TOWARDS EFFECTIVE BLENDED LEARNING WITH 3D MUVE

An Analysis of Use Case Implementations for 3D MUVE Learning

Indika Perera, Colin Allison, John McCaffery and Alan Miller

School of Computer Science, University of St. Andrews, KY16 9SX, Scotland, U.K.

Keywords: Blended learning, Learning in 3D MUVE, Learning use cases, Blended learning challenges, Learning

environment integration, Second Life, Open Simulator, Moodle, Sloodle.

Abstract: Technology enhanced learning has been a mainstream educational process, which has introduced a variety

of techniques to use in different contexts to fulfil learning objectives and goals. E-Learning has been widely

used to fulfil various learning requirements from primary education to university and research based

education. However, criticisms on the existing learning methods on competence to cater for societal and

human needs within the context of learning are not rare, associating the constraints of behavioural, cultural,

and pedagogical. 3D Multi User Virtual Environments (MUVE), in contrast, show a promising future for

dynamic learning activities, complementing blended learning methods with high collaboration and user

engagement, in which some of those would not have been possible with existing learning practices. This

paper, nonetheless, shows that the present 3D MUVE learning use cases are not well defined, yet

educationalists tend to practice and expect the conventional e-Learning use cases in 3D MUVE, creating

inconsistencies and losing the significance of 3D MUVE for learning. It also proposes a novel approach to

consider effective 3D MUVE learning use cases. The use case analysis has been done on a blended

perspective. Moreover, the paper critically argues about the effectiveness of learning activities designed

with 3D MUVE and e-Learning through selected case studies.

1 INTRODUCTION

In various segments of our society, 3D Multi User

Virtual Environments (3D MUVE) can be seen as

useful applications. In effect, 3D Learning

Environments based on Multi-User Virtual worlds

are a growing feature of UK education (Kirriemuir

2010). Virtual worlds with simultaneous interactions

of thousands of people in a shared 3D space, show

frontier and critical implications for business,

education, social and technological sciences, and

society at large (Messinger, Stroulia et al. 2009).

Leading universities are interested in, and have been

researching on, applications of this novel technology

for their learning needs. 3D MUVE support more

intuitive activities for learning complex and

advanced concepts. They are particularly appropriate

for educational use due to their alignment with

Kolb's (Kolb, Boyatzis et al. 2001) concept of

experiential learning, and learning through

experimentation as a particular form of exploration

(Allison, Miller et al. 2008). Interactive 3D virtual

environments demonstrate a great educational

potential due to their ability to engage learners in the

exploration, construction and manipulation of virtual

objects, structures and metaphorical representations

of ideas (Dalgarno, Bishop et al. 2009).

Virtual worlds provide new opportunities and

challenges for technology-enhanced learning

(Allison, Miller et al. 2010). Consequently, many

higher education courses with novel and engaging

approaches to conduct practical coursework have

been virtual worlds in academia. The knowledge on

appropriate learning use cases is essential for the

success of learning in 3D MUVE, as many

educationalists are interested in. Therefore, in this

research, we have focused on identifying key use

cases for learning with 3D MUVE support. For our

research and learning activities, we choose Second

Life - SL (Linden_Labs 2003) and Open Simulator

(The_Open_Simulator_Project 2007) as 3D MUVE

and, we consider Moodle (Moodle 2004) as an e-

Learning platform. More details about the work we

have done with these environments will be discussed

later. A significant point of interest for users of 3D

virtual worlds as learning environments is that this

paper considers popular public domain, open source

Perera I., Allison C., McCaffery J. and Miller A..

TOWARDS EFFECTIVE BLENDED LEARNING WITH 3D MUVE - An Analysis of Use Case Implementations for 3D MUVE Learning.

DOI: 10.5220/0003339500460055

In Proceedings of the 3rd International Conference on Computer Supported Education (CSEDU-2011), pages 46-55

ISBN: 978-989-8425-50-8

 2011 SCITEPRESS (Science and Technology Publications, Lda.)

platforms OpenSim, Moodle and Sloodle for

learning activities, which are capable of providing a

complete blended learning environment with 3D

virtual world support.

It is important to distinguish between virtual

worlds, such as Second Life, and massively

multiplayer online role playing games (MMORPG)

such as World of Warcraft (WoW), as many

commentators still seem to regard them as belonging

to the same genre. In virtual worlds all the content,

building and terra-forming is the result of user

activity. There are no goals per se. Unlike

MMORPGs the client does not hold a model of the

entire world. As avatars move around, their

immediate environment is downloaded dynamically

to the client. Consequently, virtual worlds require

significantly more network bandwidth than 3D

games. For example, WoW has been measured as

using as little as 7 Kbps (Sheldon, Girard et al.

