Attribution bias includes a set of more specific biases where we may attribute behaviour to fixed personality traits or characteristics of a person rather than specific circumstances or actions. For example, a child may be labelled a genius or gifted when they actually had a supportive environment and worked hard, or someone may be labelled a loser rather than recognising specific circumstances that led to some failures.

Criticism — one of the four horsemen of relationship apocalypse — can become toxic when it's attributed to someone's personality traits. For example, calling someone 'lazy' rather than sharing how it makes you feel when they don't keep the place tidy.

A form of attribution bias may be applied to whole societies via the Destiny Instinct.

Specific examples of attribution bias include the fundamental attribution error and self-serving bias.

Attribution bias

Optimism bias is a tendency to believe that things will turn out well in spite of past evidence or circumstances.

It can be extremely helpful. It allows us to attempt things that many may deem impossible. It probably helps drive entrepreneurs even when everyone is doubting them. Optimists are often healthier and happier.

But it can also undermine us. We might overestimate our chances of success because of what we want to happen rather than what's likely to happen. We may feel pressure from others to give a rosier outlook — like when you might underestimate a timeline when speaking with your boss. We might want others to succeed, or they may be paying us money and be hoping for good news. We probably think we'll get more done next week. We probably think we'll be more disciplined than we will be. We might think that this time it'll be different, or that everyone can't be wrong. Or we may be just discounting evidence without realising it.

Related: Hofstadter's Law, survivorship bias, Kitty Hawk moment

Optimism bias

The Rhyme As Reason Effect suggests that we're more likely to consider something as accurate or truthful when it rhymes. In a lovely study that asked people to consider the accuracy of both rhyming and non-rhyming aphorisms, "Woes unite foes," or "What sobriety conceals, alcohol reveals," for example, were considered more accurate than "Woes unite enemies," or "What sobriety conceals, alcohol unmasks."

A classic example outside of old sayings is OJ Simpson's trial where the defense attorney memorably noted "If the gloves don't fit, you must acquit!"

You can read the study in the delightfully titled paper, Birds of a feather flock conjointly: Rhyme as Reason in Aphorisms (pdf), by Matthew McGlone and Jessica Tofighbakhsh.

In a related note I was always curious how "Beer before wine, you'll feel fine," tallied up with what I heard in the US: "Beer before liquor, never been sicker," or even "Grape or grain but never the twain." It turns out from a neat 2019 study that, whichever way round you rhyme them, they have no effect on your hangover despite how persuasive the Rhyme As Reason effect may be — even after a few beers.

The Rhyme As Reason Effect

Brilliance bias is the tendency for people to think of 'brilliance' as generally a male trait. With fewer women generally recorded in the history books, the brilliant people that come to mind from a young age tend to be men.

Underrepresentation has consequences. Careers that are generally understood to require 'brilliance' to succeed such as maths, physics, or philosophy, are likely to have fewer women working in them. Even at school age, attitudes start to shift in both girls and boys. For example, having interest in games that require you to be 'really smart' or whether, when asked to draw a scientist, we draw a woman or a man. We need to stop teaching brilliance bias and we need to celebrate the many female role models we have.

My sister shared with me a page on drawing famous faces that showed 49 men to draw, and an example of the board game Guess Who having as many women to guess as it has men with facial hair as an attribute.

I learned about brilliance bias, including the examples here, from Caroline Criado Perez's eye-opening book Invisible Women.

Brilliance bias

In much of life we're used to things adding up — got one scoop of ice cream? Two is likely to be better. Buying one thing? Getting another free is probably a good thing. The dilution effect, however, illustrates an interesting action of our brain when evaluating our options. Niro Sivanathan showed that when displaying possible side-effects of drugs, if you were to add another side-effect, a minor one like 'itchy feet', instead of the extra negative side-effect increasing our overall perception of the risk of the drug it actually lowered it. So, people would deem a drug with the side effects heart attack, stroke, and itchy feet, as less risky than one with just heart attack and stroke.

