Self-regulated learning is not do-it-yourself learning
Self-regulated learning is a concept that has become firmly established in education in recent years. It appears in vision and mission statements, professional development courses, policy documents and teachers’ lounges. But what does it actually mean? And just as important: what does it not mean?
In an extensive conversation with educator Barend Last – primary school teacher, educationalist and author – one thing quickly became clear: much of what we think about self-regulated learning is not entirely correct.
Beyond the Hype: Why Adaptive Dialogue Systems Enhance L2 Learning
You’ve decided to practice a new language. Imagine you could do it anytime, anywhere, without waiting for a nice friend or a 24/7 tutor. Want to rehearse talking about your job or hobbies for the hundredth time? With no worries, no pressure, and a safe space to make mistakes? This is where spoken dialogue systems come in. But how can a bot really support language learning? Can an automated conversation offer the same benefits as talking to a person? Nothing replaces human connection. But access matters. Bots are always available, infinitely patient, and never tired of hearing the same sentence ten times in a row. They can adapt to each learner’s needs, making practice flexible, low-pressure, and personalized. To see if these benefits translate into real learning, Professor Bibauw and colleagues analyzed 17 experimental studies involving 803 learners, synthesizing years of research on conversational agents. Their recent meta-analysis published in Language Learning & Technology provides the strongest evidence to date on how much dialogue systems actually help people learn a language. So what does the evidence say? Do dialogue systems actually improve L2 proficiency? The short answer: Yes, they work Significant lasting learning Students who practice with a conversational agent showed significant improvement (overall effect size d = 0.59). From a cognitive interactionist viewpoint, this makes sense. These meaningful interactions with a bot create opportunities for input and output, noticing, negotiation, and feedback, all necessary ingredients for language acquisition (Gass & Mackey, 2015). Interestingly, beginner and low-intermediate students (A1-A2) benefit the most. As proficiency increases, effects decrease, suggesting that conversational agents are most powerful when students need repeated, low-anxiety, and structured communicative practice. Just as important, dialogue practice leads to long-term learning. Students not only showed immediate improvements in their L2 proficiency after conversations, but these gains remained significant when tested later. The learning that happens during these dialogues sticks. Another important question was asked: What makes some systems better than others? Design choices matter; not all bots are created equal. The meta-analysis highlights some benefits based on the interactional design: Guided, scripted dialogues led to the strongest gains, followed by goal-oriented dialogues. Why? Because structure encourages clearer communicative goals, predictable input, and better targeted feedback. Free chat may feel more “natural,” but it often lacks the scaffolding students need to actually improve. Instructional design matters just as much, if not more, than the latest technology. Corrective feedback makes a difference. Systems that offer corrective feedback outperformed those without it. Both implicit (recasts) and explicit forms of feedback improved learning, with a slight preference for explicit correction, aligned with decades of second language acquisition research (e.g., Nassaji & Kartchava, 2021). The takeaway is clear: conversational practice alone isn’t enough. Pedagogical feedback matters. Gamification boosts learning. Adding gaming elements showed a significantly stronger impact on L2 development than non-gamified ones, highlighting the importance of motivational design in dialogue systems. Rewards, challenges, and progress indicators increase motivation, sustain effort, and help students stay engaged while practicing the L2. Why does this matter? Dialogue systems are powerful tools for language learning, especially when combining strong instructional design with advanced NLP. This meta-analysis provides empirical support for the kind of adaptive, task-based, and feedback-rich conversational experiences that Linguineo builds. It reinforces several principles that match our philosophy: Dialogue systems are not just technology; they are effective when paired with thoughtful learning design, a principle at the heart of Linguineo. Original article Bibauw, S., Van den Noortgate, W., François, F., & Desmet, P. (2022). Dialogue systems for language learning: A meta-analysis. Language Learning & Technology, 26(1), 1–24. https://doi.org/10.64152/10125/73488 Other work cited Gass, S. M., & Mackey, A. (2015). Input, interaction, and output in second language acquisition. In B. Van Patten & J. Williams (Eds.), Theories in second language acquisition (pp. 194–220). Nassaji, H., & Kartchava, E. (Eds.). (2021). The Cambridge handbook of corrective feedback in second language learning and teaching. Cambridge University Press. Post photo by Shantanu Kumar on Unsplash
