Hometown U: The psychology of irrational choices

Have you ever been at a party and made a simple statement you regretted? Maybe a remark you thought was a no-brainer, like “The Seattle Seahawks are a great team” or “Life should be longer."

Then your conversation partner looked at you like you were from Mars and proceeded to tell you all the ways he or she disagreed with you.

Yasuhiro Ozuru, a professor of cognitive psychology at UAA, has documented a strong human inclination to quickly disagree with assertions like these. The implications are significant for everything from useful political discussion to teaching and learning to maintaining an open mind.

Ozuru grew curious about this because while doing post-doctoral work in psychology at the University of Memphis, two advisers responded so differently to his work. One would say, “I disagree.” The other would say, “I don’t understand.” Why the difference, he wondered.

That was back in 2007. A few years later he accepted a teaching position in the psychology department at UAA and began dreaming up experiments to figure this out.

First, he devised a study inviting 32 undergraduates to respond to simple assertions. Some were obviously true, like “Cars are useful.” Some were confusing, like “History needs light.” Others made no sense at all, like “One plus one should be three.” The undergrads were invited to say they agreed, disagreed, understood or didn’t understand the sentence.

The obviously true statements yielded expected responses; most listeners quickly agreed. But confusing or nonsensical statements delivered a surprise. People flat-out disagreed with the statements.

Take the sentence “Fish are mechanical.” Only 30 percent said either they did or did not understand the statement; 11 percent agreed with it and 59 percent disagreed with it.

“Even when a sentence is very difficult to understand, they still disagree,” Ozuru said. “This is very, very weird.”

In a second study, he decided to measure levels of agreement and levels of understanding.

This time, he had 42 undergrads review the same 126 sentences, but he allowed them a neutral answer. They could agree or disagree, but they could also say they weren’t sure. They had the same option with comprehension: They could say they understood or did not understand, or they weren’t sure.

Collectively, the responders said they didn’t understand 28 percent of the 126 statements. They also said they disagreed with 70 percent of the statements they didn’t understand. “This is not rational,” Ozuru said.

Again, why would people disagree with something they don’t understand? Since 1644, rationalists like Descartes have been telling us that people first understand something and then agree or disagree with it.

Ozuru’s research suggests a different model: We agree or disagree first, with little or no understanding. It happens automatically as we process the sentence.

Take this example: “Grizzly bears eat salmon.” This makes sense to us based on what we know about grizzly bears. It confirms our expectations about them.

But if you said, “A grizzly bear is a good pet,” that does not confirm what we already know about them. Of course, we quickly disagree.

But what if you said, “A grizzly bear is a school”? Or “A tree needs paper”? These make no sense, but based on what we already know about a grizzly bear or what we know about a tree, the sentence violates our expectations of that subject. We also can’t integrate the subject with the rest of the sentence. So we disagree outright with the statement.

“Most people think you can’t evaluate (agree or disagree) with a sentence unless you understand it,” Ozuru said. “We propose that evaluation happens earlier than comprehension because evaluation is based on the violation of your expectations.”

Here’s more evidence. Ozuru measured response times. People hit the “disagree” button much faster than they hit the “I don’t understand” button.

And only 17 percent made the most rational choice: When they indicated that they did not understand a sentence, they also said they weren’t sure if they agreed with it. And they took the longest time of all to make their choice.

“For people to be rational, they have to really think,” Ozuru said. “And that takes time.”

So is this bad? Ozuru has a couple of thoughts on that. It would be worse, he says, if we blindly agreed with statements we didn’t understand.

On the other hand, automatically disagreeing with things we don’t comprehend is no way to cultivate an open mind.

Ozuru has more studies planned, including reverse engineering the human decision-making process with several computer science colleagues.

“Think about it. We are like a computer,” Ozuru said. “Given that we have this (decision-making) function, what is our underlying software and how does it work?”

Kathleen McCoy works at UAA, where she highlights campus life through social and online media.

Leave a Reply