Adaptive learning in medical education: what works

Aerial view of a meandering river through forest

Two students open the same cardiology material. One already masters electrocardiogram and gets stuck on heart failure; the other is the opposite. A single course for both spends each one's time on what they already know. The promise of adaptive learning is to adjust the path to what each person needs.

NEXO is a daily medical diagnosis game: one clinical case a day, six progressive clues, and ICD-10 proximity feedback across five levels.

This guide explains what adaptive learning is, what types of adaptation exist, what reviews show about the effect on knowledge and skill, and how to apply the same logic to your own study.

Key points

  • Adaptive learning is adjusting content, order, difficulty or help to the performance of the learner.
  • Reviews find a clearer effect on skills than on factual knowledge.
  • The quality of studies is still variable, and results vary greatly from one study to another.
  • Adapting to what the person already knows can save time without losing learning.
  • The student also adapts their own study when using performance data to choose what to practice.

Summary

What adaptive learning is

Adjusting the path to the learner

An adaptive environment observes the performance of the learner and changes something in response: the next content, the difficulty, the amount of help or the timing of review. The idea is simple, but execution varies greatly. A systematic review described the adaptation process in five dimensions: method, objective, timing, factors considered and type of adjustment (Fontaine et al., 2019).

Types of adaptation

What is adapted Example Based on what
Content Skip what the person already masters Initial test or previous correct answers
Order and spacing Bring back what was wrong Correct answers and response time
Difficulty Harder cases for those who get more right Recent performance
Help More or less support during the task Signs of difficulty or choice of the learner

What reviews show

More effect on skill than on fact

The systematic review with meta-analysis of adaptive e-learning environments for health professionals and students gathered controlled studies with more than three thousand participants (Fontaine et al., 2019). The combined effect was 1.19 on skills, with a clearly positive confidence interval, and 0.70 on knowledge, with an interval that included zero (Fontaine et al., 2019). The authors point to a generally high risk of bias and large variation between studies, and suggest that adaptation helps more with skills than with factual knowledge (Fontaine et al., 2019).

Adapting can save time

A review on study time in online courses found a study in which adapting content to the student's prior knowledge saved time without significantly changing knowledge scores (Cook et al., 2010). In the same review, feedback and interactivity increased study time, but in many cases also improved learning (Cook et al., 2010).

Aerial view of a river delta in winter, with channels that split between snow-covered islands.
In the delta, water finds different paths to the same sea; in adaptive study, each student follows the route that leads to the goal. Credit: mattcatpurple, CC BY-SA 2.0 (license), via Wikimedia Commons (file page).

Examples of adaptation studied

Adaptive tutorials in imaging

In a crossover trial in Australia, medical students used adaptive tutorials on chest radiography and tomography, compared with a peer-reviewed web resource (Wong et al., 2015). In both arms of the study, the group with adaptive tutorials had higher scores, a gain concentrated in more advanced students and in questions about the tutorial topics (Wong et al., 2015).

Adaptive lessons in histology

In a study with dentistry students in Australia, adaptive histology lessons complemented the course, and students had higher scores on exam questions about the topics with adaptive lessons than on control questions (Yakin and Linden, 2021).

Perceptual learning with adaptive sequencing

One line of research combines perceptual learning, which trains pattern recognition with many examples, with an adaptive sequence that uses correctness and response time to decide the next item (Kellman and Krasne, 2018). The authors report large and lasting gains in reading signs and exams, such as electrocardiograms, with short training sessions (Kellman and Krasne, 2018).

When help already matches the need

In a trial with medical students in an emergency game, adaptive help did not differ from fixed help, because fixed help already matched the students' needs in 64.9% of scenarios (Faber et al., 2024). In exploratory analyses, help adjusted to need made students faster and reduced cognitive load (Faber et al., 2024). The message is that adapting is worth it when the learner's need actually varies.

Mastery before moving on

An old form of adaptation is mastery learning: the student only moves to the next objective after reaching a defined standard. In a meta-analysis on simulation, this model had a large effect on skills compared with no intervention, with an effect size of 1.29 (Cook et al., 2013). Compared with teaching without a mastery requirement, the benefit in skills was also large, with an effect of 1.17 in a few studies, but the model required more time (Cook et al., 2013).

Here adaptation is in time: everyone reaches the same standard, and each person takes the time they need.

Green lizard clinging to a tree trunk, with blurred foliage in the background.
Lizards adjust their color to the environment; adaptive study adjusts training to the learner's moment. Credit: Hiran Kavishka, CC BY-SA 4.0 (license), via Wikimedia Commons (file page).

How to adapt your own study

Use your data to choose what to train

The logic of adaptive learning also applies to those who build their own study plan. Instead of following the textbook in order, you look at your performance and decide the next step. The guide on tracking your progress shows which metrics help with this choice.

