Adaptive learning

What does adaptive learning mean?

Adaptive learning is a form of tutoring that considers students’ personal preferences, experiences, and customs content [1]. They change given courses based on the learner’s characteristics and data acquired during the studying process. Navigation behavior, knowledge, and goals are utilized to create users’ profiles. Furthermore, simple data science techniques enable the implementation of personalized content at a suitable pace and sequence [2]. An intelligent system is part of the adaptive educational system, which uses artificial intelligence (AI) to provide further and more resounding support in the learning process [1]. This kind of platform should not only deliver portions of knowledge but also assess the progress of the learner [3].

Additionally, bare-book facts are not enough in the medical field to perform well as a physician. Medical doctors need practical skills and clinical experience to become self-aware and improving practitioners [2]. However, one of the most common types of e-learning is still non-adaptive, consisting of traditional video, graphic and audio materials [4]. 

‍

Why do we need adaptive learning?

Strictly for the research use, we can simplify the division of knowledge to the declarative and procedural. While declarative knowledge consists of verbalized concepts that need to be consciously learned and recalled, procedural knowledge focuses on how to do given action; in medicine, it can be a surgical procedure or blood withdrawal [5]. However, now we know that learning curriculum is not enough to be a successful practitioner [6]. Surprisingly, even after considerable and thorough education, students still struggle to manage mundane doctors’ tasks, like interpreting ECGs, radiology images, and clinical tests [7]. The difference between medical school training and the capacity to perform well during clinical work is significant. Kellman et al. [7] underlined what distinguishes professionals from medical students. They brightly choose relevant information and process them with a low attentional load. It can be done using procedural knowledge that does not require deliberate recalling from memory. If we want to help students in their educational path, learning systems should be customized to them personally. Adaptive learning allows presenting different starting points, tailoring lesson content to students’ potential, and scheduling spacing in the learning process. Additionally, an adaptive learning platform is a prominent tool for identifying weaknesses, which may influence the ability to state the correct diagnosis. Gathered data would indicate what the learner needs to work on and focus on diligently [8].

‍

Assessment is a key to continuous improvement 

Test-enhanced learning has proven to increase the long-term retention of studied subjects [9]. In 2012 Larsen and his team conducted research that answered the question of how studying and testing could have clinical applications [10].  In the study, forty-one medical students learned three topics differently. One group worked with a simulated patient (SP), one wrote tests, and the last one used review sheets. All of them took part in four sessions. After six months, each group took a simulated patient test and a written exam week later. Results showed that the group who trained with SP and one who learned by writing tests performed better than students who reviewed sheets.

Additionally, both of the methods increased retention of the learned information, and it benefited in applying the knowledge to clinical practice. Moreover, repeated knowledge retrieval helped students in problem-solving skills, self-monitoring, and controlling clinical decisions. Questioning enables students to effectively repeat content without a teacher and points out gaps and mistakes in the thinking process [11]. Receiving feedback helps develop new abstract skills, like communication, perspective-talking, self-assessment, and, most importantly, clinical decision-making together with planning [12]. Becoming self-aware of knowledge and abilities lets physicians reflect on feelings and behaviors and gain professional empathy, which is so needed in this occupation. Quirk [11] underlies that education aims to prepare a doctor for adapting, acting, and advancing during a life-long carrier.

Additionally, quizzes help to structure the learning process. They introduce interleaved and spaced studying [11]—spacing assists in optimizing learning and, more importantly, retention [7]. Practical adaptive assessment should include spaced rehearsal, immediate analytics, structured feedback, and repetition [11]. The objective of the evaluation should be highlighted and changed in most academic environments, focusing on gained competencies and ways of improving them, not mainly on incompetencies [13]. 

‍

Case-based learning

Physicians gain abilities in the years of clinical practice. Experts have this valued trait to notice meaningful data among all information. However, no expert is needed to evaluate how doctors obtain their skills through constant contact with a patient, numerous examples, and stories heard from more experienced colleagues. We can clearly see that patient cases constitute invaluable learning material for the trainees. The merge of theory with clinical practice secured its term, vertical integration [14]. Clinical data are outlined for students and backed up with a theory so that learners can investigate various medical issues [15]. Cen et al. investigated if case-based learning (CBL) influences the student's academic performance and case analysis.

