Motivation in Many Languages: How Duolingo Supports Competence for Language-Learners

A little over a year ago, I found myself getting hooked on Duolingo, an online language learning program. I’ve always really liked languages–I took Spanish for many, many years and was once fluent, and dabbled in French in college. A few years ago I took a work trip to Germany and hated that I wasn’t able to hold more than a very basic conversation in the language, so I decided to use Duolingo to pick up some German.

I faithfully worked through the modules for several months and finally finished them and “graduated” from the German course. At the end, I was offered the chance to take an online college placement test, which of course I did. The result? I was apparently ready for third-semester college German!

At the same time, a couple of my friends were also indulging their own Duolingo additions. Granted, I am nerdy and associate with other nerds, but it’s nonetheless worth asking: How does Duolingo create such a sticky experience? The solution lies within self-determination theory.

Specifically, Duolingo does an exceptional job supporting user competence. Competence is a fundamental human need based on the idea that human beings are growth-oriented. We’re built to want to learn and grow. People are motivated when they can exercise their skills to accomplish things.

The folks at Duolingo have embedded several tactics that support a sense of competence into their user experience:

• Feedback at multiple levels, both proximal and distal
• Sequencing of tasks along a difficulty continuum
• Opportunities to bypass easier tasks if skills permit

Feedback

Feedback supports competence by letting people know when they’re successful, and offering information that can be used to calibrate performance when they’re not. In Duolingo, the user receives a lot of feedback as he goes. At the proximal level, corresponding to the specific momentary performance, the user immediately receives recognition if an answer was correct, or the correct answer if not. In the case that you make a small error (like forgetting to capitalize nouns in German), a pop-up prompts you to remember next time.

You also receive feedback at the lesson level. For each lesson, the user has four hearts which can be lost by providing a wrong answer. If you run out of hearts before completing the lesson, you must restart the lesson. You can also see progress through the questions in the lesson using the progress bar at the upper corner of the screen.

Finally, Duolingo offers feedback at the distal, or overall, level. This feedback helps users see their overall progress through the program. Users gain points as they complete lessons and “level up” in experience as they accumulate points. Users also accumulate “lingots” as they advance through the program, and cash these in for additional lessons or hearts.

At every point in the process, Duolingo offers users feedback on their past and current performance on multiple levels. This is a highly effective strategy to support user competence.

Task Sequencing

Task sequencing refers to the idea that you can put activities in an order from easiest to most difficult. As people progress through the tasks, the level of challenge grows. If the tasks earlier in the sequence are teaching people skills they need to succeed at later tasks, then this helps support a sense of competence and engage people at moving on through the sequence.

In Duolingo, the lessons within each language are ordered from the most basic to the most advanced. Each language’s lessons start with simple greetings, basic nouns like “the man” and “the woman,” and the type of vocabulary you might want to prepare for a vacation (“ein Bier, bitte”). As you successfully clear one lesson, access to the next few in the sequence is provided. You can’t skip the easier lessons to get to the harder ones (unless you pass a level exam, see below). What makes this fun instead of frustrating is how the lessons build on each other. Each more difficult lesson pulls in vocabulary and grammar from previous lessons. I’ve also noticed that when I go back to an earlier lesson to refresh my knowledge, the vocabulary from subsequent lessons sometimes sneaks back in. Basically, my easy lessons are harder because they know I’ve got more advanced knowledge, which is great because it keeps me interested.

Task Bypass

Part of supporting competence is not offering people tasks that are so easy that they provide no challenge at all. Super-easy tasks may give people a chance to be successful, but they can also quickly become boring, especially for people with a strong mastery orientation.

In Duolingo, people who have some experience with a language are not forced to start at the bottom to participate in lessons. Each set of lessons is cordoned off into a group roughly corresponding to level of difficulty. By completing a summary lesson that can be accessed at the bottom of each group, the user can bypass the easier lessons and gain access to the lessons at the next set of difficulty. This functionality ensures that advanced users aren’t bored out of their minds by being forced to practice simple greetings and food vocabulary.

But does it work?

Of course, just being engaging is not really enough. The cool thing about Duolingo is that they are also investing in the research to show that their approach actually works. In addition to having Duolingo users take online language proficiency tests, they’ve commissioned a few more academic studies to see how well Duolingo prepares people for university entrance exams in English and how Duolingo stacks up to university-based coursework.

The results are promising. One recent study showed that English proficiency as measured by Duolingo correlated significantly with TOEFL scores, suggesting that Duolingo is a useful tool to prepare for English language testing. Another study found that spending 34 hours of using Duolingo provided the equivalent language learning to a full university semester. I thought back to when I took intro French in college:

• 5 days of classes per week at 1 hour each = 5 hours
• + 2 hours language lab each week
• + homework time each week, call it 2 hours
• = at least 9 hours per week x 15 weeks in an average university semester
• = 135 hours of work and class attendance in a full university semester language course

Basically, you can get basic proficiency in a language using Duolingo in a quarter of the time it would take you at college. That’s awesome.

Having learned languages in multiple formats, I wouldn’t argue that Duolingo should replace live instruction. The big thing is lacks is the opportunity for conversation with other learners or speakers. But it provides a strong foundation in grammar and vocabulary that can equip a language learner for success in conversational attempts. It’s also great to refresh your skills in a language you’ve studied but perhaps let lapse. If you’re into language learning, add Duolingo to your toolkit. And even if you’re not, consider how Duolingo’s approach to supporting user competence might translate to your product.