An Interface Adaptation Model for LMSs According to Learning Styles

Alana Oliveira

1, 3 a

, Emanuelle Lemos

1 b

, Anderson Galv

1 c

, Carlos S. Soares Neto

2 d

and Mario M. Teixeira

2 e

Computer Engineering, Federal University of Maranh

ao (UFMA), Brazil

Departament of Informatics, Federal University of Maranh

ao (UFMA), Brazil

PhD Program in Computer Science, DCCMAPI / UFMA, Brazil

Keywords:

Learning Styles, Learning Management Systems, Adaptable Systems, User Interface, User Experience.

Abstract:

An LMS adapted to students’ learning styles has the potential to promote more meaningful learning experi-

ences. This work develops an interface adaptation model for LMSs based on Honey and Mumford’s learning

styles. Relevant characteristics of these styles are mapped and serve as basis for an Adaptation Model used

for LMS interface prototyping regarding activist, reﬂector, theorist and pragmatic students. Information orga-

nization, navigation and presentation are among the aspects considered. A teacher module is also developed

to help instructors create content adapted to different individuals.

1 INTRODUCTION

A Learning Management System (LMS) is a tech-

nological tool that serves as an intermediary in the

teaching-learning process. However, a LMS has, in

general, a standard interface, sometimes differenti-

ated by proﬁle (student, professor, administrative), al-

though it is used by individuals who have different

behaviors, attitudes, interests and motivations.

This creates the need to adapt LMS’s to the partic-

ularities of students, in order to promote more mean-

ingful learning experiences. Among the aspects for

adapting an LMS, the theory of Learning Styles is

of special interest in this work, specifying the pref-

erences of individuals when learning new content.

Knowledge of these styles makes it possible to un-

derstand and describe students’ behavior in a learn-

ing situation, which favors the development of crit-

ical thinking (Cofﬁeld et al., 2004). The choice for

learning styles is not unanimous, as pointed out by

(Kirschner, 2017), however that is the approach used

in this work.

In the literature, there are several studies that

approach the adaptation of learning using student’s

https://orcid.org/0000-0001-7870-3943

https://orcid.org/0000-0002-6451-3708

https://orcid.org/0000-0001-6821-9584

https://orcid.org/0000-0002-6800-1881

https://orcid.org/0000-0001-8771-1478

learning styles as a parameter. Their contributions re-

fer to adaptation at the level of resources, strategies,

order of component presentation and system naviga-

tion. However, there is a lack of research including

adaptation of color, screen layout, formatting, orga-

nization and arrangement of elements on the user in-

terface. Also no research was identiﬁed that helps the

teacher in content customization for different learning

styles, as well as projects developed from the perspec-

tive of user experience (UX).

In this context, this research brings contributions

in the area of adaptive learning by addressing two

main actors of an LMS: the student and the teacher.

Thus, we intend to motivate the student in their learn-

ing activities, customizing the LMS interface to their

learning style in order to improve their performance.

As for the teacher, we want to provide an environment

that assists them in the production and customization

of content to different learning styles, but that does

not generate an additional workload in their already

overloaded teaching routine.

This study aims to develop an interface adaptation

model for LMS’s using Honey and Mumford’s theory

of learning styles (Honey and Mumford, 1992) as a

reference. To do so, we use as a starting point the

mapping of the most relevant characteristics of the

different learning styles, in order to determine how

the interface elements will be adapted to the different

student proﬁles.

From the interface adaptation model, prototypes

Oliveira, A., Lemos, E., Galvão, A., Neto, C. and Teixeira, M.

An Interface Adaptation Model for LMSs According to Learning Styles.

DOI: 10.5220/0011085000003182

In Proceedings of the 14th International Conference on Computer Supported Education (CSEDU 2022) - Volume 1, pages 353-360

ISBN: 978-989-758-562-3; ISSN: 2184-5026

353

of low-ﬁdelity screens are produced for activist, re-

ﬂector, theoretical and pragmatic students. Adap-

tations include techniques for adapting information,

navigation and presentation. In addition, an interface

prototype is implemented for the teacher, since he/she

is responsible for creating and organizing the content

available in the LMS.