New Research Behind Real Adaptivity in Language Games
Imagine you’re playing a game while learning a language. You have your cute little owl (don’t worry, ours won’t threaten your streak), gathering intel, talking to characters, making choices, trying to stay fully immersed in this adventure when, without realizing it, the game quietly steps in and gives you a hand. Maybe it whispers the first letters you need. Perhaps it speaks slower? Or gives you an easier quest. Almost as if it knew you were about to hit a (virtual) wall. Too good to be true? (As good as sunshine on a Belgian winter day?) That’s the promise of adaptive learning. Not an empty one though, there’s a new empirical study that proves we can deliver on that promise (the adaptivity one, not the sunshine, yet). Researchers at KU Leuven set out to investigate whether Language Hero, our narrative-based game, can automatically assess students’ language performance in real time and adapt tasks to their level accordingly. And can it do that? [Spoiler] Yes it can. Why is this research important? From a theoretical standpoint, research in second language acquisition is clear: we learn to speak a language by using it in meaningful interaction, whether with another person or with spoken-dialogue systems. But for such systems to truly support learning, they need to adapt to each student’s needs and proficiency in real time (check Bibauw et al., 2022 to learn all about dialogue systems). Everyone promises these adaptive and personalized AI tools. But we need empirical evidence to show how such adaptation works and whether it improves language learning. This is precisely what KU Leuven researchers investigated in their recently published study. To examine how the built-in adaptivity in Language Hero could predict successful task completion, they analyzed students’ oral language using theoretically grounded measures (check out Koizumi & In’nami, (2024) for a deep dive into these measures). More specifically, Then they asked: Do these measures predict if a learner will succeed on the next task? They found out that: And why does this matter? Adaptability can support all learnersData-driven learner models like the one in Language Hero can improve micro-adaptability. It can offer better individualized support. For instance, for lower proficient students, the system can provide more detailed hints, adjust linguistic complexity, or present alternative tasks, all based on real-time indicators of students’ performance. Teachers get data they can useThese models expand pedagogical possibilities by providing interpretable linguistic data through dashboards and visualizations of students’ proficiency and progress. Teachers save time, choose what matters the most, and decide where to focus their attention. AI you can trustThis research offers a transparent, theory-informed learner model (as opposed to an opaque “black box” of off-the-shelf chatbots), that we hope can improve trust in AI-powered applications for language learning Spoken dialogue systems can do more than “talk back” …or provide speaking practice. In a way, they can tell when a learner is about to struggle before they do. And that’s when adaptivity kicks in. This study is a milestone for us. It shows that our evidence-based adaptivity is moving in the right direction. And we intend to keep building it, validating it, and sharing it with all our students. Original article Cornillie, F., Gijpen, J., Metwaly, S., Luypaert, S., & Van den Noortgate, W. (2025). Towards adaptive spoken dialogue systems for language learning: Predicting task completion from learning process data. CALICO Journal, 42(3). https://utppublishing.com/doi/10.3138/calico-2025-0035 Other Work Cited Bibauw, S., François, T., & Desmet, P. (2022). Dialogue Systems for language learning: Chatbots and beyond. In N. Ziegler & M. González-Lloret (Eds.), The Routledge handbook of second language acquisition and technology (pp. 121–135). Routledge. https://doi.org/10.4324/9781351117586 Koizumi, R., & In’nami, Y. (2024). Predicting functional adequacy from complexity, accuracy, and fluency of second-language picture-prompted speaking. System, 120, 103208. https://doi.org/10.1016/j.system.2023.103208
We tested our voicebot in Amersfoort – here’s what we learned from teenagers
