A Brief Review on Adaptive Learning Applied for Teaching Physics

Manal Elfoudali, Ayoub Ait Lahcen

Engineering Sciences Laboratory, National School of Applied Sciences, Ibn Tofail University, Kenitra, Morocco

Keywords: E-Learning, Adaptive Learning, Adaptive Feedback, Physics

Abstract: Web-based learning or E-learning is becoming more and more popular. Moreover, the need for effective

learning environments is increasing by the day. The aim of using e-learning is to enable students to not only

access learning materials outside the classroom but also to present the students with the needed help and

assistance. However, most e-learning systems lack personalization and offer the exact content to all students

when in fact, learners each have their cognitive style and prior knowledge that is different from any other

student. This is the reason behind implementing adaptive learning in online platforms and why it is considered

an essential step and is incredibly important. The purpose of this paper is to provide a literature review of

previous work related to the use of various adaptive learning environments for teaching physics.

1 INTRODUCTION

To put it simply, an e-learning system is essentially a

virtual classroom, where teachers put forward the

course materials students need in the form of text,

tutorials, simulations, etc. The materials in an e-

learning platform are the same and are meant to be

used by all students. Except students have distinct

cognitive styles and different prior knowledge, this is

the very reason behind why these e-learning

platforms were not as effective as they were supposed

to be.

This sparked interest in adaptive learning, which

has been the subject of several research papers over

the years, many of which looked at different aspects

of adaptive learning systems for learning in general

and learning physics in particular.

Adaptive feedback is without a doubt one of the

crucial aspects studied. The web pages of the ALE

(Adaptive Learning Environment) (Psycharis, 2007)

were adapted to FI (Field Independent) learners and

FD (Field Dependent) learners, being the two types of

students’ cognitive styles.

Some papers explored the text content on itself in

the form of adaptive scaffolds (Chen, 2014), both

types hard and soft:

 Soft scaffolds: During the learning process,

soft scaffolds refer to complex, situation-

specific help offered by an instructor or peer,

https://orcid.org/0000-0001-8739-3369

 Hard scaffolds: While the learner is

dynamically engaged with a challenge, hard

scaffolds are used to provide learner support at

various difficult stages.

While other papers (DeVore et al, 2017)

(Marshman et al, 2018) focused on the learning

tutorials, which are in the form of quantitative

problems that are broken down into sub-problems to

help students deepen their understandings of physics

principles.

Some research has been conducted on the

multiple challenges’ students face with self-paced

learning that relates to different factors: internal

characteristics, external characteristics, student

characteristics, and the learning tool characteristics,

which are detected using the SELF (Strategies for

Engaged Learning Framework) (Marshman et al,

2018), and that inevitably affect the students’ learning

process.

Another topic of discussion was the connection

between physics and mathematics (Nishioka et al,

2018) (Kudo et al, 2018). It was noticed that plenty of

students who face difficulties usually do not have a

good understanding of how mathematics is applied to

physics.

Furthermore, when taking an MC (Multiple

Choice) test, students may choose correct or

incorrect answers with the wrong intentions. This

was the reason behind the development of the CMR

Elfoudali, M. and Ait Lahcen, A.

A Brief Review on Adaptive Learning Applied for Teaching Physics.

DOI: 10.5220/0010731600003101

In Proceedings of the 2nd International Conference on Big Data, Modelling and Machine Learning (BML 2021), pages 227-231

ISBN: 978-989-758-559-3

227

(Coupled-Multiple Response) test (Rios et al, 2020)

that has a reasoning element, in addition to the usual

MC test, which helps in identifying which and

where interventions are needed for students when

using the adaptive learning technology.

The purpose of this paper is to provide a literature

review of previous work related to the use of e-

learning and various ALEs (Adaptive Learning

Environments) for teaching physics which will be

presented in sections 2 and 3. The conclusion of this

paper is followed by a discussion of future work in

section 4.

2 GENERIC E-LEARNING

APPROACHES APPLIED TO

PHYSICS

Learning can be pretty challenging at various levels

of education: middle school, high school, and college.

Additionally, for the sake of assisting students in this

process, text, graphics, images, videos, and

simulations were introduced to the conventional

teaching methods, hence the start of the e-learning

era.

• Basic e-learning approaches: Learning technologies

were established to facilitate the learning process for

students and the teaching process for teachers, which

were able, through web-based learning platforms, to

give out assignments to students without having to be

face to face. In the case of (Fayanto et al, 2019), the

learning technology MOODLE (Modular Object-

Oriented Dynamic Learning Environment) was

adopted by high school seniors for physics learning.

The platform allows students to access course

material outside of the classroom and enables

teachers to give out quizzes, assignments, etc. On top

of that, it offers simulations to demonstrate some

applications and encourages students to visualize

physical phenomena better. An additional feature that

is provided by this learning technology (Moodle) is

the simplicity of communication, not only between

students and their teacher but also between the

students utilizing chats and forums.

