Get it wrong first: the science of effort in learning Spanish

A few weeks after I started learning Spanish, I tried it out at a Mexican market near me. At the deli counter I asked for “Guacamole mexicano, grande, por favor.” The woman behind the counter asked if I spoke Spanish. “Estoy… aprendo?” I said. She corrected me: “Estoy aprendiendo.”

A week later I went to a Spanish meetup. Everyone at the table was well ahead of me, and I spent most of the night leaning on English and hand gestures. Words I was sure I knew wouldn’t come out in the middle of a conversation. When someone asked why I was studying Spanish, I said, “Me intereste las idiomas.” (It’s me interesan los idiomas.)

Every one of those mistakes got corrected, either by someone at the table or by me looking it up afterwards. And the corrections stuck. I still think of that deli counter every time I say estoy aprendiendo.

At the time it felt like failing. The research says otherwise.

Guessing wrong beats reading the answer

In 2009, Nate Kornell, Matthew Hays and Robert Bjork ran a simple experiment. One group was asked questions they couldn’t answer yet, made a guess, got it wrong, and then saw the right answer. Another group spent the same amount of time just studying the answers. On a later test, the people who had guessed wrong first remembered more.

This is now called the pretesting effect. One explanation is that searching for an answer pulls up everything related to it, so when the answer arrives it has something to attach to. Getting it wrong feels worse than reading the answer, but most of us don’t realize it’s helping us learn.

Speaking shows you the gaps

Reading and listening can make you feel fluent. That was me at the meetup: I knew plenty, right up until I had to say it out loud to a real person.

The linguist Merrill Swain argued that speaking makes you notice the gap between what you want to say and what you can actually say, which never happens when you’re just recognizing words on a screen. It’s also why tapping the right answer out of four can feel like progress without getting you any closer to holding a conversation.

Confidence matters too. Mistakes you were sure about are more likely to get fixed than lucky guesses, which researchers call the hypercorrection effect. If you’re certain “I’m downtown” is soy en el centro and someone tells you it’s estoy en el centro, the surprise makes you pay attention. If you only ever say what you’re sure of, you never get those moments.

It only works if you see the answer

None of this means struggling on your own is good for you. In these studies, the benefit came from trying and then seeing the right answer soon after. Struggling with no answer at the end is just frustrating, and that’s when people quit.

Bjork, one of the authors of that 2009 study, calls the useful kind of struggle a desirable difficulty. It makes learning feel slower in the moment, but it’s essential for getting things to stick.

How this shaped Poco a poco

I built Poco a poco around this idea. In each scene you talk with a character and have to come up with what to say yourself. There’s no multiple choice to tap. After every reply, you see how a native speaker would say it while your attempt is still fresh. When you’re stuck, a hint or the tutor is one tap away, so you’re never stuck for long.

The app also doesn’t grade you on being right. It scores your Effort: how far you stretch past the Spanish you’re comfortable with. Mistakes don’t count against you.

Next time you’re stuck

I’m glad I said estoy aprendo at that deli counter. So the next time you’re halfway through a sentence and the word won’t come, don’t take it as a sign you’re bad at Spanish. Make your best guess, say it out loud, and check the answer.