Interestingly, the meta-analysis showed that CBL increases academic performance compared to traditional methods. What must be noted is that there was no similar difference when problem-based learning and simulated teaching were taken into account [15]. A further benefit was that students who underwent case-based training performed better in dealing with complex problems [16]. As Cen and co-authors report, case-based learning benefits students’ mental health, giving them satisfaction from learning and enthusiasm. It also changes trainees into more aware learners, increasing their self-study and problem-solving skills [15]. 

‍

Lately, an interesting project has been evaluated, the Human Dx platform, which consists of patient cases uploaded by health care professionals [17]. This database allows practicing stating diagnoses, thinking about differential ones, and exchanging experience with physicians in various medical fields worldwide. In contrast to common multiple choice questions at universities, a user needs to write his primary thought. Afterward, instant feedback is given, and one can read the discussion on the case. Taking advantage of such a massive data set, Huffman et al. analyzed the platform’s users' characteristics and how customed content may help develop clinical skills [3]. According to their results, we can distinguish four learners’ profiles, expert doctor, emergency medicine resident, internal medicine resident, and medical student. It is out of a discussion that each of them would benefit from tailored cases with adjusted difficulty and complexity. Intriguingly, high accuracy of diagnoses positively correlated with a shorter time spent on the case. Research illustrates and proposes amendments to the platform that would change it into customed website promoting the development of physicians' skills. 

‍

Even though faculties wish to provide their own material and learning platforms, Human Dx gives an example of the advantages of a partnership with the existing company. The vastness of data already available and new technology solutions should encourage this kind of cooperation. Merging possibilities that provide medical university and cases-based platforms gives hope for changing the educational path into real adaptive medical education and opportunity for high-quality studies [18]. 

‍

Increase in efficiency 

It is assumed that the usage of the tailored curriculum improves the outcomes of the learning process [2]. An Everylearner Everywhere platform provided a detailed report about the adaptive teaching methods [19]. They analyzed the survey about adaptive courseware and its implementation in institutions. 

An adaptive learning platform is transparent; students can easily follow the course syllabus, check what is expected of them, and what topics are covered in the curriculum. It enables learners to set their goals and expectations on what they want to achieve. 

The most prominent feature of adaptive courseware is quick and targeted feedback. It helps students with workload distribution and enables repetitions and knowledge retrieval [11,19]. Furthermore, real-time data analysis gave students accurate instruction and indicated a personalized learning path [19]. 

Secondly, students become active creators of their education; they can take control of studying. Additionally, they acquire self-awareness and the ability to control development outside of the university, as an active practitioner [11,19].

At the same time, faculty can administer trainees' progress and analyze it, providing personalized support when needed and planning amendments for the following semesters. Moreover, gathered data is a huge help in identifying gaps and inequity between students. Issues like poverty, racism, failure, or withdrawal can be captured early and investigated to provide the necessary assistance [20]. 

‍

Conclusions

Facing the need for modern medical education and the update of deprecated curriculums, and teaching methods, adaptive learning presents one of the possibilities. It offers indisputable advantages as an online form, flexible schedule, personalized approach, and much more. The growing amount of medical knowledge and fast pace of changes in the guidelines is an obstacle for old-fashioned presentations or videos. When we have tools like databases, artificial intelligence, and machine learning that can be used to optimize the learning process, thorough consideration should be given in their favor. Significantly, case-based learning correlates strictly with the platforms and is a method approved by students and teachers. 