• Game-based learning (GBL): Various video games

are used for learning for the reason that GBL explores

novel learning techniques and is considered one of the

ICT tools (Information and Communication

Technology). One of the most popular games among

middle schoolers is Angry Birds (AB). It is a well-

known physics-based puzzle game in which the aim

is to destroy all of the pigs in a 2D level location by

using a large number of birds. To go to the next level,

each level has a collection of tasks/goals that illustrate

a range of topics. Moreover, it was noticed that

students play the AB game for long periods that vary

from 6 hours to 8 hours a day. Furthermore, to use the

wasted time productively, (Department of Computer

Science, Shah Abdul Latif University, Khairpur,

Sindh, Pakistan & Umrani, 2020) used the AB

interface as a tool to teach basic physics principles.

This technique was feasible since students have

access to mobile devices; this way not only do they

get to play their favorite game, but they also learn

along the way.

However, moving from the usual classroom to a

virtual one and allowing self-paced learning was not

enough. There was always the problem of having the

exact content delivered to all students; in other

words, a one size fits all platform. This was when

research moved the focus to the adaptive aspect of

learning to personalize the instructional material for

each student individually.

3 ADAPTIVE LEARNING

APPLIED TO PHYSICS

Along with adaptive learning, numerous challenges

have emerged which affected students’ motivation,

confidence level, etc…, that undoubtedly influenced

students’ learning process, especially for physics

learning.

• Students’ cognitive styles: E-learning environments

lack customization, so identical content is offered to

all the students. Nevertheless, the web pages of the

ALE (Adaptive Learning Environment) (Psycharis,

2007) program were developed to help students in

learning physics and were adapted to two types of

students’ cognitive styles: FI (Field Independent) and

FD (Field Dependent). This way, FI learners were

presented with physics concepts from specific to

general, as they tend to have an analytical approach

and are more autonomous when it comes to the

development of their cognitive skills. However,

information from general to specific was provided to

FD students who usually need more assistance in their

learning process and approach their environment

from the global scope. Additionally, the ALE focused

mainly on Problem-Based Learning (PBL), which

used real-world problems to help students develop

their problem-solving skills.

• Students’ proficiency: The adaptive learning

platform WPO (Wiley Plus ORION) (Basitere &

Ivala, 2017) is one of the online-based feedback

systems used for learning physics. When using the

platform, learners start by taking a diagnostic test to

BML 2021 - INTERNATIONAL CONFERENCE ON BIG DATA, MODELLING AND MACHINE LEARNING (BML’21)

228

evaluate their proficiency level on the physics lessons

taught in the classroom. And according to learners’

level of performance, the online system adapts the

difficulty level of the diagnostic test questions. The

data generated from the diagnostic test is provided to

the students and broken into three categories: low

performance, medium performance, and high

performance. The teachers can also use this data to

grant necessary assistance and support to the students,

give Web-Based Proficiency Homework (WBPH),

and assignments that are graded automatically to

strengthen their level of proficiency.

• The text content: Physics subjects such as

velocity and acceleration appear simple to teach

when in fact, they are complicated topics. While

students are familiar with the terminologies, the

concepts behind them are often difficult to teach due

to misconceptions and dependence on existing

knowledge. (Chen, 2014) addresses the process of

designing adaptive scaffolds that take into account

cognitive aspects of learning, such as students’

current level of proficiency and their prerequisite

ZPD’s (Zone of Proximal Development). In addition,

Brophy’s ZMPD (Zone of Motivational Proximal

Development) suggests that the adaptive scaffolding

e-learning system should also pay attention to

learners’ motivational needs that can be empowered

through scaffolding. This study further proves that

one size does not fit all.

• Self-paced learning challenges: Both articles

(DeVore et al, 2017) and (Marshman et al, 2018)

discuss the challenges that students face with

physics self-paced learning. Three e-learning

tutorials on introductory mechanics were used in the

(DeVore et al, 2017) investigation. Each tutorial

contained a quantitative problem that was broken

down into a series of sub-problems to help students

develop their problem-solving skills and improve

their self-reliance. However, the self-paced learning

tutorials remain challenging for students. This is

why the SELF (Strategies for Engaged Learning

Framework) (Marshman et al, 2018) was put in

place to detect what are precisely the factors, be

them internal and external factors, that influence the

learning process and how they can be taken into

consideration and implemented in the development

of the self-paced learning tutorials, as well as using

quantitative problems that are divided into sub-

problems to further improve the efficacity of

learning for students and enhance their

understandings of physics principles.