This study complements the methodology we dis-

cussed in (Oliveira et al., 2020) where learning styles

are used as a guideline to automatically recommend

educational content to students of similar proﬁles by

leveraging their learning style similarities.

Section 2 discusses the theory of Learning Styles

under the perspective of Kolb, and Honey and Mum-

ford’s proﬁle categorization. Section 3 discusses

some aspects and strategies for LMS adaptation while

Section 4 introduces the adaptation model used in this

work. In Section 5, a comprehensive LMS interface

prototyping is carried out considering different as-

pects within an LMS and their mapping to speciﬁc

learning styles. Section 6 discusses teachers’ role in

this adaptive scenario and Section 7 sums up with

conclusions and future work.

2 LEARNING STYLES

Personal behavior, reasoning process, forms of ex-

pression and strategies for carrying out tasks are some

of the aspects in which human beings differ from one

another. In this perspective, several scholars advocate

that the learning process is not the same for everyone,

so it is necessary to adapt it to people’s particularities,

such as their level of knowledge, motivation, cogni-

tive style and learning style.

In a review carried out by (Cofﬁeld et al., 2004),

71 learning models were identiﬁed, divided into ﬁve

categories. This work takes Honey and Mumford’s

theory as a reference, which extends Kolb’s initial

proposal.

2.1 Kolb and the Experiential Theory

For David Kolb, learning is a cycle of four stages:

Concrete Experience (CE), Reﬂective Observation

(RO), Abstract Conceptualization (AC) and Active

Experimentation (AE) (Kolb, 1976). This has been

called an experiential learning model, a proposal em-

phasizing the relevance of experience and different

skills during the process. It is observed that individu-

als show resourcefulness in certain stages, while they

have signiﬁcant inaptitude in others.

Thus, Kolb proposes a Learning Styles Inventory

(LSI) for the purpose of measuring the weaknesses

and strengths in this cycle and to identify learning

styles (Kolb, 1976). In general terms, the above stages

are organized as a Cartesian plan and four quadrants

are determined referring to the styles identiﬁed by the

LSI: divergent, assimilating, convergent and accom-

modating.

2.2 Honey and Mumford’s Learning

Styles

Kolb’s learning cycle, associated with experiences in

the management area, motivated Peter Honey and

Alan Mumford to develop their own cycle and then

the Learning Styles Questionnaire (LSQ) (Honey and

Mumford, 1992) was born. In general terms, Honey

and Mumford deﬁned four learning styles, each one

being directly related to two stages of Kolb’s cycle.

As Figure 1 shows, the four learning styles identi-

ﬁed are activist, reﬂective, theoretical and pragmatic

(Rosewell, 2005).

Figure 1: Honey and Mumford’s Learning Styles.

3 LMS ADAPTATION ASPECTS

According to (Pereira et al., 2007), Learning Manage-

ment Systems emerged as a technological tool to meet

educational demands, both from schools and universi-

ties, and from the industry, positioning themselves as

an intermediary between the educator and the student.

Traditionally related to distance learning, LMSs have

also been used as a support for face-to-face teaching,

given the variety of resources available.

LMS resources can be classiﬁed into four main

axes: information and documentation; communica-

tion; pedagogical and administrative management;

and production (Pereira et al., 2007). In the 2020’s,

with the prominence of remote teaching due to the

COVID-19 pandemic, virtual spaces sometimes be-

CSEDU 2022 - 14th International Conference on Computer Supported Education

354

came the only alternative for maintaining academic

activities, thus LMSs have established themselves as

signiﬁcant instruments for education as a whole.

LMS customization can help improve learning

experience (Cossul et al., 2020; Thyagharajan and

Nayak, 2007) and is based on criteria related to users’

proﬁles. Among them, learning styles stand out,

as they reveal relevant characteristics of individuals

(Schmeck, 1998) and can be used to customize teach-

ing strategies.

3.1 Adaptation Methodology

The methodology used to provide the adaptation

model based on learning styles can be divided into

three stages as shown in Figure 2:

Figure 2: Proposed methodology.

The ﬁrst step maps the most relevant characteris-

tics of Honey and Mumford’s learning styles. In the

second step, we seek to determine how interface el-

ements will be adapted. Finally, interface prototypes

are developed according to the adaptation model.