Simple adjustment rules

  • Got it right early and with ease: increase the difficulty or change area.
  • Got it right late: repeat the topic with different cases before moving on.
  • Missed close: compare neighboring diseases side by side.
  • Missed far: go back to the content before training more cases.
  • Area with few correct answers: increase its frequency in the coming weeks.

Adjust the help too

When a topic is new, it is worth studying with support: tables, summaries, the explanation right after each case. When it solidifies, remove the support gradually and train as you would on shift. This gradual withdrawal of help is a form of adaptation that you control yourself. The guide on differential diagnosis offers a way to train with clues that fits this adjustment.

Training choices in NEXO

In NEXO, the player has at hand choices that adjust training to their own moment. In Adrenaline, the dose is chosen before starting and defines the time and help: the low dose brings the area and ICD range help, the medium brings the area help, and the high leaves only the case clues. Raising or lowering the dose as the topic solidifies is the same gradual withdrawal of help described above.

On the home screen, the distribution of correct answers by clue and the specialties where you get the most right give the data to decide the next step. The Modes screen brings together the Archive, with previous daily cases, Multimedia and Continents, to choose the type of case each study phase calls for. Each guess shows the proximity in ICD-10 at five levels, which helps to know if the error was close or far. The details for accessing each mode are on the NEXO economy page.

If you research adaptive learning and want to study the topic with NEXO, check out NEXO Researcher. To bring NEXO to your medical school, student league or residency program, there is NEXO Closer. Doctors who review cases in their specialty take part through NEXO Curator, and those who create health content on social media can take part through NEXO Creators.

Frequently asked questions

What is adaptive learning?

It is a mode of teaching that changes content, order, difficulty or help according to the learner's performance. The goal is to spend each person's time on what they still need to learn.

Does adaptive learning work in medicine?

Reviews indicate a clearer effect on skills. In a meta-analysis with health professionals and students, the effect was 1.19 on skills and 0.70 on knowledge, the latter with a confidence interval that included zero (Fontaine et al., 2019).

Does adaptive learning save time?

It can save time. In a review of internet-based courses, adapting content to prior knowledge saved time without significantly changing grades (Cook et al., 2010).

How can you adapt your study without an adaptive platform?

Use your performance to decide the next step: increase the difficulty where you get it right early, repeat where you get it right late, and go back to the content where you get it wrong by a lot. Remove the help gradually as the topic solidifies.

How can you adjust your training in NEXO?

In Adrenaline, the player chooses the dose before starting, with different time and help options in each one. The distribution of correct answers by clue and the specialties where you get the most right help you choose what to train. To learn about the game, read what NEXO is.

Adapted study checklist

  1. Measure before studying: do cases or questions on the topic.
  2. Skip what you already master with ease.
  3. Repeat with different cases what you get right late.
  4. Go back to the content when you get it wrong by a lot.
  5. Increase the frequency of areas with fewer correct answers.
  6. Start with support on new topics.
  7. Remove the help gradually as the topic solidifies.
  8. Reassess every week and adjust your routine.

Want to choose your level of help in a case per day? Play the case of the day in NEXO.

References

  1. Fontaine G, Cossette S, Maheu-Cadotte MA, et al. Efficacy of adaptive e-learning for health professionals and students: a systematic review and meta-analysis. BMJ Open. 2019. https://pubmed.ncbi.nlm.nih.gov/31467045/
  2. Cook DA, Levinson AJ, Garside S. Time and learning efficiency in Internet-based learning: a systematic review and meta-analysis. Adv Health Sci Educ Theory Pract. 2010. https://pubmed.ncbi.nlm.nih.gov/20467807/
  3. Wong V, Smith AJ, Hawkins NJ, et al. Adaptive Tutorials Versus Web-Based Resources in Radiology: A Mixed Methods Comparison of Efficacy and Student Engagement. Acad Radiol. 2015. https://pubmed.ncbi.nlm.nih.gov/26292916/
  4. Yakin M, Linden K. Adaptive e-learning platforms can improve student performance and engagement in dental education. J Dent Educ. 2021. https://pubmed.ncbi.nlm.nih.gov/33825184/
  5. Kellman PJ, Krasne S. Accelerating expertise: Perceptual and adaptive learning technology in medical learning. Med Teach. 2018. https://pubmed.ncbi.nlm.nih.gov/30091650/
  6. Faber TJE, Dankbaar MEW, van den Broek WW, Bruinink LJ, Hogeveen M, van Merriënboer JJG. Effects of adaptive scaffolding on performance, cognitive load and engagement in game-based learning: a randomized controlled trial. BMC Med Educ. 2024. https://pubmed.ncbi.nlm.nih.gov/39210381/
  7. Cook DA, Brydges R, Zendejas B, Hamstra SJ, Hatala R. Mastery learning for health professionals using technology-enhanced simulation: a systematic review and meta-analysis. Acad Med. 2013. https://pubmed.ncbi.nlm.nih.gov/23807104/

Back to the blog