At Linguineo, we have one mission: to make language learning more fun and more effective. Not just by reading or listening, but by doing. By speaking, to be exact. And that’s often where things get tricky in the classroom. Because let’s be honest: do you know that feeling too? You think, “Okay, I speak pretty decent German,” until you actually have to say something. Your phone’s dead, no data left, and your only option is to talk to a stranger… and suddenly, buying an old-school city map sounds like the better idea. 😬 That’s exactly the kind of situation we want to eliminate at Linguineo. So, together with publisher Noordhoff and the team behind the German curriculum, we headed to a school in Amersfoort. There, we had two classes of 13- to 15-year-olds from 2 HAVO/VWO and 3 HAVO/VWO try out our voicebot. We wanted to know: does it work in the classroom? Do they actually speak? And most importantly: do they dare to? Spoiler alert: they do. And how. 1. Talking to a voicebot is less scary than talking to a real person What we suspected turned out to be true: students find it way less intimidating to talk to a voicebot than to a teacher or classmate. “It’s nice to practice without the stress,” one of them told us. And she wasn’t the only one.By putting on a headset and practicing calmly with the bot, most of that speaking anxiety just disappears. No judgment. Just practice, at your own pace. More confidence, less pressure. 2. There’s simply not enough time in class to let everyone speak Creating a truly safe space to practice speaking in class is hard. You’ve got 25 students, limited time, and a jam-packed curriculum. Not everyone gets the chance to speak equally. And when they finally do, over 20 pairs of ears are listening in.Our voicebot changes that. Students practice one-on-one, using earphones or a headset. No one’s listening. No one interrupts mid-sentence. And they get immediate feedback, clear and tailored to their level.What’s more, our bot is highly adaptive: it adjusts to each student individually. Whether you’re a beginner or more advanced, you get the right feedback and support, exactly when you need it. The bot literally guides you step by step toward your speaking goal.This way, students can grow and practice independently, without pressure. And the teacher? They finally have some breathing room to focus on what AI can’t replace: coaching, deeper learning, and tailored support. 3. Practical, goal-based tasks We don’t believe in just drilling random sentences or memorizing word lists without context. Language only comes to life when you use it in real, meaningful situations. That’s why our voicebot works with realistic, goal-oriented speaking tasks: introducing yourself, asking for directions, giving your opinion, or describing something.During our test in Amersfoort, we saw students fully engage with these tasks. Sure, a few ran into small tech issues, like a glitchy mic, but that was the exception.What really stuck with us? They dared. They spoke. And they learned. What did we take away from this? We’re genuinely excited about the outcome of this test day. It confirmed what Linguineo is all about: more speaking confidence, more practice, less fear. Thanks to tech that doesn’t replace teachers, but supports them. And thanks to an approach that empowers rather than overwhelms. A big thank you to the school in Amersfoort, to Noordhoff Publishers, and of course to the students who participated so openly (and honestly). Want your students (or yourself!) to speak with more confidence? To get unstuck? Or just practice more with feedback that actually helps? Let us know! We’ll get the voicebot ready for you.
5 things we improved about Linguineo Pro that you need to know
Summer is coming to an end. *Cries in Belgian weather* While we’ve enjoyed the occasional sunny moments, we have been working *read: the whole year* behind the scenes to improve Linguineo Pro. We know, summer isn’t done yet, but we sure are with the update of Linguineo Pro. We have done a major update to Linguineo Pro, incorporating all the most important previous user feedback. Yay! We literally can’t wait to share it with you! So, keep on reading to find out what we improved.
A day in the life of a Natural Language Processing intern
In July, we warmly welcomed a new person to the Linguineo team. *Drum roll* None other than Kobe, who completed a 5-week internship in Linguineo’s small but cozy team. As his final week of his internship wraps up, we interviewed him to hear about his experience at Linguineo.
OKAN pupils learn Dutch with voicebot POL
If you move here from another country, the best thing you can do is of course learn the language. Easier said than done? Not with POL, our personalized voicebot tailored to OKAN students.