Moreover, to carry out the adaptation of a sys-

tem, such as an LMS, three components are needed:

the user model, the domain model and the adaptation

model (da Silva, 2017). The User (or Student) Model

in learning systems represents the signiﬁcant charac-

teristics of the student, including learning styles. The

Domain Model “speciﬁes the conceptual design of

an adaptive hypermedia application” (Aroyo et al.,

2006), since it structures the material to be made

available to the student. The Adaptation Model de-

ﬁnes the adaptation semantics, i.e, it relates the data

from the student model to the domain model in order

to provide the desired adaptation.

In the case of the Adaptation Model, (da Silva,

2017) highlights the following items: Exploration

Form (EF), Detailing Order (DO), Composition Or-

der (CO) and Resources. EF refers to content structur-

ing, which can be linear or networked. DO determines

whether content should be presented from general to

speciﬁc or vice versa. CO, in turn, establishes the or-

der in which the elements of content are revealed to

the student. Finally, resources refer to content to be

made available to students.

4 ADAPTATION MODEL

In this research, the ﬁrst step consists in mapping

the most relevant characteristics of Honey and Mum-

ford’s learning styles, which will determine how the

interface elements will be adapted. Eventually, proto-

types of LMS interfaces customized for each learning

style are developed.

By applying the characteristics of the student

model as input parameter to the domain model, the

Adaptation Model is obtained, which in this work in-

cludes activist, reﬂective, theoretical and pragmatic

learning styles (Figure 3).

Figure 3: Student, Domain and Adaptation models.

In order to adapt the LMS interface, aspects such

as Exploration form, Detailing order, Composition or-

der, Resource types, Color palette, Feedback, Con-

tent organization, and Fundamental components are

considered. Table 1 summarizes these characteristics

organized by learning style as a contribution of this

work.

In general, an activist student is characterized

by a preference for embarking on new experiences

and learning opportunities with an open mind, given

their priority for participation, while, at the same

time, their learning is compromised in situations

that involve well-deﬁned instructions and guidelines.

Therefore, the network exploration form is estab-

lished for this learning style.

Detailing order, in this case, is set from speciﬁc

to general and composition order should be activity-

based (Magoulas et al., 2003). These choices for DO

and CO are deﬁned because of students’ preferences

for the active experimentation and concrete experi-

ence stages of the experiential cycle, which indicate

their understanding of the whole is enhanced by ex-

periences and activities in speciﬁc situations.

In the case of a reﬂector, the experimentation form

should be linear, since this student does not beneﬁt

from carrying out activities without proper prepara-

tion, therefore their exploration of the content should

occur in stages. Detailing order is deﬁned as general

to speciﬁc, since they need a general overview to in-

vestigate and establish their own conclusions. Com-

position order is based on example (Magoulas et al.,

An Interface Adaptation Model for LMSs According to Learning Styles

355

Table 1: Interface adaptation model.

Activist Reﬂector Theorist Pragmatic

Exploration

Form

Network Linear Network Linear

Detailing Order Speciﬁc to General General to Speciﬁc General to Speciﬁc Speciﬁc to General

Composition Or-

der

Activity-based Example-based Theory-based Exercise-based

Resources Competition, Chal-

lenges, Films, Fo-

rum, Infographics,

Games, and Mind

Maps

Handouts, Digital

Books, Papers,

Documentaries,

Films, Exercise

lists, Podcasts, Rid-

dles, Video lessons,

and Webcasts

Handouts, Digital

Books, Papers,

Documentaries,

Video lessons,

Demo videos, and

Webcasts

Documentaries,

Films, Infographics,

Games, Exercise

lists, and Tutorials

Colors Yellow, orange, red Pink, brown, purple,

green

Blue, gray Crimson, yellow,

green

Feedback Individual vs Class Self-analysis vs

Grade

Timeline Pie chart

Content Organi-

zation

Image and content

title

Image, title and

problem

Image, title and

learning objectives

Image, title and ap-

plications

Fundamental

Components

Forums and Last

updates

Notices and Tasks Last updates and

Forums

Tasks and Forums

2003) due to the fact that their learning is better when

allowed to observe the phenomena under study.

