Learning & thinking · a life skill
Deliberate Bias Spotting
a spotter's month, with someone telling you when you're wrong
Reading about biases is easy, and it mostly teaches you to see them in other people. Spotting them as they happen takes practice with feedback, the way you'd learn birds: a few field marks, short drills that tell you when you guessed wrong, and real sightings you check later. For two to four weeks you work on six biases, one sighting a day, and spend most of your drill time on the one you miss most.
- Moderate to start
- High impact
- Learn it in 2-4 weeks
- Six biases, one sighting a day
Base rates got four misses in the month, more than any other, so next week's drills go there. The dots come from real sightings checked later; the drills come from the cards.
Two to four weeks
Try it now.
Ten minutes a day: five drill cards, one real sighting, and a check on an old one. Fifteen minutes on Fridays.
-
Learn six field marks
Six biases, each with a one-line tell you could spot in a conversation. Six is enough to start, and few enough to keep in your head.
-
Drill with feedback
Five short cases a day. Name the bias before you see the answer, then read why. Wrong answers are the useful ones.
The drill below has twelve cards, two for each bias. -
Log one real sighting
Once a day, catch one in a meeting, the news or your own head. Write the moment, the bias, and what you'd expect to happen if you're right.
-
Check it later
A day or a week on, look again. Did the anchored price stick? Did the person you blamed have a reason? Mark the dot filled or hollow.
-
On Friday, drill your weakest
Count the hollow dots for each bias. The one with the most gets most of next week's drill time.
Spot your own first. If every sighting is someone else's, that's the blind spot at work. Aim for at least one a week from your own decisions.
Today's drill
Name the bias, then read why.
Twelve short cases, two for each of the six biases, in a new order each time. Pick the one you think is at work. You'll see straight away whether you were right, and the tally shows which one to drill next.
Case
Your tally
Today's sighting
The bias
Check it
Nothing you type is saved or sent anywhere, and the tally resets when you leave. Copy the sighting into your notes or calendar so you remember to check it.
Why drill rather than read
One training session cut bias for months, and practice with feedback cut it more than a video.
Carey Morewedge and colleagues gave people a single training session on three biases, in two experiments with six biases between them. Half played a computer game that had them make judgments, told them how biased they had been, and gave them more practice. The other half watched an instructional video.
Bias was measured before training, right after it, and again online two or three months later. Both kinds of training reduced bias straight away, and the reductions were still there at the later test. The game did better than the video.
The gains also showed up on problems set in other contexts and in formats the training hadn't taught, which is what you want from practice: not just recognising the examples you studied.
The honest limits: bias was measured with test questions, not real decisions. In the second experiment the game's edge over the video was only marginal by the later test, and for anchoring the two did equally well. The authors say more work is needed to know exactly why the game did better.
Experiment 1
Bias blind spot, confirmation bias, blaming the person (fundamental attribution error)
Experiment 2
Anchoring, base rates (representativeness), assuming others think like you (social projection)
The idea
Why it works.
Knowing a bias's name doesn't mean you'll catch it. You catch it when a situation looks like one you've practised on, and practice only helps if something tells you when you got it wrong. The drill does that in seconds. The world does it more slowly, which is what the later check is for.
Writing a prediction keeps the sightings honest. "That's anchoring" costs nothing to say. "So the final price will land near the first offer" can turn out wrong, and the hollow dots are how you learn which tells you're misreading.
When it goes wrong
Habits that help.
- Guess before you look. Reading the answer first feels like learning and isn't. Commit to a name, even a wrong one.
- Keep the list short. Six biases you can spot beat sixty you can define. Add one only when the first six feel easy.
- Don't use it as a weapon. "That's just confirmation bias" ends conversations. Use the names on your own thinking first, and on others' only when asked.
- Count the hollow dots. A month of only filled dots usually means the predictions were too safe. Make them specific enough to be wrong.
Practise this in HourLife.
In Decisions: Bias check. Write down a decision, weigh the options, mark the one you chose, and come back later to record how it turned out.