OKAN stands for ‘Onthaalklas voor anderstalige nieuwkomers’, or classes for non-native newcomers between ages 6 and 18 who have not yet mastered Dutch. The goal is to support them and make their learning experience more enjoyable. We partnered with D-Teach and KU Leuven’s Centrum voor Taal en Onderwijs to do that! Currently, we are in the testing phase of POL, and we would love to share you a little bit more about him.
Building speaking confidence in a foreign language
Learning a foreign language is challenging but rewarding. While learning, some of us may experience speaking anxiety. Overcoming this is not easy, but crucial to master a new language. In fact, according to scientific research from Defense Language Institute, it is one of the most important things to overcome to master a new language. Luis von Ahn, founder of our big brother Duolingo, says it in this video. Overcoming speaking anxiety as a major challenge Imagine yourself in a sunny village in the south of France, a remote but charming village. Everything is in French, no one speaks English. Fortunately, you have a good basic French knowledge. This immersion should in fact refine your French speaking skills. On your first day in the morning, you decide to go look for a local bakery. You forgot to charge your phone and therefore have to ask locals for directions first. There are many people outside, but you don’t approach them. You don’t dare to ask for directions in French. The threshold to speak in French is too high. You decide to look for a bakery yourself. After walking for a while, you come across a local bakery. Great! You would like some pastries, but again you can’t get over the threshold to carry out the conversation in French. With some pointing, you get rid of the uncomfortable situation. Although you supposedly know enough French, speaking it in real-life situations is very difficult for you. These situations highlight the importance of building up speaking confidence. Task-based language teaching To be better equipped to deal with situations like this and build up this necessary confidence, task-based language teaching (TBLT) has proven to be an effective method. Tasked-based language teaching (TBLT) is a methodology where the focus is on learning skills rather than language knowledge. This by doing practical tasks in the language one is learning. Exactly what we are up against in our French village situation. The language knowledge is there, but in practice the language skills are lacking, after which the fear of speaking emerges. One reference for this approach is the book ‘technology-mediated TBLT’ written by researchers Marta González-Lloret and Lourdes Ortega. It highlights even more benefits besides “doing it yourself”. A safe language learning environment for practice Many people are too self-conscious when they have to speak a foreign language. Especially because they are afraid to make mistakes and therefore be judged by others. The threshold to speak that new language is then often too high. As in the case of the French village. Thankfully, that fear of speaking can be overcome. One of the ways to overcome speaking anxiety is by creating a safe environment for language learning. In a digital learning environment with chatbots and voicebots, there is no fear of being judged by another person and we can focus on actively practicing the language. Contextualised environments with computer-based game characters seem to provide an ideal context when teaching a language. This fits perfectly into the task-based language teaching method. That is exactly what we do at Linguineo. Digital learning environment with conversational AI to actively practise a foreign language in real-life situations. Scene: asking for directions in French. Conversation for advanced learners with pronunciation feedback. Scene: Ordering at the bakery. Our conversational AI enables language learning through games, apps and learning environments such as Language Hero and Linguineo Pro to practice speaking skills. Considering the time constraints teachers experience when supervising students in language classes, this can provide additional useful speaking opportunities. In this way, students or even employees can develop their speaking skills and accuracy with the support and feedback of our system. Fear of speaking a foreign language is a difficult threshold to cross and yet it is essential for mastering a language. By identifying those problems and offering a solution, speaking anxiety will drastically reduce. Follow us on Facebook, Twitter or LinkedIn to get weekly updates on Conversational AI for language learning.