For the theorist, the form of exploration should be

in a network, since they need to inquire, question and

investigate. As their learning depends on their un-

derstanding of theories, detailing order is from gen-

eral to speciﬁc and composition order is theory-based

(Magoulas et al., 2003). For these students, the learn-

ing process is based on theories, models, concepts and

facts. Recommended Resources are class notes, dig-

ital books, article, documentaries, videos, and web-

casts.

Pragmatists have a linear exploration form, as

they are more motivated when confronted with clear

guidelines and procedures, thus learning in stages is

more appropriate for them. They are more interested

in practical beneﬁts, so the detailing order should

be from speciﬁc to general. Finally, exercise-based

composition order is recommended (Magoulas et al.,

2003), as they prefer learning the content by solving

practical tasks. Since they need to visualize the appli-

cation of the studied content, documentaries, movies,

games, lists of exercises, tutorials, and infographics

are recommended as educational resources.

5 INTERFACE PROTOTYPING

Based on the Adaptation Model, we prototyped three

distinct screen layouts typically found in Learning

Management Systems: student’s view, course view,

and content view.

A common template is used for all screens with

ﬁve distinct elements: i) a header, ii) a navigation bar,

iii) a footer, iv) a left sidebar, and v) the main content

area. In (i), we ﬁnd information identifying system

and user proﬁle. Item (ii) indicates where the user is

located in the environment while the footer (iii) pro-

vides additional information. Items (iv) and (v) have

a ﬂexible layout adaptable to a speciﬁc learning style

as directed by the Adaptation Model.

Since colors are important elements in a graphic

representation, we chose to build palettes with ﬁve

colors for each style, two of which are ﬁxed and three

are deﬁned according to Table 1 and can be used as

primary, secondary or tertiary colors. Thus, areas (ii)

and (iii) as well as the details of areas (iv) and (v) are

adapted accordingly.

Figure 4: Learning styles color palettes.

5.1 My Courses Screen Adaptation

My Courses module represents all courses a given stu-

dent is enrolled in. The main area is divided into four

CSEDU 2022 - 14th International Conference on Computer Supported Education

356

parts: a menu at the left side (1), a larger center top

area (2), and two smaller areas at the center bottom (3

and 4) as shown in Figure 5.

The left sidebar (1) shows the same items for each

of the learning styles. These items have the same

symbols associated with their descriptions in order

to ensure consistency and system patterns, thus en-

hancing users’ action recognition and system usabil-

ity (Nielsen and Molich, 1990).

The larger area (2) lists a student’s in-progress

courses while the two other areas are adaptable, re-

served for components targeted to speciﬁc learning

styles. For an activist proﬁle, Forums and Latest Up-

dates are prominent (Figure 5-a) while, for a reﬂector,

Notices and Tasks are highlighted (Figure 5-d). For a

theorist, Latest Updates and Forums are visible (Fig-

ure 5-c) while, for pragmatists, Tasks and Forums call

their attention (Figure 5-b).

Besides interaction objects, the color palette is

distinct for each learning style as shown above (Fig-

ure 4). This form of adaptation is compliant with the

page variants technique since acreen layout and other

visible components are presented in different ways for

distinct users (Batista, 2008).

5.2 Course Screen Adaptation

In the Courses module screen (Figure 6), the left side-

bar is divided into two regions: course menu items

are displayed at the top two-thirds, and the student’s

performance is highlighted at the bottom one-third.

Performance information is presented differently ac-

cording to the learning style. The content of the main

area varies in accordance with the Adaptation Model.

An activist obtains feedback on their performance

through a bar graph that shows their individual statis-

tics compared to the class average. Content, in turn, is

presented with an image and title without any access

restrictions (Figure 6-a).

For a reﬂector, performance feedback compares

student self-analysis to the system’s analysis over

time. Content topics start with a question or problem

for reﬂection purposes. Since learning style explo-

ration pattern is linear, new topics are unlocked only

upon completion of previous ones (Figure 6-d).

For a theorist, feedback is presented in a line chart

so that they can visualize their performance over time.