2003), whereas Second Life and OpenSim will

readily use more than 500 Kbps. This can be an

issue for a limited resource blended learning

environment that looks for 3D MUVE engagement.

Appropriate learning use cases should be designed

considering 3D MUVE network bandwidth

requirements as well. Delivering the desired Quality

of Service to a virtual world is a live and challenging

subject (Oliver, Miller et al. 2010).

This paper is arranged into the following

sections: in section 2 we describe background details

along with our experiences on learning in 3D

MUVE; section 3 explains the high level model that

we use to analyse 3D MUVE supported learning in

the context of existing learning methods. Section 4

elaborates appropriate learning use cases for 3D

MUVE learning while considering the research

environments we have used. Sections 5 and 6 briefly

present two case studies we have worked on to

examine the proposed model while presenting the

use case mapping challenges and domain specific

comparisons, respectively. The section 7 describes

the relevance of research findings for security policy

development on 3D MUVE learning as the future

research work, before concluding.

2 BACKGROUND

AND RELATED WORK

E-learning gives many advantages, but there have

been criticisms on using it as a mainstream method

of education. In fact, this concern was examined by

Graf and Kinshuk in their work on e-Learning

adaption to standard learning styles (Graf and

Kinshuk 2009). A study on e-Learning impact for

successful learning activities was done using an

approach of task-technology fit (McGill and Klobas

2009). It was found that the perceived benefits of e-

learning utilization are higher than that of the actual

outcome in the form of student grades, in light of the

argument that the technical constraints and

underutilization of the possible use cases could have

resulted in such observation, through poor

collaboration and irrational learning methods, due to

overwhelming technology perception of the users.

Additionally, technological limitations on providing

learning content and activities without 3D rich

formats can be a significant cause for a failure of a

learning activity. Moreover, trying to map traditional

models of learning into e-learning has resulted in

weaknesses that we experience in some available e-

learning applications (Teo and Gay 2006).

Importantly, monotonous ways of interacting

students, without their preferred personalization has

resulted in poor engagement on learning activities,

as well.

Blended learning refers to instructional

approaches with multiple learning delivery methods,

including most often face-to-face classroom with

asynchronous and/or synchronous online learning.

Rich collaboration and user friendliness are expected

norms on multiple platforms in blended learning

(Brenton 2009). Blended learning is characterized as

maximizing the best advantages of face-to-face and

online education (Wu, Tennyson et al. 2010). As a

result, the possibility of incorporating 3D MUVE as

a complementary learning platform with existing

learning environments has been widely used in this

work. However, we insist that the new blended

learning paradigm should only consist of productive

learning use cases of 3D MUVE to avoid

redundancies and suboptimal practices. Weippl’s

extensive set of factors and use cases for e-Learning

security management (Weippl 2005), has been used

for this analysis in a blended approach.

The University of St Andrews has ventured in a

diverse set of educational and research projects

incorporating 3D MUVE. The educational projects

mainly include the course delivery and assessment in

3D MUVE with a significant success. The Laconia

Acropolis Virtual Archaeology (LAVA) (K.

Getchell, Nicoll et al. 2007), a joint project between

schools of Classics and Computer Science, allows

students to engage with a simulated archaeological

excavation, and then explore a recreation of the site

in Second Life. The WiFi Virtual Laboratory in

Second Life project (WiFiSL) (Sturgeon, Allison et

TOWARDS EFFECTIVE BLENDED LEARNING WITH 3D MUVE - An Analysis of Use Case Implementations for 3D

MUVE Learning

al. 2009) is used as a learning aid for wireless

networking. It facilitates collaborative explorations

and visual simulations of wireless traffic using 3D

virtual world interfaces. Another computer

networking related research project is to examine

Second Life network traffic. It has been a validating

study of previous researchers' findings whilst

offering new insights of SL traffic management, as a

client end measurement analysis (Oliver, Miller et

al. 2010). Teaching Human Computer Interaction

(HCI) to honours students has been a successful

effort, once with Second Life (Perera, Allison et al.

2009) and another instance with OpenSim (Perera,

Allison et al. 2010). Figure 1 shows some of the

creative mechanisms students have developed.

Figure 1: Students’ coursework for HCI in Second Life

and OpenSim – Dijkstra’s shunting algorithm, and

interactive door systems for enclosures.

The Module Management System (MMS)

interoperates with Second Life in order to maintain

an association of institutional and virtual world

identities while being a learning management

system. The university is in the process of

introducing Moodle in place of WebCT, which

provide a Single-Sign-On based Moodle-MMS e-

Learning platform for a seamless course

management service. Therefore, with the experience

on using 3D MUVE for teaching, we suggest that

incorporating 3D MUVE along with existing e-

Learning environments would generate better

blended learning results for students and teachers.