In another clever experiment by Christopher Hsee at the University of Chicago, people were willing to pay less for a new dinnerware set when the set was actually larger but also included some broken items. Even though a set had more intact items the presence of a few broken ones decreased the set's perceived value.

And the same may be true when making an argument to persuade someone to see our point of view. If you have 2 strong points and 1 weaker one, your argument may actually be made weaker by adding the extra point. As Niro Sivanathan, an Associate Professor of Organisational Behaviour at London Business School, explains: Arguments don't add up. They average out.

Next time you're making your case, consider sticking to your strong arguments.

Hear Niro Sivanathan explain the dilution effect in his 10 min TED talk The counterintuitive way to be more persuasive

The paper: Sivanathan, N., Kakkar, H. The unintended consequences of argument dilution in direct-to-consumer drug advertisements.Nat Hum Behav 1,797–802 (2017).

The dilution effect

There are a lot of reasons why it can make sense to go with the majority. Going with the popular view is a shortcut for taking the time and effort to make our own decisions from scratch. We might reason that if others believe it, then it's probably true. But we can also keenly feel the desire to conform to the majority view because it's uncomfortable to hold a different opinion — the easy choice is so often to go with the flow even if, deep down, you disagree. So we get the bandwagon effect.

The bandwagon effect can lead to the Abilene paradox — where a group can make a decision that no one in the group agrees with. On a night out it may lead to just following the crowd, but in a company it could lead to launching a product that no one thinks is good or, more seriously, in safety critical industries like airlines, if people don't feel comfortable speaking up against the group or authority it could lead to accidents.

The bandwagon effect

So the story goes, as told by Stephen Sigler, Nature May 1989:

"The US military studied fighter planes returning from missions to try to improve their survival rate and were considering adding heavy armour to those parts of the plane that tended to show the greatest concentration of hits from enemy fire, until statisticians pointed out the fallacy of that argument. The more vulnerable parts of the plane were those with the fewest hits; planes hit there tended not to return at all. The single most vulnerable part, the pilot’s head, was without serious scar in the sample of planes that returned."

It’s easy to draw correlations from what we see in front of us. But what we see usually represents just a small part of what has happened. Focusing on the evidence we can easily see at the expense of that we can’t leads to survivorship bias. To say it another way, when Bill Gates drops out of college and starts Microsoft it might seem like dropping out is a path to success for others too, but that ignores all the dropouts who didn’t create Microsofts and consequently you didn’t hear about.

Silent evidence — the evidence that we don’t or can’t easily choose to consider — is a term from Nicholas Nassim Talleb.

When trying to discover the veracity of the planes story I enjoyed Bill Casselman’s American Mathematical Society article on The Legend of Abraham Wald. The postscript points to some of the source behind the story including a mention of Stephen Sigler’s letter in Nature quoted above.

I covered survivorship bias before but like this story so much I thought it was worth doing again.

Survivorship bias

It turns out that in general, the intensity of emotion, or the affect, of our bad memories seems to fade faster than for our good memories. This means that, say, our happy memories keep us happier for longer and our bad memories don’t affect us as much in the future. It’s called the fading affect bias as the fading is biased more strongly for negative memories.

This is handy as it naturally gives us a more positive disposition. So, for instance, after some time has past, we may find we remember the good parts of a holiday — like an amazing beach — more clearly than the bad — like the mosquitoes at the beach. And thinking about the good parts keeps making us happy longer than thinking about the bad parts might make us sad. Brilliant!

PS: This is a different use of affect than the common affect/effect confusion.

Fading affect bias

We have a natural tendency to think that big events must have big causes. For instance, in one study, participants who were told that a plane had crashed and everyone died were more likely to believe it was a terrorist incident than one in which they were told a plane had crashed but people survived. It might also help explain why, when we want to roll that big lucky number to win, we might find ourselves giving the dice an extra shake before rolling, as if the extra effort might lead to the better outcome.