‍

1. Brusilovsky P, Peylo C. 2003. Adaptive and intelligent web-based educational systems. Int J Artif Intell Educ. 13:159–172.
2. Fontaine G, Cossette S, Maheu-Cadotte MA, Mailhot T, Deschênes MF, Mathieu-Dupuis G, Côté J, Gagnon MP, Dubé V. Efficacy of adaptive e-learning for health professionals and students: a systematic review and meta-analysis. BMJ Open. 2019 Aug 28;9(8):e025252. doi: 10.1136/bmjopen-2018-025252. PMID: 31467045; PMCID: PMC6719835.
3. Huffman M, Gustafson S, Chatterjee S, Rabner M, Nundy S, Gerkovich MM, Wright SM. Bolstering diagnostic reasoning skills with adaptive learning. Med Teach. 2018 Aug;40(8):845-849 doi: 10.1080/0142159X.2018.1484561. Epub 2018 Aug 9. PMID: 30091646.
4. Cook DA, Levinson AJ, Garside S, et al.. Internet-Based learning in the health professions. JAMA 2008;300:1181–96. 10.1001/jama.300.10.1181
5. https://unt.univ-cotedazur.fr/uoh/learn_teach_FL/affiche_theorie.php?id_concept=90
6. Bereiter C, Scardamalia M. 1998. Rethinking learning In: Olson DR, Torrance N, editors. The handbook of education and human development. Malden (MA): Blackwell; p. 485–513. 
7. Kellman PJ, Krasne S. Accelerating expertise: Perceptual and adaptive learning technology in medical learning. Med Teach. 2018 Aug;40(8):797-802. doi: 10.1080/0142159X.2018.1484897. Epub 2018 Aug 9. PMID: 30091650; PMCID: PMC6584026. 
8. Huffman M, Gustafson S, Chatterjee S, Rabner M, Nundy S, Gerkovich MM, Wright SM. Bolstering diagnostic reasoning skills with adaptive learning. Med Teach. 2018 Aug;40(8):845-849. doi: 10.1080/0142159X.2018.1484561. Epub 2018 Aug 9. PMID: 30091646.
9. Larsen, D. P., Butler, A. C., & Roediger, H. L. (2008). Test-enhanced learning in medical education. Medical Education, 42(10), 959–966.
10. Larsen DP; Butler AC; Lawson AL; Roediger HL 3rd, Advances in health sciences education : theory and practice [Adv Health Sci Educ Theory Pract], ISSN: 1573-1677, 2013 Aug; Vol. 18 (3), pp. 409-25; Publisher: Springer Netherlands; PMID: 22618856
11. Quirk M, Chumley H. The adaptive medical curriculum: A model for continuous improvement. Med Teach. 2018 Aug;40(8):786-790. doi: 10.1080/0142159X.2018.1484896. Epub 2018 Jul 22. PMID: 30033792.
12. Quirk M. 2006. Intuition and metacognition in medical education: keys to developing expertise. New York (NY): Springer Publishing Company.
13. Schuwirth L, Ash J. 2013. Assessing tomorrow's learners: in competency-based education only a radically different holistic method of assessment will work. Six things we could forget. Med Teach. 35:555–559.
14. Eisenstein A, Vaisman L, Johnston-Cox H, Gallan A, Shaffer K, Vaughan D, O’Hara C, Joseph L. Integration of basic science and clinical medicine: the innovative approach of the cadaver biopsy project at the Boston University School of Medicine. Acad Med 2014; 89: 50-53.
15. Cen X-Y, Hua Y, Niu S, Yu T. Application of case-based learning in medical student education: a meta-analysis. European review for medical and pharmacological sciences . 2021;25(8):3173-3181. doi:10.26355/eurrev_202104_25726
16. Grauer GF, Forrester SD, Shuman C, Sanderson MW. Comparison of student performance after lecture-based and case-based/problem-based teaching in a large group. J Vet Med Educ 2008;35: 310-317.
17. https://www.humandx.org/ 
18. Tackett S, Wright SM, Quirk M. Adaptive medical education research. Med Teach. 2018 Aug;40(8):783-785. doi: 10.1080/0142159X.2018.1490705. Epub 2018 Jul 13. PMID: 30005580.
19. Neisler, J., & Means, B. (2021). Teaching practices of faculty adopting adaptive courseware. Digital Promise Global.
20. https://www.everylearnereverywhere.org/blog/disaggregating-learning-data-adaptive-learning/

‍