3 BLENDED LEARNING WITH

3D MUVE – STRATEGIC VIEW

The following abstract model indicates learning

environment approaches and possible technology

applications with a high level perspective. The

model is used to analyse how 3D MUVE fit into the

existing learning environments, and to evaluate

feasible solution stacks to form a productive learning

environment. This model analysis will be considered

for the use case analysis, later in this paper.

Figure 2: High level model to analyze 3D MUVE

integration with learning practices.

The model uniquely identifies three core areas of

learning methods: traditional learning methods, e-

Learning methods and 3D MUVE learning activities.

According to the model, for a productive blended

learning experience, 3D MUVE should be

introduced in a complementary nature to the existing

e-Learning and traditional learning system suites.

Let us briefly discuss typical characteristics and

issues on each of the different combinations that

teachers can practice along with selected system

environments. Moreover, for this analysis, we

presume the individual methods, i.e. traditional

learning or e-learning or 3D MUVE learning alone,

would only provide suboptimal learning experience.

Another important consideration made with this

model analysis is that the role of mobile learning or

commonly known as m-learning. There is no doubt

that the portability and the ubiquity presented by

hand held devices such as PDAs and mobile phones

add a significant value to the existing learning

activities. Many researchers tend to consider it as a

sub category of broad e-Learning application

domain. Lefrere (Lefrere 2009) has examined

activity based scenarios for ubiquitous mobile e-

Learning. In his findings, he concludes that

CSEDU 2011 - 3rd International Conference on Computer Supported Education

prevailing resource constraints of mobile

infrastructure and complex nature of existing e-

Learning use case scenarios have prevented mobile

learning being considered as a broad and

independent learning technology, but a supportive

rich extension for existing e-Learning practices.

Moreover, learning with the mobile device support

does not replace existing learning practices, but it

offers a way to extend the support of learning

outside the classroom, to the conversations and

interactions of everyday life, as a value addition to

e-Learning (Boja and Batagan 2009). We also

understand that considering mobile learning as a part

of e-Learning for this model analysis would not

affect negatively the accuracy of the model, but

provide a summative and straightforward insight on

the arguments being presented.

Most of the present virtual environment

supported learning activities can be seen as

complementary approaches of e-Learning and

traditional learning combinations. Apart from a pure

e-Learning based distance learning activity, all the

other learning practices have traditional learning

methods such as classroom teaching, in person

interactions, practical and laboratory projects,

assessment and feedback. Even though e-Learning

methods provide learning process optimization

through automation and usable content reusing

approaches, it cannot entirely replace traditional

learning activities that require user collaboration and

physical engagement. On the other hand, beyond

video content support, e-Learning does not provide

simulation facilities to streamline 3D aspects in the

learning experience. A learning environment with

Moodle support can be considered as an example

scenario for this category.

Undoubtedly, the traditional learning has

benefited by using 3D MUVE as a supportive tool

for 3D simulations and user engagement. Specially,

when it comes to explaining complex concepts such

as computing algorithms, natural and physical

science phenomena and 3D modelling, 3D MUVE

provide unequal value additions for traditional

learning. Moreover, 3D MUVE can be used as an

alternative simulation tool to train students virtually,

before their actual laboratory experiments. In some

instances, this can be the only possible option due to

various constraints on real experiments. However,

we do not see a comprehensive integration with the

learning processes, as 3D MUVE are merely used as

supportive tools. Isolated learning activities that use

Second Life or Open Simulator virtual regions can

be considered as example scenarios for this.

The combination between 3D MUVE and e-

Learning also show better results, but it misses the

important aspects of traditional learning such as

classroom participation, examination and physical

engagement. The data consistency and content

integration between the two environments have

made this option the most effective out of the three,

yet it is not the optimal scenario. Sloodle (Sloodle

2007) integration between Moodle and Second

Life/OpenSim is the best example for this type.

However, we will further discuss certain

inappropriate use cases designed in Sloodle, which

could be practiced productively with e-Learning

systems than in 3D MUVE.

Therefore, it is understandable that for a

successful learning experience, there should be

complementary facilitation of these three learning

environments. The proposed new blended learning

paradigm facilitates straightforward analysis on the

learning processes and the degree of their

dependence upon e-Learning, traditional and 3D

MUVE domains. Moreover, the finer granular

functionality analysis derived from this abstract

model would help to fine-tune the existing learning

methods and models to avoid potential redundancies

and functional gaps, when systems are in use. For

that, we further analyse effective use cases for 3D

MUVE learning in the context of existing blended

learning environments in the next section.

