How does Gymglish use adaptive learning to design effective online courses?

We’re glad you asked. Adaptive learning is an innovative educational technology which provides personalized learning paths for students.

At Gymglish, adaptive learning has been an integral part of all of our courses since 2004. Today, Antoine Brenner, Gymglish co-founder, takes us behind the scenes of this innovative pedagogical approach. 

A popular EdTech buzzword

Most commonly associated with modern technology, adaptive learning has been around for decades, dating back to the 1970s. Designing a course today (especially language courses) without adapting the learning path is virtually impossible because each learner has their own specific needs. As a matter of fact, the oldest form of adaptive learning is a teacher naturally adjusting to each student’s level, expectations and areas to review.

However, given the major technological breakthroughs of the past two decades, the popularity of adaptive learning is booming – it has managed to change the landscape of learning, democratize education and now provides a never before seen form of adaptive learning.

Let’s outline the different types of adaptive learning:

• Single point adaptivity (or placement tests), which evaluates a student’s performance at one point in time in order to determine the level of instruction or material he receives from that point on. Once the student has completed a placement test, they will begin a course consisting of answers followed by corrections. This describes the most basic, traditional method.
• The learning model directly depends on the learner’s answering path, which helps determine a student’s proficiency using a predetermined path of questions: the learner is assessed throughout the course. Depending on their answers, the student is guided towards different types of course material. All possible paths must be anticipated beforehand by the course creator/educational specialist. However, the number of potential paths can be exponential.
• Continuous adaptive learning in real-time depending on the learner model (regardless of the chosen path, which is why this method is different from the previous one) described below.

Gymglish uses adaptive learning technology – it’s one of the four pillars of our pedagogical approach. Our artificial intelligence system uses our learner’s daily interactions called “traces” – such as answers to questions or participation dates – which helps foster long-term memorization. Every day, depending on each model and each student, our AI tool designs a customized learning experience for each learner.

Our approach is first and foremost based on the learning model which aims to take a snapshot of different concepts which have either already been learned, are to be reviewed or need to be learned. The more the learner completes daily lessons, the more they will instruct us on their needs, and the better our AI software will be at determining what the student already knows and what the student needs.

Research has consistently demonstrated the many benefits of adaptive learning (provided the students are consistent). But in fact, the learner almost never notices the type of adaptive learning technology used in their training. It is mostly the course creators (or educational specialists) who will see an impact. According to them, the linear approach is easy (but can become too easy, too rigid and restrictive, as each learner has their own way of memorizing the knowledge acquired and each learner has his or her specific strengths and weaknesses). When they talk about adaptive testing using the learner’s answers path, they report that an army of consultants and engineers have to help the course creators design said courses because of the exponential number of possible paths.

On the other hand, adaptive learning based on the learner model is goal-oriented; relevant content is picked from a database depending on the learner’s profile at a given time. The course creators don’t have to worry about paths, but can focus more on knowledge and ways to assess that knowledge, making the task much easier.

For us, the most effective way to learn is having access to a genuine adaptive learning approach, excluding placement tests followed by a linear approach. That being said, the adaptive testing method using the learner’s answering path requires a huge amount of work from the course creators. In our opinion, customizing a learning path based on the learner model that responds in real-time to each student’s performance and activity is the method most likely to be a sturdy foundation for 21st century adaptive learning. Actually, adaptive learning based on the learner model is at the very heart of our new course offers Gymglish Studio which provides AI technology, training, advice and support services to our partners so that they can create and edit their online course “the Gymglish way”.

Each day at Gymglish, we strive to offer our users the best experience possible to promote user engagement, memorization and participation. If you want to try some of our courses, click here.

Related articles:

• Spaced repetition: how to foster long-term memorization?
• Microlearning: How to foster user participation in 2022?
• Why do companies choose to set up Blended Learning programs?
• The many promises and challenges of e-learning from 2000 to 2022
• Learning and motivation: an essential tandem
• The benefits of being bilingual
• The benefits of learning a language for your brain