About
The College of Science, Engineering and Technology offers degree programs that prepare students for high-demand professions in science, technology, engineering and math (STEM) fields. With an emphasis on Grand Canyon University’s Christian worldview, our college believes in instilling social awareness, responsibility, ethical character and compassion. Our blog, BrainSTEM, focuses on topics related to science, engineering and technology, with engaging contributions from students, staff and faculty. On the blog, you can find helpful resources relating to STEM fields and learn more about current events occurring globally, locally and within GCU. We hope to provide our readers with information that helps them learn about the necessary knowledge, skills and mental disciplines to succeed in today’s job market.
Let's get started on your degree

* Do you have a high school, college or university credits from outside the U.S.?
* Are you a U.S. Citizen?
* Are you a licensed, registered nurse in the U.S.?
(example: 777-777-7777)
Browse

* Required field

** Required field if international

Request More Information

Online Learning and Diverse Learning Styles: Part 2 – An Adaptive Algorithm

0
Young woman working on laptop

By Isac Artzi, PhD
Faculty, College of Science, Engineering and Technology

The challenge presented in part 1 of this blog can be addressed by an algorithm I developed as part of my ongoing research on learning styles. It is illustrated in the flowchart below.

A particular learner is measured on n different learning characteristics, p1, p2 … pn. For illustration purposes, the profile used in the flowchart is a learner whose profile consists of a visual ability rated at 2, audio ability of 8, expressive ability of 6 and technical ability of 9.

This profile can be initially stored in the database by the learner or generated arbitrarily. A proof by induction (outside the scope of this blog) demonstrates that even an arbitrary initial assessment of the learning styles of a learner quickly (after a few iterations) modifies itself into the actual mix of styles of that learner. This shows that there is no need at all for pre-assessing an individual learning style. This is a major advantage over existing practices, based on studies that dedicate a considerable amount of resources to planning a learning styles assessment strategy and implementing it.algorithm

This is how the algorithm works:

  • Step 0: This is the initialization phase, during which a simulated learner profile is presented to the simulated LMS, with profile attributes p1, p2 … pn.
  • Step 1: Upon being presented with a learner profile, the simulated authoring system creates a Reusable Learning Object (RLO) based on initially requested criteria and using existing building blocks. Since this is a simulation, the actual content building blocks need not exist, but only XML-based templates that can be automatically generated in code.
  • Step 2: Upon creation, the RLO is added to a RLO database or repository. Any number of RLOs can then be delivered from the database in order to assemble an educational module suited to the initial learner profile presented to the system.
  • Step 3: When the educational module is complete, it is presented to the learner, followed by a test to assess proficiency on the topic. The test consists of questions related to each RLO used in the educational module.
  • Step 4: If the learner achieves a predetermined passing threshold score, then it can be assumed that the media types of RLOs chosen for the given educational module were appropriate. The content media types satisfactorily match the learner profile. GO TO Step 0. If the learner did not pass the threshold grade CONTINUE to Step 5.
  • Step 5: An evaluation of performance on each RLO is performed.
  • Step 6: The system identifies the particular RLO on which the learner did not pass the threshold. There could be any number of RLO in any given module.
  • Step 7: The system adjusts the learner profile stored in the system and requests different RLOs from the RLO database. GO TO Step 0.

The algorithm has now completed a full iteration and steps 0-8 are repeated. The learner is presented with a modified educational module, which the system believes to more closely match the particular learner profile. The cycle of presentation and assessment is repeated until the learner passes the set threshold.

The system can be programmed to save any number and types of characteristics, circumstances and situational factors. Thus, if a particular learner exhibits different preferences at different times of day or on different topics, these constraints can be easily added to the requests from the RLO database.

It is worth mentioning again that this is a simulator. As such, it only has the ability to generate RLO templates and templates for educational modules. However, the simulator is an excellent testing and proofing mechanism for both the theory and algorithm detailed throughout this study. Furthermore, it can serve as a foundation for an actual implementation of an adaptive learning environment.

The faculty in the College of Science, Engineering and Technology strive to provide all students with the tools and knowledge that they need to succeed. To learn more, visit our website or contact us using the green Request More Information button at the top of the page.