4 USE CASE ANALYSIS – 3D

MUVE LEARNING

Comprehensive use case analyses on learning with

virtual environments have not been performed in a

larger scale, so far. The main reason for that may be

the intrinsic properties of learning use cases that

directly map with the pedagogical and traditional

learning processes, which have resulted in

researchers considering those as they are for virtual

environments. However, this lack of analytical

understanding on appropriate use cases for a given

learning environment creates difficulties for

integrating 3D MUVE with existing learning

environments. Furthermore, it results in educators to

expect inefficient use cases from 3D MUVE, and

often makes them to practice those meaninglessly.

Table 1 summarises the default user roles in the

Moodle e-Learning environment while indicating the

appropriate corresponding roles from Second Life

and OpenSim 3D MUVE. It shows the abstract user

role definition in 3D MUVE, compared to Moodle

TOWARDS EFFECTIVE BLENDED LEARNING WITH 3D MUVE - An Analysis of Use Case Implementations for 3D

MUVE Learning

or similar e-Learning systems, results in poor

granularity on defining learning use cases in 3D

MUVE. Learning activity management for complex

use cases with distinct roles can be a challenging

task to achieve in 3D MUVE. As a result, a revision

of complex e-Learning scenarios is needed for an

effective learning experience with 3D MUVE

support. Furthermore, access control and permission

models in 3D MUVE are designed for 3D content

and land access (Perera, Allison et al. 2010), which

may not be possible to map directly with access

control models of e-Learning systems. This creates

further discrepancies when users expect exact e-

learning use case behaviours in 3D MUVE.

Table 1: The comparison of default user roles in Moodle

with the 3D MUVE Second Life (SL) and OpenSim (OS).

Moodle Role Description SL OS

Administrator

system

administration

(all courses)

Linden

Labs

System

Owner

Course

creator

create courses,

teach in them

Land

owner /

Resident

user

Land

owner /

Resident

user

Teacher

teach in and

modify assigned

courses

Land

owner /

Resident

user

Land

owner /

Resident

user

Non-editing

teacher

teach in assigned

courses

Resident

user

Resident

user

Student

resource access

and course

participation

Resident

user

Resident

user

Guest observation only Visitors Visitors

Although we can consider all major user roles in

the table 1, due to the limited space, let us consider

only the student role for the use case analysis, here.

In fact, for 3D MUVE, beyond administration tasks

of the system and the virtual environment, most of

the other use cases are common to different roles;

hence the common user role would be the ‘Resident

User’ in the virtual region. Therefore, default student

role is taken as a resident user, and considered the

common use cases available for a resident user in

default, which are compared in the figure 3 with the

Moodle student role.

With the system support for rich text based

content management and integration, e-Learning

environments such as Moodle can incorporate a

diverse set of student activities as shown in the

figure 3. Moreover, these activities can be extended

easily with additional functions to form

comprehensive end-to-end learning processes. On

the other hand, 3D MUVE user activities are more

abstract and emphasis on 3D simulation and

dynamic nature through programming than advanced

textual features. The 3D MUVE student use cases

shown in figure 3, indicates this abstract nature and

gives a clear view on how difficult to achieve e-

Learning use cases as they are, in 3D MUVE.

This validates the proposed blended learning

model and the arguments, as 3D MUVE should be

incorporated with its competent learning use cases,

whilst e-Learning and traditional learning practices

being considered for the rest. Moreover,

inappropriate use case integration between e-

Learning systems and 3D MUVE can result to

inconsistent data and critical security issues on the

role based access control. The next section

elaborates the use case mapping challenges

considering a set of unproductive learning feature

implementations in 3D MUVE through the one-to-

one mapping of e-Learning use cases.

Figure 3: The comparison of learning use cases for the

student role in Moodle (version 1.9.9) and generic 3D

MUVE. (UML 2.0 use case standard).

5 CASE STUDY I – USE CASE

MAPPING CHALLENGES

Use case mappings can be seen in two types. The

first type accounts for the popular activities of using

CSEDU 2011 - 3rd International Conference on Computer Supported Education

3D MUVE for trivial learning activities such as

mere gatherings or to impose 3D flavour on existing

2D learning contents. These activities do not induce

additional inefficiencies to learning process, but add

variety and dynamism, although the learning

activities may be not practiced to the optimum

potential. On the other hand, the second type of

learning use case mapping seems crucial for learning

environments since it can introduce a set of

complimentary uses with 3D MUVE instead of

major Web based e-Learning activities. Although

these practices are becoming popular, they are yet to

provide productive learning use case

implementations for a complete replace of selected

2D Web tasks.