• Adaptive feedback: Feedback is essential during

physics problem-solving in an adaptive learning

environment, where the three main knowledge

components are usually treated in isolation. In

(Bimba et al, 2018), these three components are

portrayed in the form of models, using the OAR

model (object, attribute, and relations): the

pedagogical model that represents the technique and

knowledge of teaching, domain model which

constitutes the facts, rules, equations, feedback, and

student model containing information about

students learning style and their understanding of the

domain. The relationships between these

components are hard to illustrate using existing

methods that can only represent the relationship

between a pair of concepts. (Bimba et al, 2018)

proposes a concept operator that can represent the

relationships between multiple criteria and therefore

represents the relationships between the three

knowledge components.

• Learning management systems (LMS):

Instructional methods (e.g., advice from a teacher or

specific instructions) that are useful for novices in a

given field may lose their efficacy or even be

detrimental when applied to experts. This

phenomenon which is often referred to as the

Expertise Reversal Effect demonstrates how crucial

it is to tailor the learning process to the needs of the

learners, which in the case of (Imhof et al, 2018) are

college students. Moreover, depending on the

students’ prior knowledge and their online activity,

meaning the number of tasks solved daily, the LMS

Moodle used for the physics module, otherwise

known as the problem module in the 2015/16 and

2016/17 semesters of the Swiss Distance University

of Applied Sciences (FFHS), implemented an

adaptive task set combined with a simple

recommender system that gives feedback to students

according to the tasks they chose, either detailed

step-by-step tasks or non-detailed tasks. The former

ones performed well with low to medium prior

knowledge students. The latter ones were sort of

effective with high prior knowledge students even

though they had less learning progress compared to

other students.

In Russia, the Moodle course developed by the

Elabuga Institute of Kazan (Volga region) Federal

University (Shurygin & Krasnova, 2016) not only

makes the teaching material available outside of the

classroom but also indispensable in regards to the

self-education and the self-development of students.

The system also helps in monitoring students’ online

presence and provides real-time assistance through

A Brief Review on Adaptive Learning Applied for Teaching Physics

229

private messages with the teacher or through forums.

Furthermore, within e-learning applications,

especially blended learning (Krasnova & Shurygin,

2020), the LMS Moodle is an incredible tool that can

be used in the design of refresher courses for teachers

in general and physics teachers in particular.

• Remedial instruction system: Designed for

senior high school physics instruction, the remedial

instruction system’s (Lin et al, 2009) learning

material is presented in the form of units for

different concepts, which are introduced to the

learners one at a time so that they are not

overburdened with information. The duration of

each unit varies between 1 to 3 minutes and includes

audio, graphics, text, and video. The learning

material is also available anytime and anywhere.

The remedial instruction system consists of a

comprehensive tracking mechanism for

documenting the participants’ entire learning

experience to gain a greater understanding of their

learning process. In addition, when users log out, the

system asks them to complete a real-time survey

about whether the learning session helped them

understand the subject, whether there were any

unanswered questions, and so on. The efficacy of

learning can be measured through the survey, and

the results can be used to plan individualized

remedial training to improve learning effectiveness.

Both the framework’s tracking mechanism and the

survey were used to give an insight into the concepts

students faced difficulties with and where they

needed more help.

• The connection between mathematics and

physics: Usually, e-learning websites must have an

adequate hyperlink structure, which allows quick

access to information and reference to documents

spread around the internet using a search engine and

a collection of keywords. Based on this approach,

the Japanese Kanazawa Institute of Technology, in

short KIT, developed different technologies for

Mathematical Navigation KITMN and Physics

Navigation KITPN in 2004 and 2016, respectively

(Nishioka et al, 2018). Moreover, to enhance the

learning performance of high schoolers, university

students, and even engineers, KIT grants an

effective learning environment of physics by

providing the connection between mathematical and

physical knowledge, as well as visualizing equations

by using simulations. The concept behind these

technologies was that one webpage should have one

topic for optimal searchability; this way, visitors can

browse the webpages from basic knowledge to

advanced knowledge and thus deepen their

understanding of physics and grasp its connection

with mathematics. These frameworks were also

evaluated (Kudo et al, 2018) based on access logs

obtained from visitors who had consulted a KIT

environment webpage at least once to draw out the

browsing paths of visitors who were curious about

the webpage’s content. Furthermore, a cluster

analysis was used to determine which topics were of

interest to the website visitors based on the number

of visits and the length of those visits.

• The evaluation process: MC (Multiple Choice)

tests have questions with one correct answer and

multiple incorrect distractors. However, CMR

(Coupled-Multiple Response) test items have two

parts: the usual MC part and a reasoning element

which renders the students with the option of giving

a justification for their answer. Some students may

choose the correct answers for the wrong reasons;

however, others may choose the wrong answer with

the correct reasoning. (Rios et al, 2020) examines the

reasoning patterns of students when taking an MC

test that led them to choose incorrect, partially

correct, and correct answers. The reasoning element

of the CMR test items was proven very helpful in

identifying which and where interventions are

needed in the adaptive learning technology for

learning physics.

