Forgetting curve: how to delay the decline in retention using cognitive science?

In a professional context where the continuous acquisition of skills has become crucial, online training has emerged as an essential solution. However, a major challenge persists: how to ensure that the knowledge acquired during training courses is retained by learners on a sustainable basis? Advances in cognitive neuroscience are shedding light on memory mechanisms and paving the way for innovative educational solutions to optimize learning retention. Let's explore the phenomenon of the forgetting curve together and discover how artificial intelligence technologies can help us overcome it.

The forgetting curve: a well-documented neurobiological phenomenon

The forgetting curve, conceptualized by Hermann Ebbinghaus in 1885 in his book “Memory: A Contribution to Experimental Psychology,” illustrates the alarming speed with which we forget newly acquired information. Studies show that without active review, we lose up to 75% of new information in just 48 hours, and up to 90% after a week.

This phenomenon is explained by the neurobiological mechanisms of memory. During initial learning, our brain creates synaptic connections between neurons, forming what is called the “memory trace.” Without reactivation, these connections gradually weaken due to brain neuroplasticity - the brain's ability to constantly remodel itself. Neuroscience has shown that the consolidation of long-term memory requires the repetition and reactivation of these neural circuits.

The work of Eric Kandel, awarded the Nobel Prize in Medicine in 2000 and detailed in “In Search of Memory: The Emergence of a New Science of Mind” (2007), revolutionized our understanding of the molecular mechanisms of memory. His research shows that the practice of active recovery triggers specific molecular cascades in neurons, activating genes that sustainably strengthen synaptic connections through the process of long-term potentiation.

The limits of traditional e-learning in the face of oblivion

Cognitive science research has demonstrated the superiority of active recovery for long-term retention. In their landmark study published in Science in 2008, Karpicke and Roediger showed that students who regularly practiced active retrieval retained 80% of the information after one week, compared to only 36% for those who just reread their courses. A more recent meta-analysis by Dunlosky et al. (2013) in Psychological Science in the Public Interest confirms that retrieval practice is one of the most effective learning strategies.

Unfortunately, traditional e-learning is struggling to incorporate these scientific principles into its design. The majority of online courses are based on a passive format: the learner watches videos, reads content, and then answers a few basic questions every 15-20 minutes. This approach has several major flaws:

• The absence of moments dedicated to the active recovery of knowledge throughout the learning journey
• Overly superficial quizzes that test recognition rather than active recall
• A lack of contextualization that limits the transfer of learning to real professional situations
• The absence of a spaced revision system, which is nevertheless crucial in countering the forgetting curve

A study by the University of Maastricht (van Merriënboer et al., 2018) shows that this type of training leads to very low retention: less than 20% of the content is retained after one month, making investment in training unprofitable for companies.

Didask innovation: an approach based on cognitive science

This is where Didask's educational artificial intelligence comes in. For training designers, our technology automatically generates learning paths that strategically integrate two types of exercises: flashcards spaced out over time to regularly reactivate the knowledge acquired, and contextualized practical cases that allow this knowledge to be applied in realistic professional situations.
‍

(We are so sorry this video is only in french, we are working on it!)

On the learners' side, our platform goes beyond a simple quiz by offering complex and authentic scenarios. Whether formulating a business strategy or solving a delicate customer case, the learner is constantly engaged in an active recovery process. We are also the first on the market to have developed an AI coach who supports the learner in solving complex tasks (formulating a product, closing a commercial negotiation, writing a cover letter), analyzes his production and gives him precise feedback. Enough to intensely re-mobilize the learning seen during training and effectively counter the forgetting curve!

Based on the scientific principles of memory consolidation and active learning, Didask transforms digital training into a truly effective experience for the long-term retention of professional skills.
‍