For this case study, we have installed and tested

a prototype learning environment. We have used

Moodle (version 1.99) as the e-Learning platform,

Sloodle (version 1.0) and OpenSim (version 0.68) as

the 3D MUVE. Another important fact with this

arrangement is all these systems are free and open

source; the most complete open solution available

present for academia to work with on this regard.

Therefore, this system suite based learning activities

should be thoroughly studied in order to optimise the

proposed blended learning with 3D MUVE.

Sloodle learning features include 11 activities to

map selected Moodle activities such as chat, forum,

glossary, choice, content display and limited support

of quiz and 3D content submissions for assignments.

A summary of main Sloodle tools for learning, their

functional descriptions along with the corresponding

Moodle task, is mentioned in table 2.

Table 2: The comparison of default user roles in Moodle

with the 3D MUVE Second Life (SL) and OpenSim (OS).

Sloodle tool

Functionality description

Moodle

Webintercom

Synchronize chat Chat

Presenter

Display media files

Content

display

Toolbar

Blog, Gestures, Avatar list

display

Forum

Quiz chair

MCQ question support Quiz

Pile on quiz

MCQ question support Quiz

Prim drop

Submit prim for

assignment in Moodle

Assignment

MetaGloss

Access glossary in Moodle Glossary

Sloodle Choice

Synchronize choice Choice

Vending machine

Distribute SL objects --

Awardsystem

Connects Sloodle points

with Moodle Gradebook

Gradebook

Picture Gloss

Connects texture from

glossary to prim

Glossary

Synchronised user communications and Moodle

content display in 3D MUVE are rational features

that add value to learning. However, using 3D

Figure 4: Publishing rich text entries through direct access

of Moodle without Sloodle support.

MUVE chat channels to publish student

compositions in Moodle forum, glossary and wiki,

can be a question as those entries are supposed to

feature rich text and content, which cannot be

supported through 3D MUVE chat or IM channels,

at present. Furthermore, asking students to

participate in quizzes, assignments and text based

learning activities in 3D MUVE instead of Moodle

can introduce further difficulties to student work. In

most of the instances, students require re-login to

Moodle afterwards of their initial submission, to

enrich the entries that have been done through raw

chat interfaces while they were inside 3D MUVE.

By accessing Moodle through in-world media

browser directly would allow a certain level of rich

text support as shown in Figure 4. However, due to

JavaScript support issues with the present in-world

media browsers, students can try external browsers

of their systems for advanced text compositions

while being inside the 3D MUVE.

Another important fact to consider is the

accurate map of user collaboration facilities

provided in 3D MUVE with the e-Learning

environment. Specially, Web based learning

activities by their nature are not designed for rich

collaboration in real time, but a user session based

individual browsing. Some argue that confining

certain 3D virtual world learning activities to the in-

world browser, would eventually hinder the

advantages of 3D MUVE for collaborative learning.

A collaborative web browsing (‘co-browsing’) tool

support is being developed as an extension to the

Sloodle system (Crowe and Livingstone 2009).

However, it is too early to evaluate such a tool for

seamless web browsing on serious learning activities

with 3D MUVE. For the time being, accessing

textual content through the in-world browser or

external browsers (Firefox, Safari, IE, etc.) would be

advised, and it provides better results for rich and

comprehensive composition tasks (Figures, 5 and 6).

Moreover, Sloodle mediation only supports MCQ

TOWARDS EFFECTIVE BLENDED LEARNING WITH 3D MUVE - An Analysis of Use Case Implementations for 3D

MUVE Learning

questions! Therefore, accessing Moodle directly in

2D form is essential for formal assessment and

feedback.

Figure 5: Attempting a quiz, published on a Moodle

course, in OpenSim with Sloodle mediation through in-

world Note-cards.

Figure 6: Attempting the same quiz mentioned in fig. 5,

through the OpenSim in-world media browser. Here the

avatar accesses the quiz directly without Sloodle support.

However, when it comes to mediating the 3D

content ownerships, their associations and avatar

identity registration with Moodle course modules,

Sloodle does reasonably well through its tools. ‘Prim

drop’ and ‘Vending machine’ are important tools for

this purpose and unarguably implements essential

use cases for 3D MUVE integration with e-

Learning. Moreover, ‘Registration Booth’, ‘Access

Checker’ and ‘Login Zone’ functions are effective

for platform integration with accurate identity and

access control mapping between 3D MUVE and

Moodle.

Therefore, trying to achieve all learning use

cases of e-Learning systems in 3D MUVE is not

advised for serious learning requirements. With the

case study, we have identified that the textual and

2D based learning content and activities are not yet

fully supported for effective use in 3D MUVE.