4 CONCLUSIONS

E-learning tools are a great way to help students

engage in self-paced learning outside the classroom.

These tools enable them to access learning material

anytime and anywhere. Students can also be given

assignments online.

However, self-paced learning has its drawbacks

as students can face a lack of motivation, their

productivity can be affected by internal and external

characteristics, which are easily identified using the

SELF framework. Speaking of students, there are

two types of learners: FI (field-independent), FD

(field dependent), and two types of adaptive

scaffolds, which play an essential part in the

adaptive learning environment. Both hard scaffolds

and soft scaffolds provide learners with the needed

support at the different stages of problem-solving

depending, of course, on their learning styles and

their prior knowledge. Students may also struggle

with the material that can be too complex for them

to understand independently or due to their

academic shortcomings in mathematics as they do

not understand its connection with physics.

BML 2021 - INTERNATIONAL CONFERENCE ON BIG DATA, MODELLING AND MACHINE LEARNING (BML’21)

230

Moreover, one of the essential parts of the

adaptive learning environments mentioned

previously is adaptive feedback, which goes hand in

hand with recommender systems, as a means to help

students in their learning process to deepen their

understanding of physics principles.

Future work will focus mainly on the use of

recommender systems to develop an adaptive

learning environment for teaching physics to high

school students in Morocco.

ACKNOWLEDGEMENTS

This work was supported by the Al-Khawarizmi

Program funding by Morocco’s Ministry of

Education, Ministry of Industry, and the Digital

Development Agency (ADD) under Project No.

451/2020 (Smart Learning).

REFERENCES

Basitere, M. M., & Ivala, E. N. (2017). Evaluation of an

Adaptive Learning Technology in a First- year Extended

Curriculum Programme Physics course. South African

Computer Journal, 29(3).

Bimba, A. T., Idris, N., Al-Hunaiyyan, A.

A.,Mahmud, R. B., & Shuib, N. L. B. M. (2018).

Design of an Algebraic Concept Operator for Adaptive

Feedback in Physics. Challenges and Opportunities in

the Digital Era (Vol. 11195, pp. 181–190). Springer.

Chen, C.-H. (2014). An adaptive scaffolding e- learning

system for middle school students’ physics learning.

Australasian Journal of Educational Technology.

Department of Computer Science, Shah Abdul Latif

University, Khairpur, Sindh, Pakistan, & Umrani, S.

(2020). Games based learning: A case of learning

Physics using Angry Birds.Indian Journal of Science

and Technology.

DeVore, S., Marshman, E., & Singh, C. (2017).

Challenge of engaging all students via self- paced

interactive electronic learning tutorials for introductory

physics. Physical Review Physics Education Research.

Fayanto, S.-, Kawuri, M. Y. R. T., Jufriansyah, A.,

Setiamukti, D. D., & Sulisworo, D. (2019).

Implementation E-Learning Based Moodle on Physics

Learning in Senior High School. Indonesian Journal of

Science and Education.

Imhof, C., Bergamin, P., Moser, I., & Holthaus, M. (2018).

Implementation of an adaptive instructional design for

a physics module in a learning management system.

Krasnova, L. A., & Shurygin, V. Y. (2020). Blended

learning of physics in the context of the professional

development of teachers.

Kudo, T., Nishioka, K., & Nakamura, A. (2018). Evaluation

for e-learning website of physics by browsing path

analysis and cluster analysis of access log. Journal of

ICT, Design, Engineering and Technological Science.

Lin, L.-F., Huang, C.-Y., & Hsieh-Chih, C. (2009). The

Construct and Evaluation of High School Physics

Online Remedial Teaching System. 2009 International

Conference on Research Challenges in Computer

Science, 214–217.

Marshman, E. M., DeVore, S., & Singh, C. (2018).

Challenge of Helping Introductory Physics Students

Transfer Their Learning by Engaging with a Self-Paced

Learning Tutorial. Frontiers in ICT.

Nishioka, K., Kudo, T., & Nakamura, A. (2018). Learning

Support Website of Physics with Emphasis on

Connection with Mathematics.

Psycharis, S. (2007). Designing Adaptive Learning

Environment according to cognitive stylesand its

influence on students’ achievement and beliefs for

Physics.

Rios, L., Lutz, B., Rossman, E., Yee, C., Trageser,

D.,Nevrly, M., Philips, M., & Self, B. (2020). Creating

coupled-multiple response test items in physics and

engineering for use in adaptive formative assessments.

IEEE Frontiers in Education Conference, 1–5.

Shurygin, V. Y., & Krasnova, L. A. (2016). Electronic

Learning Courses as A Means to Activate Students’

Independent Work in Studying Physics.

A Brief Review on Adaptive Learning Applied for Teaching Physics

231