Previous: Magic Isn't Magic: It's Psychology
Next: The Dark Side of Disgust



View count:1,456
Last sync:2020-12-03 20:30
You've probably been told at some point or another to "trust your gut", but is that actually good advice?

Hosted by: Anthony Brown
Support SciShow by becoming a patron on Patreon:

SciShow has a spinoff podcast! It's called SciShow Tangents. Check it out at
Huge thanks go to the following Patreon supporters for helping us keep SciShow free for everyone forever:

Jb Taishoff, Bd_Tmprd, Harrison Mills, Jeffrey Mckishen, James Knight, Christoph Schwanke, Jacob, Matt Curls, Sam Buck, Christopher R Boucher, Eric Jensen, Lehel Kovacs, Adam Brainard, Greg, Ash, Sam Lutfi, Piya Shedden, KatieMarie Magnone, Scott Satovsky Jr, Charles Southerland, charles george, Alex Hackman, Chris Peters, Kevin Bealer
Looking for SciShow elsewhere on the internet?
[♪ INTRO].

You’re taking a multiple choice test, and immediately you know -- the answer is B! No, wait.

D. It’s totally D, right? Should you change your answer, or go with your first instinct?

Even teachers and professors are likely to advise students to trust their gut and stick with their first answer. And though you probably shouldn't change every answer, data suggests most changes actually help us. We’re not saying you should never go with your gut.

Our first instinct is definitely right sometimes. But our memory can trick us into putting a little too much stock in our first answer. Back in 2005, some researchers published a series of studies in the Journal of Personality and Social Psychology to test this idea.

They started by looking at multiple choice exams from intro psych classes. They had a team scour the exams for changed answers -- eraser marks and the like -- and they found over 3000 changes in more than 1500 tests. They noted which of these were a correction from wrong to right, from right to wrong, or from wrong to wrong.

What they found was that over half of changes were from wrong to right. Only 25% were changes to the wrong answer, and the rest were from wrong to wrong. They also gave a smaller group of about 50 students a survey asking them to predict how many wrong-to-right and right-to-wrong changes there would be.

Students' predictions were about split. 42% thought there would be the most changes from right to wrong -- meaning, switching your answer is a bad idea. Another 33% thought it would help, with the rest on the fence. But the data suggested that switching did help more often than not.

Psychologists call this the first instinct fallacy -- the belief that sticking to your first gut instinct will pay off. So why are our impressions so wrong? It has to do with how we react emotionally to mistakes, and how we remember them.

Those same researchers asked a group of 27 students to report what their first instinct was on any question where they weren't sure of the answer. Then, the researchers gave them feedback on which answers were right and which were wrong. And yes, when those students stuck with their first instinct, they were more likely to be wrong than when they switched.

Then, a few weeks later, the students were invited back to test their memory. All of them remembered more switches to a wrong answer than they had actually made -- and fewer times that they stuck with the wrong answer. It feels worse to take an extra action, like changing your answer, and still get it wrong.

That difference in our emotions seems to affect how we remember things. When we actually wind up taking the test, we want to avoid that bad feeling again. This is an example of the availability heuristic, which is when you use whatever is easily available in your mind to guide your reasoning.

It's the same reason that people are often more worried about plane crashes than car accidents. Plane crashes are usually scary events that get covered in the news. Car accidents are far more common, but rarely make headlines.

The right-to-wrong answers stick in our minds a bit like a plane crash does. We regret them more, but they're far less common. Because they feel worse, they stick around in our minds more -- so we think they happen more than they really do.

So if we know our memory is tricking us into thinking the more regrettable outcome is more likely, how can we adjust? First, it’s important to note that staying put isn’t terrible -- it’s just not as accurate as we think it will be. In a smaller study, published in Metacognition and Learning in 2016, students kept track of when they considered switching answers but didn't.

It turned out they had the right answer 59% of the time. So going with their guts panned out just under two thirds of the time. Another thing to do is to pay attention to why you want to switch.

If you find you want to change your answer because you misread a question or found a clue in a later question, that may well mean that switching will net you a correct answer. But if your first answer was just a guess, and now you think you've got a better guess, that's less likely to be a reliable switch. Teachers can help out, too.

For example, a pretest that gives you positive feedback when you switch your answer can help -- a little. But mostly on easy questions that you might have rushed through. One study from 2003 designed a test that gave students a chance to go back and check for any errors.

And having a brief chance to review helped students catch some mistakes and avoid that misleading gut instinct. In the end, there’s no one rule to follow if you’re trying to figure out whether to go with your gut on a test. It depends on how well you know the material, which is the point of a test, after all.

And though maybe sometimes going with your gut can be good, it almost never hurts to think things through and double-check when you can -- on tests, and in life in general. Hey, you know who’s great at thinking things through? Our patrons.

At least when it comes to helping us make free videos about psychology -- and space and everything else -- for everyone to enjoy. You guys are awesome. Seriously.

If you want to help out too, check out [♪ OUTRO].