Educationalists should rationally decide appropriate

learning use case implementations according to the

learning domain requirements. We have further

studied on such a specific implementation of a

domain requirement and will be discussed in the

next section. Moreover, students should be

encouraged to use the e-Learning environment for its

competent functions while the 3D MUVE for its

best, in a mutually independent manner. The system

infrastructure should ensure the seamless data

integration between the environments underneath for

a smooth learning experience, without forceful

implementations of inappropriate practices.

6 CASE STUDY II – A DOMAIN

SPECIFIC COMPARISON

For this, we have selected to compare two distinctive

implementations of the link state protocol OSPF

(Open Short Path First) simulations as learning aids.

OSPF uses Dijkstra’s Shortest Path algorithm as its

mechanism to build and calculate the distances to all

known destinations. Importantly, understanding

Dijkstra’s Shortest Path algorithm can be a

challenging task for an unaided student due to its

complex nature. Like many algorithms, it can leave

some students bored and disengaged, so the goal was

to demonstrate a user-friendly learning resource to

help students who would otherwise lose interest.

The first attempt was to implement a Web based

public accessible learning resource with animation

support to indicate the algorithm behaviour. Also

this implementation facilitated a certain level of

interactivity with feedback to self assess user’s

understanding on the algorithm behaviour. A

working instance of the Dijkstra’s algorithm

animation with highlighting the relevant step is

shown in Figure 7.

Figure 7: An operational instance of the interactive Web

based OSPF algorithm simulation learning resource.

CSEDU 2011 - 3rd International Conference on Computer Supported Education

Although this Web based OSPF learning

resource was a better alternative to the static textual

explanations of OSPF algorithm, we have identified

several obstacles associated with its domain

limitations, which are challenging to overcome for

an engaged and collaborative learning participation.

Moreover, the Web based OSPF learning aid could

only provide the dynamism within a pre arranged set

of graph topologies to indicate the algorithm

behaviour. If a student wants to practice own

network topology with varying constraints to

examine the algorithm, it is not possible with this.

However, with the intrinsic support for textual

content of the platform, this version of OSPF

animation provides usable text based learning

content and dynamic activities such as formatted

routing table, node – arc summary at each step and

textual user input and feedback for self assessments.

It also provides additional resources (external) for an

enthusiastic student to refer and further study.

Alternatively, an OSPF simulation learning

resource has been developed in a locally installed

OpenSim island (McCaffery, Miller et al. 2010).

The Web based OSPF learning resource

demonstration has been replicated with the 3D

content and scripting for the required animation (Fig

8). Further extensions and attractive learning use

cases have been associated with this learning aid

beyond the basic preset algorithm simulation.

Figure 8: An avatar observes the Dijkstra’s algorithm

simulation example.

The implementation of OSPF algorithm

simulation in OpenSim has several additional

functions when compared with its Web based

predecessor. Importantly, one of the key use cases of

this learning resource is the truly dynamic network

topology creation at students’ preference. Student

avatars can create various complex network

architectures using routers and end nodes as they

wish (Fig. 9). This is an excellent opportunity for

students to reflect on what they have learnt and to

self examine individual concerns to overcome

liminal spaces related to a threshold concept of their

learning. A threshold concept lets students into

transformational liminal states(Meyer and Land

2005). Indeed a complex algorithm related concept

such as OSPF can be a threshold concept for some

students, in which if we could not provide adequate

dynamic learning aids as liminal states emerge when

students are in the process of understanding, they

easily tend to mimicry the algorithm flow shown in

the static demonstrations, without a thorough

knowledge. While creating their own desired

network topologies, it is obvious that students think

a shortest path by reflecting their knowledge.

However, instantaneous observation of the correct

answer through intuitive simulation would help

students to reinforce their learning accurately. Had

they anticipated an erroneous answer, their curiosity

would let them reinvestigate their answers once the

simulation indicated the correct path of the packet

flow, as shown in the figure 10.

Figure 9: A student avatar creating a dynamic network

topology with routers and end nodes.

Figure 10: Observing the packet flow in the shortest path

of the user created network topology.

Another important advantage with this

implementation is the extensive facilitation for

collaborative group activities. Certainly, it is due to

the intrinsic trait of the 3D MUVE platforms; the

TOWARDS EFFECTIVE BLENDED LEARNING WITH 3D MUVE - An Analysis of Use Case Implementations for 3D

MUVE Learning

learning resources should be carefully designed to

exploit it, however. Students’ collaboration on

network topology creation, setting various weights

and resources, and observing and commenting

others’ use of the learning aid are new use cases that

have been introduced compared to 2D Web version.

