let’s see this effect in action. Suppose you would like to hire a tutor. When introducing you to an applicant, I describe him with these four adjectives:
intelligent
persistent
cunning
arrogant
The first two traits seem quite good, the third is not great (but maybe okay if used to help students?), and the fourth is unequivocally awful. Nobody wants an arrogant tutor. But maybe you could train him to be nicer?
Now suppose I had introduced him slightly differently, arranging the adjectives in this order:
arrogant
cunning
persistent
intelligent
I’m immediately not interested when I hear that he’s arrogant. Even ‘persistent’ seems bad to me…maybe he’ll argue too much and refuse to admit when he’s wrong! And by the time I get to ‘intelligent’, the trait no longer seems very positive (he’ll just use his intelligence to annoy me and our students). The four traits are exactly the same, but somehow my experience of them, and therefore my opinion of the candidate, is different.
This same cognitive bias affects us when we consider answers on standardized tests. I’ve noticed that many students don’t actively make sense of this paragraph:
They get a vaguely positive impression of John Huff, but they don’t note the hyperbole (like “...the only god living in the whole of Green Town…”). As they scan the answers, they are mostly waiting for something to jump out at them, and they are not disappointed:
Students immediately like the beginning of the first answer choice (“emphasizes John’s physical strength and intelligence”). In fact, they like it so much that they don’t even read the second part of that answer choice, which is clearly wrong (the passage doesn’t indicate John’s view of himself). If they read the second choice at all, it now seems a bit negative (“exaggerates” doesn’t quite match their vaguely positive impression), even though it’s correct.
Students can avoid this error by engaging System Two. A careful re-read of the paragraph might alert the student to the exaggerated description of John Huff. Methodically reading to the end of each answer choice might shatter the halo of familiarity created by the opening phrase of the first choice. But rather than answer the hard questions (“What does this paragraph say? How do these answers relate to the paragraph?”), students will often answer a much easier question: Which answer gives me a positive feeling?
Great explanation Mike !