Tech Can’t Do It Alone
The term ‘disruptive innovation’ is everywhere. Although the term is overused – read this article in which Harvard professor Clayton Christensen who originally coined the term disruptive innovation explains why Uber is not a disruptor but Netflix is – we live in times in which startups and behemoths alike are actively looking for ways to shake up entire industries and they often succeed. As we use disruptive digital technologies, we sometimes blur the boundaries between our physical and virtual worlds. One example is getting a mortgage loan. Years ago, that required multiple trips to the bank and lots of paperwork, but now you can handle the entire process online from your couch. Many years ago, I was travelling the world as a ‘digital nomad’, a term for someone who works remotely while travelling. One friend called me a homeless freelance programmer instead, in an attempt to make me feel less hip. In the morning I was coding under a palm tree at the beach. In the afternoon – once the wind picked up – I was kitesurfing. Now, years later, my wife and I order groceries online. I still collect them with the car, but I expect they’ll be delivered soon enough. We also get dinner boxes with recipes delivered to our door, allowing us to eat vegetarian twice a week without any effort. Half the week I work remotely. We have huge electronic mailboxes in which shipping companies can pick up or drop off any package using a pin code without ever seeing us. All these conveniences help free up time to raise our twin boys, Michaël and Daniël. Thanks to technology we can pursue lifestyles that were not possible before. I love technology; I’m what you might call a gadget freak. My wife has said to me more than once, “Oh, you are not going to buy the latest version of ? I’m surprised.” I believe every technology is worth inventing and makes us better at solving problems. But what is lost by taking away all this human interaction? American novelist Jonathan Safran Foer writes that technology may be ‘diminishing us’: “Let’s assume, though, that we all have a set number of days to indent the world with our beliefs, to find and create the beauty that only a finite existence allows for, to wrestle with the question of purpose and wrestle with our answers. We often use technology to save time, but increasingly, it either takes the saved time along with it, or makes the saved time less present, intimate and rich. I worry that the closer the world gets to our fingertips, the further it gets from our hearts.” Some time ago – it was in the middle of my digital nomad years – I created some language learning apps. Originally, they were not intended for others to use, because I was travelling in South America and I created the apps to supplement my Spanish and Portuguese language learning. They covered all the basics – vocabulary, phrases, grammar, verb conjugations – but I never imagined they would reach over 2 million downloads in the App Store. Suddenly I had a small language learning business without knowing much about language learning. Sure, learning languages has been a constant in my life. I was born in Belgium near the French-Dutch language border, and as a Dutch-speaking teenager, I had to know French. I had to attend university in English; I started travelling after my studies, for which I learned some Spanish. Later, I ended up working in Brazil, where I had to learn some Portuguese. Some years ago, I was working with companies in Belarus and Ukraine and I had to learn some Russian. But I am not a polyglot; my track record and language skills are very common. I began to receive positive feedback on the apps, but I was also getting frequent emails that said, “Your app does not provide any guidance on learning this language. What do I need to do first?” I politely replied that, as the app description clearly mentions, it is designed to support people taking language classes, not people who are learning a language from scratch. When learning a language I always attended language classes or had a private teacher. French and English I learned at the Onze-Lieve-Vrouwecollege Halle; my first Spanish and Russian I learned at the CLT in Leuven. I continued learning Spanish at ECELA in South America, in Lima, Cusco and Buenos Aires. Portuguese I learned with a private teacher in Brazil.The app users writing in had a point, but their messages made me wonder whether they were expecting to learn a language just by using an app. Apps are excellent tools for augmenting learning, and they are often inexpensive to purchase. Lack of access to education is a serious driver of inequality in many countries, and it’s a noble goal to bring free or low-cost education to the world through a digital platform. I’ve tried many language learning apps myself. Duolingo and Busuu are two of them; I love the gamification and interface of Duolingo, and I really like the peer review of speech in Busuu. But when it comes to my memories of language learning, I remember the real life experiences more vividly and fondly.My high school French teacher made us think about much more than French alone. He facilitated discussions in class that made each of us think about politics and helped build our view of the world. I remember the Spanish lessons at CLT with Rita, our wise and funny teacher who lived for years in Barcelona. And I remember the Wednesday evenings after the class even more, as we found ourselves in Villa Ernesto, a bar here in Leuven, at 2am and still having to work the day after. I remember the thrill of total immersion courses at ECELA and making my first South American friends there. I remember attempting to speak Spanish in the Amazon region and getting compliments for trying,
Learn words! is better than ever – and we couldn’t have done it without you!