Normally, proportionality bias makes a lot of sense, but sometimes it leads us astray. Working with software, for example, might help train this out of us as a mere misplaced comma can bring entire systems grinding to a halt. And small changes to ecosystems, or, say, a small temperature rise on a planet, can lead to massive consequences.

Proportionality bias is considered one of the drivers of conspiracy theories. For example, we might be inclined to think that someone like Lady Diana couldn’t possibly have just died in a driving accident — there must be more to it. Or JFK couldn’t have been killed just by a lone gunman — there must be a larger plot at work.

Proportionality bias

Change of large complex things like people, countries, religions or cultures tends to happen slowly. And because the change is small we might think they’re not changing at all. Our instinct may be that they can’t in fact change because of their innate characteristics. But the destiny instinct misleads us.

Small change adds up to big changes over enough time. And while things may not seem to change in a week or a month or a year, when you look back over decades huge change can happen. Changes in poverty, live expectancy, economies, women’s rights, or attitudes to homosexuality — things that may have seemed at one time that they were hardly changing at all —  have all transformed when we look back over sufficient time. Change that may not have seemed like it was adding up to much is still change.

Read more from the Roslings: Culture, nations and religions are not rocks — they’re always changing.

And, at a time of the year perfect for reflection, here are 4 ways to control our destiny instinct from Gapminder:

Keep track of gradual improvements. A small change every year can translate to a huge change over decades.

Update your knowledge. Some knowledge goes out of date quickly. Technology, countries, societies, cultures, and religions are constantly changing.

Talk to Grandpa. If you want to be reminded of how values have changed, think about your grandparents’ values and how they differ from yours.

Collect examples of cultural change. Challenge the idea that today’s culture must also have been yesterday’s, and will also be tomorrow’s.

The Destiny Instinct

A deceptively simple, almost too simple to notice, heuristic —  a mental shortcut to solve a common problem — where we consider the likelihood of events by the ease with which relevant instances come to mind. In simple terms, things that we can easily recall or where we can think of examples, we generally consider as more important or more common than other things.

If we were asked to consider the divorce rate in the community, we might evaluate that by considering how many of your acquaintances are divorced. If none of your friends are divorced you’re naturally less likely to think the divorce rate locally will be high.

In many situations it makes good sense to give some credence to what you remember easily. However, in situations involving probabilities in particular the heuristic will lead us predictably astray such that more often I think of it as an availability bias — a predictable pattern of thought and behaviour leading to incorrect conclusions. 

So, the vividness of shark attacks causes us to overrate how likely they are to actually occur. Or we’re more likely to worry about lions than hippos, even though hippos kill more people each year than lions. Or the risk of being struck by lightning vs stuck by a car crossing a road. Or, the prevalence of violent crime on the news compared to the likelihood of you being a victim. Or the fact that it’s cold outside and yet the planet is gradually warming.

Here’s a 1973 Kahneman and Tversky paper: Availability: A Heuristic for Judging Frequency and Probability

The availability heuristic (and bias)

The Implicit Association Test (IAT) is an amazingly simple way to uncover if and how strong your implicit biases may be. For example, you could test if you have a bias on gender roles (as in the sketch), race, religion or in fact any attitudes you may have. You could forget about reading this and simply go take the Implicit Association Test online now and you’ll get the idea and perhaps surprise yourself.

In the example I gave in the sketch, you could see how strong your biases are for traditional gender roles by measuring the difference in time it takes to sort first career words towards men and home and family words towards women, and then sorting the other way around.

It’s fascinating, and a little uncomfortable, to see your implicit attitudes so clearly come to light. Don’t forget to try taking a test.

The Implicit Association Test

I like this one because it basically shows that most of us can just relax a bit and not stress about what other people think — because they’re probably not thinking about us, they’re probably thinking about themselves.