In fact, these use cases of learner collaboration are

essential for a broader understanding of a complex

concept, which would otherwise be limited with

individual insights. User collaboration occurs

through interaction between learners while

interacting with the 3D MUVE, through the learning

interface that provides supports rather than barriers

to learning (Girvan and Savage 2010). Students have

been given the opportunity to use 3D MUVE

communication channels (chat, voice, IM) to share

their ideas while being engaged with their learning.

This was not possible with the previous Web based

animation, either.

The main emphasis given on this OSPF

simulation implementation was to depict the

associated concepts in 3D visuals as much as

possible instead of the textual alternatives. Although

the learning experience could be implemented on

other platforms without 3D MUVE support, the

learner experience would be lost, and users feel

contrived (Girvan and Savage 2010). Therefore, a

reasonable amount of textual content and activities

have been appropriately redesigned and mapped to

3D content to exploit the 3D MUVE benefits to the

fullest. By doing so, however, certain learning use

cases that are primarily associated with the textual

contents and could not completely mapped to 3D

content, seem to be deprived when observed along

with the rich 3D visuals. Using 3D MUVE raw text

channels to display routing information or algorithm

simulation updates may not be aesthetic without rich

formatting. Indeed, this strengthens our argument of

using complement systems for a blended learning on

what they are competent of. An appropriately

designed Web based e-Learning resource to provide

rich textual content and user activities would

complement the 3D MUVE with sufficient

integration. Moreover, as we discussed in the

previous section, such a resource could be accessed

interactively through the in-world or external

browsers, while participating in the 3D MUVE

learning activities.

Another challenge, with 3D MUVE learning

resources, is the provision of external resources of

the same calibre for further studies. Being a novel

and growing technology for learning, 3D MUVE do

not yet possess a reasonable amount of 3D learning

resources to be referred for a particular area of

study. Even if there are 3D contents available from

external sources, accessing and execution of those

content objects is not a straightforward activity, as

we access Web based public resources. This is

primarily due to the heterogeneous content

ownership management policies and platform

constraints. However, we can expect for a brighter

future as 3D MUVE systems further develop and are

widely used in blended learning environments.

7 CONCLUSIONS

AND FUTURE WORK

This paper has rationalized the use case issues

associated with learning in 3D virtual worlds, when

users expect identical use cases as they practice with

e-Learning activities. Either the situational

approaches for utilizing 3D MUVE for learning, or

forceful integrations of inappropriate use cases of e-

Learning systems with 3D MUVE, would not

furnish sustainable solutions; this paper has

introduced a strategic model to analyze these issues

considering prime aspects. The analysis on use case

implementations and selected case studies in this

research would only guide the pathway, but further

research is encouraged for standardising and

applying productive use cases for various blended

learning requirements with 3D MUVE.

3D virtual worlds have a great potential for

engaging students in innovative, immersive learning

environments. Identifying appropriate use cases for

learning in 3D MUVE will facilitate the future work

of this research. With this research, we are looking

forward to provide comprehensive security

management policies for generic learning

requirements in 3D MUVE. For a sustainable

security management policy definition on 3D

MUVE learning, it is essential to consider every

associated main platform of blended learning. Not

only it allows us to develop policy models for 3D

MUVE learning in a holistic manner, but also

provides an efficient set of complimentary learning

use cases that has been already filtered for

redundancies from other platforms. The proposed

security policy models will then be implemented as

prototypes to evaluate and optimize for large scale

deployments. These will be done at the application

level, independent of the underlying platform

constraints to ensure seamless customization and

reuse, as required.

CSEDU 2011 - 3rd International Conference on Computer Supported Education

ACKNOWLEDGEMENTS

This research is supported by the UK

Commonwealth Scholarship programme and the

Scottish Informatics and Computer Science Alliance

(SICSA). Second Life region rental was supported in

part by the University of St Andrews Fund for

Innovations in Learning, Teaching and Assessment

(FILTA). The Higher Education Academy for

Information and Computer Sciences (HEA ICS)

supported part of the work on OpenSim.

REFERENCES

Allison, C., A. Miller, et al. (2008). "Exploratory Learning

for Computer Networking." Advances in Web Based

Learning – ICWL 2007: 331-342.

Allison, C., A. Miller, et al. (2010). Educationally

Enhanced Virtual Worlds. 40th ASEE/IEEE Frontiers

in Education Conference (FIE), IEEE.

Boja, C. and L. Batagan (2009). "Analysis of m-Learning

Applications Quality." WSEAS Transactions on

Computers 8(5): 767-777.

Brenton, S. (2009). E-learning - an introduction. A

Handbook for Teaching and Learning in Higher

Education. H. Fry, S. Ketteridge and S. Marshall: 85-

98.