Ideally, this graph should be interactive in order to al-

low the student to explore speciﬁc mistakes or suc-

cesses. Contents listed present an image, title and ed-

ucational objective. As their form of exploration is

networked, content can be acessed in any order (Fig-

ure 6-c).

A pragmatist has performance data presented as

pie charts for a better understanding. Their form of

exploration is linear, therefore certain navigation re-

strictions between content topics are imposed. Con-

tent list shows a title with an image and it highlights

practical applications and beneﬁts of the content (Fig-

ure 6-b).

5.3 Content Screen Adaptation

Last but not least, the Content module has been

adapted considering Resources, Exploration Form,

Detailing Order and Composition Order, as com-

mented in Section 3.1. The Content screen shown in

Figure 7 exhibits two areas on the left sidebar: the

content navigation area (top) and additional resources

(bottom). The main area on the right side is reserved

for course content.

Content navigation area presents contents so that

the student can understand the structure of the course.

This area has been adapted according to the explo-

ration form (EF). If EF is networked, the whole con-

tent will be readily available to the user for naviga-

tion purposes. If EF is linear, content will be released

gradually according to student progress. In this case,

Link Annotation navigation technique has been used.

The content under study is highlighted in larger bold

letters while other content not currently relevant ap-

pears in a lighter font.

Additional resources refer to materials that are

complementary to student learning. They are indi-

cated based on the types of resources recommended

to their learning style and also based on the detailing

order of these resources. For instance, an activist stu-

dent could ﬁnd a Mind Map, whose detailing is from

speciﬁc to general. To a theorist, on the other hand,

the LMS might recommend a paper discussing a cer-

tain content from a general to a more speciﬁc perspec-

tive,

The main area, or content area, presents content

divided into Activity, Example, Exercise and Theory

distributed in four tabs. Tag order is determined by

the learning style composition order. For instance, a

reﬂective student will see the sequence Example, The-

ory, Exercise and Activity (Figure 7-d). A theorist

student, on the other hand, will ﬁnd Theory, Exam-

ple, Exercise and Activity, in conformance with their

learning style (Figure 7-c). The exploration form dic-

tates whether all tabs will be open for navigation (the

case of activists and theorists) or will be gradually

made available for reﬂectors and pragmatists (observe

the padlocks on the tabs).

An Interface Adaptation Model for LMSs According to Learning Styles

357

Figure 5: My Courses screen adapted to distinct learning styles.

Figure 6: Course screen adapted to distinct learning styles.

CSEDU 2022 - 14th International Conference on Computer Supported Education

358

Figure 7: Content screen adapted to distinct learning styles.

6 TEACHER ROLE IN AN

ADAPTABLE LMS

A Learning Management System can be character-

ized as an intermediary between the student and the

teacher, since it mediates the teaching-learning pro-

cess (Pereira et al., 2007).

The teaching practice in an LMS requires the

teacher to be the author of the contents made avail-

able to the students. In the interface adaptation model

proposed in this work, content is presented accord-

ing to the speciﬁc Composition Order for the student’s

learning style, as well as complementary materials are

recommended. The recommendation of these materi-

als follows the criteria of the types of Resources indi-

cated associated with the Detailing Order.

As seen above, different learning styles dictate

different content selection, presentation, organization,

exploration and so forth. Management of these dif-

ferent elements, speciﬁed in the adaptation model,

should be carried out by the system, not by the

teacher, whose role is to organize and provide con-

tent. The LMS will determine the target student and

how this content will be shown and consumed.

In the adaptable LMS architecture proposed, the

teacher should organize a Lesson Plan to serve as a

road map for a given course. This plan is divided into

three areas: general information, content, and teach-

ing resources.

Teacher’s Course Management screen (Figure 8)

presents the content listing of a speciﬁc course as-

signed to an instructor. On the left sidebar, there is

a menu similar to that of students with the insertion

of a Class Proﬁle item in order to present class pref-

erences in terms of learning styles. This serves as a

guideline for teachers when organizing course mate-

rial.

Figure 8: Course Management: content selection.

An Interface Adaptation Model for LMSs According to Learning Styles

359

Figure 9: Course Management: planning.