It has been an exciting journey to get where we are today. At Linguineo, our goal is to make learning a language as effective and easy as possible, so we launched our Learn Words! app in December 2016 as a supplement to Linguineo’s suite of products. Even though the launch was quite successful, there’s always room for improvement, so after the release of Learn words!, we asked more than 40 language teachers to give us feedback on the app. One of the first things they said was that they needed to be able to share word lists with their students. We agree that sharing lists is important, so that was one of the first things we changed in our upgrade. Sharing lists is now one of many new and useful features in the app. If you would like to see how to share a list, see a quick video tutorial here. We received many useful suggestions, and agreed that if three or more teachers made the same suggestion, we needed to make those suggestions part of the upgrade. The new and improved version of Learn Words! is now live and available at the app store. We think it is important that people can see how we are improving the app, so below is a list of the most common suggestions, and our solutions. 1. Users felt they had to pay for everything – this was never our intention! Challenge: Most features in Learn words! are completely free, but to pay for the running costs of our servers we do have to charge the heavy users. Unfortunately, this principle of “pay only if you use it a lot” was not properly reflected in the app. Beginner users were constantly confronted with the ‘unlock content’ dialog, although it was usually for the “Use for free” option. Despite this, users felt like we were asking them to pay for each feature and they were understandably getting frustrated. Christie Vanorsdale: “I think that the system of paying for each add on separately could be improved. Maybe just have everything for purchase as one ‘package’ upgrade as it seems that you need all of those functions to use the app anyway.” Solution: Now, when a user is able to access a free feature, this dialog is no longer displayed. We also increased the usage threshold that applies to heavy users by a factor of 4, so people can use the app for free quite a bit longer. In addition to reducing the number of times users encounter the ‘pay for content’ dialog, and raising the frequent user threshold, we have also bundled all permanent upgrades into one package (they were 3 separate upgrades before). 2. Difficulty creating word lists from images Challenge: Although we tried to make creating a word list from an image as easy as possible in the initial release, users who weren’t very tech-savvy found the process too complicated. Keon Esky: “As any other teacher would probably tell you, time is of the essence, so you may want to find a way to render the app a bit more shallow (as in user-friendly).” Deirdre Steenekamp: “I like the app but uploading a list is a bit cumbersome.” Solution: For each image scan, a user had to indicate whether to use the entire image or only part of it. With our upgrade, it is assumed that the entire image will be used – which will be sufficient in most cases. Users still have the option of cropping the image, but they don’t have to actively decide not to crop with every image upload. We also disabled the default offline image recognition. Offline recognition was only working well in 10% of the cases and although the app clearly indicated that “in general offline recognition produces low quality results” this was still confusing for our users. Many thought the text could not be recognized at all, but server recognition would have worked in these cases. Automatic recognition mode is a brand new feature! Before, users had to manually process the recognized text. Although this was easy to do since there were already some context-aware macros available that could remove all duplicates or punctuation with one touch, the process was still confusing. In the new automatic mode the app analyses the recognized text and decides itself which of these macros to execute to get a word list. Now, users can skip this “text processing screen” altogether. Lastly, the app was not retrieving translations automatically once the recognized text was converted to a list of words even though every user wants these translations. The app will now automatically retrieve the translations. The result is that a user now really only needs to select an image, press “Recognize text” and then decide to add the entire word list with translations or not, which is a much simpler flow. The old recognize words flow – 9 screens The new recognize words flow – 4 screens 3. Too many options Challenge: Before the upgrade, users had 7 choices when adding words. Ana Pecanha: “As a student, I think there are lot of options and sometimes having so many options at one’s disposal can make one lose the point.” Solution: We have reduced this to 5 clearer options, postponing certain decisions to later in the flow. Previously, users had to choose Learn, Exercise or Play, when doing an exercise. Depending on their selection, 10 more options appeared for the exercise type. Now, users only press the Exercise button, and then select one of 5 options. The additional configuration options are still there, but there are no forced selections, and users can begin their exercises right away, using the defaults. 4. Labeling lists was difficult Challenge: A remark many teachers made was that they wanted to give a label to their word list, but they could not see how to do it. Although this was already possible, the feature was hard to identify. Graziani Correa: “It is not clear where I