Crowe, M. and D. Livingstone (2009). Collaborative Web-

Browsing in a Virtual World. In Proceedings of 2009

Conference in Games and Virtual Worlds for Serious

Applications, VS-GAMES, IEEE.

Dalgarno, B., A. G. Bishop, et al. (2009). "Effectiveness

of a Virtual Laboratory as a preparatory resource for

Distance Education chemistry students." Computers &

Education 53(3): 863-865.

Girvan, C. and T. Savage (2010). "Identifying an

appropriate pedagogy for virtual worlds: A Communal

Constructivism case study." Computers & Education

55: 342-349.

Graf, S. and Kinshuk (2009). Advanced Adaptivity in

Learning Management Systems by Considering

Learning Styles. Proceedings of the 2009

IEEE/WIC/ACM International Joint Conference on

Web Intelligence and Intelligent Agent Technology -

Volume 03, IEEE Computer Society.

K. Getchell, J. Nicoll, et al. (2007). "Evaluating

exploratory learning in LAVA." in Proceedings of the

sixth conference on IASTED International Conference

Web-Based Education 2.

Kirriemuir, J. (2010). "Virtual world activity in UK

universities and colleges - Virtual teaching in

uncertain times, Snapshot #8 of virtual world use in

UK Higher and Further Education." Virtual World

Watch, Eduserv Snapshot #8(Spring 2010): 1-45.

Kolb, D. A., R. E. Boyatzis, et al. (2001). Experiential

Learning Theory: Previous Research and New

Directions. in Perspectives on Thinking, Learning and

Cognitive Styles. J. Sternberg and L. F. Zhang,

Lawrence Erlbaum: Mahwah: 227.

Lefrere, P. (2009). "Activity-based scenarios for and

approaches to ubiquitous e-Learning." Personal and

Ubiquitous Computing 13(3): 219-227.

Linden_Labs. (2003). "Second Life." from

www.secondlife.com [cited November 2009].

McCaffery, J., A. Miller, et al. (2010). Visualising

Routing Algorithms in Virtual Worlds. in Proceedings

of the 11th PGNet symposium, Liverpool.

McGill, T. J. and J. E. Klobas (2009). "A task-technology

fit view of learning management system impact."

Comput. Educ. 52(2): 496-508.

Messinger, P. R., E. Stroulia, et al. (2009). "Virtual worlds

- past, present, and future: New directions in social

computing." Decis. Support Syst. 47(3): 204-228.

Meyer, J. H. F. and R. Land (2005). "Threshold concepts

and troublesome knowledge: Epistemological

considerations and a conceptual framework for

teaching and learning." Higher Education 49(3): 373-

388.

Moodle. (2004). "Moodle." Retrieved February 2010,

from http://moodle.org/.

Oliver, I. A., A. H. D. Miller, et al. (2010). Virtual worlds,

real traffic: interaction and adaptation. The first annual

ACM SIGMM conference on Multimedia systems,

MMSys'10. Phoenix, Arizona, USA, ACM: 305-316.

Perera, I., C. Allison, et al. (2010). Secure Learning in 3

Dimensional Multi User Virtual Environments –

Challenges to Overcome. In Proceedings of the 11th

PGNet symposium, Liverpool.

Perera, I., C. Allison, et al. (2009). Towards Successful 3D

Virtual Learning – A Case Study on Teaching Human

Computer Interaction. 4th International Conference

for Internet Technology and Secured Transactions

(ICITST-2009). London: 159-164.

Perera, I., C. Allison, et al. (2010). "Managed Learning in

3D Multi User Virtual Environments." International

Journal of Digital Society 1(4): 323-332.

Sheldon, N., E. Girard, et al. (2003). The effect of latency

on user performance in Warcraft III. NetGames ’03:

Proceedings of the 2nd workshop on Network and

system support for games.

Sloodle. (2007). "Sloodle." from http://www.sloodle.org/

moodle/.

Sturgeon, T., C. Allison, et al. (2009). "802.11 wireless

experiments in a virtual world." SIGCSE Bull. 41(3):

85-89.

Teo, C. B. and R. K. L. Gay (2006). "A knowledge-driven

model to personalize e-learning." J. Educ. Resour.

Comput. 6(1): 3.

The_Open_Simulator_Project. (2007). "Open Simulator ".

Weippl, E. R. (2005). Security in E-Learning, Springer.

Wu, J., R. D. Tennyson, et al. (2010). "A study of student

satisfaction in a blended e-learning system

environment." Computers & Education 55(1): 155-

164.

TOWARDS EFFECTIVE BLENDED LEARNING WITH 3D MUVE - An Analysis of Use Case Implementations for 3D

MUVE Learning