Also, the Planning screen in Figure 9 contains a

lesson plan with a main area divided into three tabs re-

ferring to different aspects of a class. In addition, the

sidebar has a content navigation area the teacher may

use to organize course material. Finally, the teacher

can choose how to visualize content on the perspec-

tive of different learning styles.

7 CONCLUSION

In this work, the characteristics of learning styles

were mapped in relation to favorable and unfavorable

learning situations, as well as aspects related to pri-

oritization, problem-solving perspective, interaction,

and personal preferences in relation to content. These

characteristics, moreover, serve as parameters for the

development of an LMS Adaptation Model.

The referred model was used for LMS interface

prototyping for activist, reﬂector, theorist and prag-

matic students. The proposed adaptations included

content selection, navigation and presentation. In

addition, it was provided a teacher interface, in the

form of a lesson plan, to facilitate the organization of

course material for different student proﬁles.

This resarch brings contributions to the area of

adaptive learning in relation to strategies and re-

sources for learning management environments. As

future work, it is intended to validate the proposed

approach by organizing real courses in an adaptable

LMS.

REFERENCES

Aroyo, L., Dolog, P., Houben, G.-J., Kravcik, M., Naeve,

A., Nilsson, M., and Wild, F. (2006). Interoperability

in personalized adaptive learning. Educational Tech-

nology & Society, 9:4–18.

Batista, C. R. (2008). Modelo e diretrizes para o pro-

cesso de design de interface web adaptativa. Pro-

grama de P

os-Graduac¸

ao em Engenharia e Gest

ao do

Conhecimento, Universidade Federal de Santa Cata-

rina (UFSC), Florianop

olis.

Cofﬁeld, F., Moseley, D., Hall, E., and Ecclestone, K.

(2004). Learning styles and pedagogy in post-16

learning: A systematic and critical review. Learning

and Skills Research Centre, Londres.

Cossul, D., Fagundes, B. J., Ferreira, G., Frozza, R., Sil-

veira, W. A. N., Clavijo, M. L. T., and Orozco, W.

J. G. (2020). Ambiente virtual de aprendizagem:

uma abordagem baseada em mediac¸

ao tecnol

ogica

personalizada. Brazilian Journal of Development,

6(12):101874–101888.

da Silva, Z. C. (2017). Adaptac¸

ao de apresentac¸

ao de

conte

udos de objeto de aprendizagem considerando

estilos de aprendizagem. Universidade Federal do

Paran

a, Setor de Ci

encias Exatas, Programa de

osGraduac¸

ao em Inform

atica, Curitiba.

Honey, P. and Mumford, A. (1992). The learning styles

questionnaire. Industrial and Commercial Training,

24(7):10–13.

Kirschner, P. A. (2017). Stop propagating the learning styles

myth. Computers & Education, 106:166–171.

Kolb, D. A. (1976). Management and the learning process.

California management review, 18(3):21–31.

Magoulas, G., Papanikolaou, Y., and Grigoriadou, M.

(2003). Adaptive web-based learning: accommo-

dating individual differences through system’s adap-

tation. British journal of educational technology,

34(4):511–527.

Nielsen, J. and Molich, R. (1990). Improving a human-

computer dialogue. Communications of the ACM,

33(3):338–348.

Oliveira, A., Teixeira, M. M., and Neto, C. d. S. S. (2020).

Recommendation of educational content to improve

student performance: An approach based on learning

styles. In CSEDU (2), pages 359–365.

Pereira, A. T. C., Schmitt, V., and Dias, R. A. C. (2007).

Ambientes virtuais de aprendizagem. In Ambientes

Virtuais de Aprendizagem em Diferentes Contextos,

chapter 1, pages 4–22. Editora Ci

encia Moderna Ltda,

Rio de Janeiro.

Rosewell, J. (2005). Learning styles. Times Educ. Suppl,

(4659):11–14.

Schmeck, R. R. (1998). Individual differences and learn-

ing strategies, chapter 10, pages 171–191. Academic

Press.

Thyagharajan, K. and Nayak, R. (2007). Adaptive con-

tent creation for personalized e-learning using web

services. Journal of Applied Sciences Research,

3(9):828–836.

CSEDU 2022 - 14th International Conference on Computer Supported Education

360