The Invisible Gorilla

Change blindness: we fail to compare what is there now with what was there before.

Most people believe they would notice an unexpected change. In practice, many do not.

My takeIn product terms: a banner nobody was looking for is a banner nobody sees. Test it instead of assuming it.

The illusion of memory is the gap between what we think we remember and what we actually remember.

Our intuitions about everyday memory failures can be fairly accurate. Our intuitions about the detail and persistence of our memories are often wrong.

Source-memory failure: we remember the information but lose track of where it came from.

In the authors' survey, 63% of Americans agreed they were more intelligent than the average person.

Research updateThe authors later repeated this with two nationally representative samples (Heck, Simons & Chabris, PLOS ONE, 2018) and found 65%.

The most dangerous overconfidence shows up while we are still unskilled, not once we are highly skilled.

People who succeed early can become especially prone to believing they are more capable than they are.

My takeWorth rereading after every good quarter.

The least competent can be the most overconfident, yet we keep using confidence as a proxy for ability.

Starting a war and then losing it is an extreme case of overconfidence in judgment.

An eyewitness's confidence is not, by itself, evidence that the identification is accurate.

Research updateLater research (Wixted & Wells, 2017) found that a witness's first confidence statement, under fair lineup conditions, does predict accuracy. Confidence grows less reliable the later it is measured.

In the late 1950s, Herbert Simon and Allen Newell predicted a computer would beat the world chess champion within 10 years. It took until 1997.

If experts can make badly mistaken predictions, the rest of us should be careful about overestimating what we know.

Investors given more information can grow more confident even when that information tells them nothing new.

Curse of knowledge: knowing something makes us overestimate how much other people know.

My takeThis is why onboarding written by the people who built a product so often confuses the people new to it.

We prefer experts who seem to know more, even when their actual knowledge does not justify that confidence.

When two things happen together, we assume one caused the other, even when nothing links them.

Arthritis patients are often sure the weather affects their pain. They remember the cold days that hurt and forget the rest, and tracking pain against the weather finds no link.

A vivid story convinces us more than statistics, because stories supply a cause and data does not.

The Mozart effect began as a small, short-lived boost on one spatial task in college students and grew into the claim that classical music makes babies smarter. Later studies found the effect small or absent.

Practising a brain game makes you better at that game, but the gains rarely carry over to general thinking ability.

Aerobic exercise has better evidence for protecting thinking in older adults than brain-training products do.