I probably only remember it because of the effect effect.

The spotlight effect

It’s named Streisand after Barbara Streisand tried to have an aerial photo of her house removed from a set of California coastline photos intended to document coastal erosion. Views of the photo apparently went from 6 to 420,000.

The Effect Effect

It turns out when our body gets excited about something, we can’t reliably tell what caused the excitement.

In David McRaney’s brilliant article, he explains how we routinely misattribute physiological arousal and how the most common target of misattribution is other people. It’s part of why doing challenging or exciting activities with your partner can help strengthen your relationship, and why roller coasters and scary movies can make great first dates — when we reflect on the time, we often assume that it was the person we were with that helped get our heart racing rather than the excitement of the activity.

The misattribution of arousal

Our tendency to believe that the worst possible thing that can happen is the worst possible thing we know has happened — the name given by Nicolas Taleb in his book Antifragile. So, should we plan to stress test an investment portfolio, for example, we might test it against how it fared in the worst market crash. Or stress test a building for the biggest known earthquake, or a nuclear reactor for the biggest known tsunami. And yet, when each of these events happened, they were bigger than what had come before, and the same could happen again.

Lucretius was a Roman philosopher who wrote that:

The fool believes that the tallest mountain in the world will be equal to the tallest one he has observed.

Also known as The Lucretius underestimation.

See other sketchplanations on cognitive biases.

The Lucretius problem

We find it increasingly hard to empathise with the plight of larger numbers of people. So, a story about a family who needs help is much easier for us to relate to than, say, a story about 1000s of people displaced or dying. As the number increases the situation gets more abstract and we just can’t connect with it.

See the work of Paul Slovic, or this super article by Brian Resnick: A psychologist explains the limits of human compassion

Other uses of psychic numbing are for the general withdrawal of people and societies from potentially major catastrophes that seem unlikely to happen. Individually, it may also be reduced engagement with a past traumatic experience.

Psychic numbing

The name given to the observation that people who watch more TV seem to believe the world is a more dangerous place than people who watch less. Probably due to the prevalence of violence on TV and coverage of extreme events that are actually quite rare in most people’s lives. The term comes from cultivation theory — broadly the study of the long-term effects of television — which has a more general idea that people who watch a lot of TV tend to see reality more like the reality they see on TV.

Mean world syndrome

It turns out that, when considering how enjoyable, or indeed how painful something was, we don’t mentally do an average enjoyment check to decide if it was good or bad. Our minds are swayed by the peak and end levels. So, an experience that may have been dull for most of it, if it had a cracking bit in the middle and ended on a high will generally be remembered as more enjoyable than something that was decent all the way through but never reached those highs.

Here’s a classic study of this — on the pain side — from Dan Kahneman:

Kahneman and his associates showed, in 1993 that participants exposed to 30 seconds of 14 degree ice water (very cold!) rated the experience as more painful than participants exposed to 90 seconds of exposure to 60 seconds of 14 degree ice water plus 30 additional seconds of 15 degree ice water. In other words, participants found the 90 seconds of ice water exposure less painful than those exposed to 60 seconds of nearly equally cold water because the 90 seconds ended with exposure to a “warmer” stimulus. We will rate an experience as less painful, then, if it ends on a slightly less painful way. The “peak end” in this case was a one degree difference in water temperature.

Susan Krauss-Whitbourne PhD, Happiness: it’s all about the ending, Psychology Today.

The peak-end rule

The nocebo effect is how the expectation of pain or negative effects of a treatment increases the chances people will feel them — even if they're aren't any. So, tell someone that a treatment will hurt or have side-effects and they're now more likely to feel that pain or find something wrong.

Taken together with the placebo effect I guess we might do well, for normal affairs, to let people know that what they will get will probably help them and not raise unnecessary expectations that something might hurt.

Sketchplanations tagged with cognitive-bias