Dataset Information

Enhancing human learning via spaced repetition optimization.

ABSTRACT: Spaced repetition is a technique for efficient memorization which uses repeated review of content following a schedule determined by a spaced repetition algorithm to improve long-term retention. However, current spaced repetition algorithms are simple rule-based heuristics with a few hard-coded parameters. Here, we introduce a flexible representation of spaced repetition using the framework of marked temporal point processes and then address the design of spaced repetition algorithms with provable guarantees as an optimal control problem for stochastic differential equations with jumps. For two well-known human memory models, we show that, if the learner aims to maximize recall probability of the content to be learned subject to a cost on the reviewing frequency, the optimal reviewing schedule is given by the recall probability itself. As a result, we can then develop a simple, scalable online spaced repetition algorithm, MEMORIZE, to determine the optimal reviewing times. We perform a large-scale natural experiment using data from Duolingo, a popular language-learning online platform, and show that learners who follow a reviewing schedule determined by our algorithm memorize more effectively than learners who follow alternative schedules determined by several heuristics.

SUBMITTER: Tabibian B

PROVIDER: S-EPMC6410796 | biostudies-literature | 2019 Mar

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Enhancing human learning via spaced repetition optimization.

Tabibian Behzad B Upadhyay Utkarsh U De Abir A Zarezade Ali A Schölkopf Bernhard B Gomez-Rodriguez Manuel M

Proceedings of the National Academy of Sciences of the United States of America 20190122 10

Spaced repetition is a technique for efficient memorization which uses repeated review of content following a schedule determined by a spaced repetition algorithm to improve long-term retention. However, current spaced repetition algorithms are simple rule-based heuristics with a few hard-coded parameters. Here, we introduce a flexible representation of spaced repetition using the framework of marked temporal point processes and then address the design of spaced repetition algorithms with provab ...[more]

PMID: 30670661

Similar Datasets

Project description:BackgroundPharmacology is a cornerstone of medical education as it underlies safe prescribing practices. However, medical students have reported unease regarding their perceived proficiency in medical pharmacology. Despite the significant impetus to improve student outcomes, there is little analysis available of the techniques used by medical students to learn, retain and apply medical pharmacology knowledge.MethodsA mixed-methods, student-focused approach was conducted to design and evaluate specific resources developed to address gaps in pharmacology education. This methodology comprised an anonymised scoping survey, followed by semi-structured focus group interviews. We then developed a flashcard resource as an intervention to support long-term revision for academic and clinical success. This resource was released to a cohort of 100 graduate entry preclinical medical students who were invited at the end of year to evaluate the intervention via a subsequent anonymous survey.ResultsThe scoping survey received 103 complete responses. Surveys and focus group interviews revealed that only 50% of students engage in ongoing revision. Amongst our cohort, we identified that the evidence-based technique of spaced-repetition was particularly well regarded. Hence, we developed and evaluated a bespoke resource utilising Anki™, an open-source, spaced-repetition flashcard program. A total of 1208 flashcards spanning 156 distinct classes of drugs with supplementary summary tables, diagrams and explanatory video and summary guides were created. Designed as a strategic revision tool to reinforce learning, evaluation showed students greatly appreciated the "comprehensive" and "well formatted" Anki™ resource that supported existing teaching modalities, with a global rating of 3.8 out of 5.ConclusionsStrategic and personalised resources for medical pharmacology education that assist with in-semester revision and long-term retention are highly valued amongst students for examination preparation and preparedness for practice. Collectively, these results reflect a novel approach to identifying and addressing weaknesses in existing learning resources in a manner that is inclusive of, and acceptable to, medical students.

Project description:BackgroundPrevious research has strongly supported the utility of spaced learning in enhancing memory, but its effectiveness in complex surgical procedures has largely been unexplored. The main objective of this study was to evaluate whether, in comparison to concentrated learning, spaced learning improves the short-term acquisition and long-term retention of cosmetic suturing skills as outcomes of surgical resident training courses.MethodsThis randomized controlled trial was conducted from February 2023 to June 2023. Surgical residents were recruited from a teaching hospital in Guangzhou, China. The participants were randomly assigned at a 1:1 ratio to either the spaced training group (40 min of training followed by a 20-minute break) or the concentrated training group (3 h of continuous training), in which they received one-on-one training for cosmetic suturing skills. The short-term acquisition and long-term retention outcomes were evaluated by three independent raters using an objective scoring scale to assess the participants' cosmetic suturing skills before the training (pretraining test), within one hour after the training (posttraining test), and three months after the completion of the training (follow-up test). The score for each participant was calculated as the average of three independent scores.ResultsThe study included 23 surgical residents, 12 in the spaced training group and 11 in the concentrated training group. The pretraining test revealed no significant difference between the groups. However, in the post-training test, the spaced training group achieved a significantly higher total score than did the concentrated training group (74.06 ± 5.87 vs. 63.43 ± 10.73, p = 0.0070). Specifically, the suture technique scores were 28.46 ± 1.78 and 22.85 ± 3.75, respectively, which were significantly different (p = 0.0002). During the long-term follow-up test, the spaced training group consistently outperformed the concentrated training group by having significantly higher total (75.60 ± 4.78 vs. 60.68 ± 10.40, p = 0.0001), suture quality (32.26 ± 4.01 vs. 26.23 ± 4.16, p = 0.0019), suture technique (28.68 ± 2.63 vs. 22.18 ± 3.94, p = 0.0001), and suturing time scores (14.67 ± 1.15 vs. 12.27 ± 6.07, p = 0.0460).ConclusionsIncorporating the principles of spaced learning into the instructional process of obtaining cosmetic suture skills for surgical residents not only significantly enhances short-term skill improvement but also contributes to the long-term retention of training outcomes.

Dataset Information

Enhancing human learning via spaced repetition optimization.

Publications

Enhancing human learning via spaced repetition optimization.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets