NeuroK: A Collaborative e-Learning Platform based on Pedagogical

Principles from Neuroscience

Fernando Calle-Alonso

, Agustín Cuenca-Guevara

, Daniel de la Mata Lara

Jesús M. Sánchez-Gómez

, Miguel A. Vega-Rodríguez

and Carlos J. Pérez Sánchez

Research & Development Department, ASPgems SL, Spain

Catedra ASPgems, University of Extremadura, Spain

Department of Technologies of Computers & Communications, University of Extremadura, Spain

Department of Mathematics, University of Extremadura, Spain

Keywords: Educational Data Mining, e-Learning, Learning Analytics, Learning Management Systems, Neurodidactics,

Computer-Supported Collaborative Learning, Social Learning, Social Networks.

Abstract: The use of online education platforms has grown extensively and most education centers and companies use

them for their learning programs. Although technology has changed the learning environment, the

pedagogical model has mostly remained the same as it was many years ago. Therefore, another education

paradigm should arrive to online platforms in a generalized way. In this paper, NeuroK is presented as a

new e-Learning platform leveraging the latest technologies and, moreover, implementing new tools that

support pedagogical principles from neuroscience. While most of traditional platforms focus on content,

content management and applying teacher-centred methodologies (everything goes through the teacher),

NeuroK focuses on students, and uses collaborative learning, motivational processes and a “learning by

doing” perspective to achieve a long-term relevant learning. The proposed NeuroK framework describes the

already implemented tools and the new ones to be included in next versions. An active R&D process allows

new methodologies from the fields of Learning Analytics, Data Mining and Social Learning to be proposed

and implemented. The main contribution of this platform is to deploy a significant improvement in the e-

Learning process based on a neurodidactics approach and data analysis research results.

1 INTRODUCTION

During the last decade, there has been significant

changes in the educational field, especially

strengthened by new technological developments.

Tools as e-learning platforms or Learning

Management Systems (LMS) are now common for

educational institutions and companies. Indeed,

NMC Horizon Report (Johnson et al., 2014) pointed

out that online, blended and collaborative learning

will be widely adopted in the short term, and that

learning analytics environments, using predictive

modelling, will be adopted in the mid-term.

The main advances happened around online

training frameworks, enabling people to learn almost

everything remotely and costless. This new concept

of learning is called e-learning. It is defined as a

remote way of learning which facilitates and makes

more flexible the teaching/learning process, adapting

it to the skills, needs and availability of each

different student. E-learning saves time, money,

materials, and resources by keeping everything

online. As technology continues to push e-learning

forward, education and training will become more

popular and reach wider audiences. But it is not only

the technology alone what makes e-learning so

important (Haythornthwaite et al., 2016), but also

the inclusion of innovations from different fields

such as the use of educational gaming and

gamification (De Marcos et al., 2016), big data

techniques (Anshari et al., 2016), collaborative

learning (Masud, 2016), among others.

There are many e-learning platforms allowing

the communication and interaction between

teachers, students and the study contents. They

include different learning systems such as instructor-

led learning, e-books, tests, video tutorials and so

on. And they can be served from web browsers,

mobile apps or desktop applications mainly (Fallon

and Brown, 2016). Also, e-learning platforms can be

550

Calle-Alonso, F., Cuenca-Guevara, A., Lara, D., Sánchez-Gómez, J., Vega-Rodríguez, M. and Sánchez, C.

NeuroK: A Collaborative e-Learning Platform based on Pedagogical Principles from Neuroscience.

DOI: 10.5220/0006378705500555

In Proceedings of the 9th International Conference on Computer Supported Education (CSEDU 2017) - Volume 1, pages 550-555

ISBN: 978-989-758-239-4

extended with administrative resources, and these

systems are called Learning Management Systems.

In this paper the interest will be focused on web-

based applications.

Some of the most recognized platforms in this

field are commercial as Blackboard, WebCT,

OSMedia, Saba, eCollege, Fronter, SidWeb, E-

ducativa or Catedr@, among others. Nevertheless,

the market of open source solutions has grown a lot

in the last years and platforms as ATutor, Dokeos,

Claroline, dotLRN, Moodle, Ganesha, ILIAS or

Sakai are as known and useful as the commercial

ones. In some cases, offering even more

functionalities than commercial solutions.

Massive Open Online Courses (MOOCS) have

also experienced a great improvement with

platforms as Udacity, Coursera, Udemy, edX,

Ecaths, Wiziq or Edmodo. But this kind of platforms

are more limited than the general ones mentioned

before. They are focused only on video-based

lessons and tests/exercises, with the purpose to

obtain badges demonstrating some kind of

knowledge.

Analyzing the market for all these applications, it

can be highlighted that more than 90% of the US

higher education institutions are using LMS’s (De

Smet et al., 2016). But OECD reports that

universities use LMS for administrative and

communicative purposes, instead of offering new

pedagogical ways of teaching (Dalsgaard, 2006). So,

there is still a lot to research to improve learning

processes, experiences and students’ satisfaction.

One of the main problems for online education is

the attrition and dropout rate of students. Many

researchers are focusing on this subject (Monteiro et

al., 2016). The issue is that students motivation fall

down after the first two weeks and they leave the

course. Also, the completion rates are too small,

especially for MOOCs where the completion rate

almost always falls between 2% and 5%, and in few

occasions exceeds 10% (Jordan, 2015). Moreover,

the information learnt by the students is forgotten in

a short period of time (Bacon & Stewart, 2006).

The issue underneath all the advances in

education is that technology has changed a lot, but

the way of teaching remains the same. Platforms

such as Moodle, Blackboard or even MOOCs have

just displaced traditional education from classrooms

to websites, but without changing the way of

teaching knowledge.

Neurodidactics (Anastasia 2016; Sabitzer, 2011),

also called brain-based teaching, tries to overcome

classical education with some new tools, and starting

to teach from a different point of view. “Knowledge

cannot be transferred, it must be newly created in

the brain of each student” (Roth, 2004). According

to this, neurodidactics proposes a learner centred

education. It avoids just giving materials to study,

instead of that, it offers examples, associations and

linking information to build knowledge. The reason

is that just giving new information, the long term

memory is not activated. The more often

associations and links are used and practiced, the

better the knowledge will be acquired (Westerhoff,

2010). The magic recipe to learn with the long term

memory is motivation (Di Ges

ù & Seminara, 2012;

Rivas, 2009), which gives the students the incentive

to learn and the ability to concentrate (Hermann,

2009 and Harandi, 2015). There are several ways to

achieve this goal, for example, using gamification,

learning by doing, flipped classroom, cooperative

learning, small group learning, peer tutoring, and the

use of real life challenges, among others (Hattie,

2009).

Most of the current e-learning platforms don’t

implement any of these tools to manage motivation

as a keystone of the education, nor follow the

neurodidactics perspective. Therefore, there is a

research opportunity in the e-learning field. The

overall objective of this paper is to address on-line

education from the perspective of the new trends in

e-learning and the principles of the neurodidactics

provided by the proposed new platform NeuroK

(https://neurok.es/en). This could provide an

important step in changing traditional online

learning paradigm, allowing to learn in a

collaborative and cooperative way by means of an

attractive and motivating social network

environment. In addition to the already implemented

tools, there are many others that will be incorporated

in the next versions. These tools are based on new

research lines in the field of Learning Analytics and

Social Learning that match the neurodidactics

approach.

2 NeuroK FRAMEWORK

After many successful experiences in software

development, ASPgems started collaborating with

some Universities trying to develop an innovative e-

learning solution. NeuroK platform was built based

on the latest methodological advances of

neurodidactics (Sousa, 2014; Edelenbosch et al.

2015). As other social networks, it pays special

attention to debate, new opinions and engaging

features as notifications, hashtags or mentions. All

NeuroK: A Collaborative e-Learning Platform based on Pedagogical Principles from Neuroscience

551

of them encouraging to learn by motivation (Waelti

et al. 2001).

NeuroK focuses on the student, not on the

content or the teacher like traditional learning

paradigms do. The students interact and collaborate

to build the information they will learn. They

choose, analyze, compare materials, and argue about

the proposed learning units. They also peer review

the activities of their classmates, and give opinions

about them. The teachers are not expendables, as

other online platforms intend. They are not eligible

for expense reduction. In NeuroK, the teacher has a

fundamental role as a “learning guru”, providing the

course criterion, the main guidelines to follow,

managing the information provided by the students,

and personally supporting them. He tries to achieve

the goals of the course, following the path the

students set up, and redirecting according to his

experience.

The NeuroK environment is a social network. It

gives the teachers the tools to teach by doing. The

knowledge is built by all the participants of the

course in a continuous way, thanks to their

comments, favorites, reviews, documents sharing,

cooperative problem solving and much more. This

kind of social learning boosts their motivation and

makes easier reaching a long term learning.

The main features of NeuroK are oriented to

serve the three C’s: Communication (to share the

knowledge), Community (to connect students and

build groups) and Cooperation (to propose ideas and

solve problems with cooperative learning).

NeuroK is a cloud platform always accessible

from any device. It is very easy to use, and the

learning curve is very soft, according to the students’

opinions who have already tested it. It supports from

small groups to a large number of students. Many

different kinds of documents can be published from

Youtube, Vimeo, Drive, Dropbox, or local materials

from students and teachers.

Every course starts on the home page (see Fig.

1). The central space is devoted to the social activity,

the same as Facebook or Twitter do, denoting the

importance given to collaboration and interaction

among students. The students follow the teacher and

other students who they think are especially

relevant. The social interactions appearing in the

central timeline can be filtered including everything,

just the people they are following, or just the

teacher. This collaborative way of creating contents

by publishing documents, debating, solving

problems and peer based rating, strictly follows the

neurodidactics principles, and it is very attractive

and motivating to the students (Elsenbaumer, 2013).

They see NeuroK as another social network, easy to

use, engaging, offering new materials and comments

all the time, and challenging their skills to solve

problems. Students are intended to learn with

motivation, like if they were playing.

Figure 1: NeuroK home page.

The main actions to take in NeuroK are accessed

from the home page. At the top, the one-on-one

contact and the notification buttons can be found.

Just below, there is a small summary about the user

participation on the course. In the central panel the

timeline shows the latest publications, but the user

can also filter by only his contributions, favourites

(marked before), or the noticeboard

(communications from the platform or the teacher).

The student can directly read, comment, recommend

or mark as favorite any publication. It is very easy

and intuitive to interact giving opinions, rating

others comments, or even peer reviewing activities

from other classmates.

In the left panel the access to Learning Units can

be found. It includes the proposed contents to

generate a debate, Learning Activities and practical

questions and problems about the contents of each

learning unit. Everything to “learn by doing” both in

an individual or collaborative way.

Figure 2: Global statistics.

Some other important options in the left panel are

Library (with contents shared by the teacher), Travel

CSEDU 2017 - 9th International Conference on Computer Supported Education

552

to the past (if the student leaves the course for some

time, then he can restart from that point and follow

the evolution of the posts, comments, etc), Statistics

(more detailed global and individual statistics) and

Video call (allowing a group of students to contact

by video conference).

The global statistics table shows the participation

level of the students in a color scale (Fig. 2). It gives

the teacher quick information about the development

of the course in general and lets him know which

students are at risk of abandonment. With the

individual statistics (Fig. 3), the teacher can observe

the students at risk, and make some engaging actions

to try to motivate them to continue participating in

the course. Some actions to re-engage students can

be performed based on these data. Some actions can

be automated (with rules e.g., sending an automatic

e-mail) and others can be done directly by the

teacher.

The objective of these descriptive measures is to

offer all the information available to the teacher, so

he/she follows the students, motivates them,

customizes the contents, and redirects the courses.

As we mentioned before, dropping out is one of the

main concerns for e-learning. To reduce the dropout

rate, some new measures are being developed, as the

use of percentiles, the estimation of probabilities of

abandonment or the evaluation of a social index to

detect influencers. Other new tools have also been

researched and/or developed at Cátedra ASPgems

(University of Extremadura), i.e.: word clouds,

multi-text summarization and social graphs (see

Section 3). They will be used to evaluate students,

identify possible dropouts, and to direct the course in

case that data show the learning units are deviating

from their objectives.

Besides, other features such as gamification have

already been implemented. The students earn points

with the actions they do in the platform. Then, a

ranking by points is shown at the global statistics.

This motivates them to earn more points

participating in the course. It could be possible to

expand this feature to include badges.

In conclusion, NeuroK not only has implemented

tools supporting neurodidactics theories, but also it

obtains very interesting information by using

learning analytics to follow and evaluate students in

real-time. With the integration of new educational

tools, a robust e-learning platform is built, leaving

behind the traditional ones just focused on contents.

Figure 3: Individual statistics.

NeuroK: A Collaborative e-Learning Platform based on Pedagogical Principles from Neuroscience

553

3 NEW EDUCATIONAL TOOLS

There are some educational tools that will be

considered for inclusion in the NeuroK platform in

next versions. Specifically, one module for natural

language processing and another one for social

network analysis are under implementation or under

a research phase.

The first module comprises two different, but

related functionalities. On the one hand, a word or

tag cloud tool has been implemented (see, e.g.,

Resendes et al. 2015). The objective is to provide a

fast and visual representation of the contents

provided by the students within the different

learning units. After considering all the messages

provided by the students in a certain learning unit,

the instructor will be able to analyze if they are

using the right concepts in their comments. Several

metrics to measure the deviation from the theoretical

situation defined by the instructor have been also

implemented. On the other hand, an automatic multi-

document summarization approach is under

research. Multi-document summarization consists in

the extraction from multiple texts about the same

topic. The problem consists in a multi-objective

optimization to obtain the maximum coverage with

the minimum redundancy (see, e.g., Saleh and

Kadhim, 2016). In this case, the multiple texts would

be the messages of the students in a learning unit

about a concrete topic. This allows the instructor to

analyze both, the comments in a learning unit and

the contribution of each particular student to the

summary. Multi-document summarization is a

current research line with a great potential for this

kind of online platforms. Both the word cloud and

the multi-document summarization tools share some

aspects of the natural language pre-processing step.

The second module is based on social network

analysis applied to NeuroK software. Nowadays,

social network analytics is playing an important role

in online platforms (see, e.g. Buckingham and

Ferguson, 2012). A lot of important information can

be extracted from the relationships among the

members of the social network. Firstly, a good

graphical implementation should be displayed,

followed by a number of metrics. These graph and

metrics should represent the relative importance of

each student in the social network based on his/her

comments, observations or ratings, providing a great

information about the structure of the network and

how the students relate. This would help in the

identification of influencers or key players, who may

be good knowledge brokers. Identification of a set of

key players in a given social network is of great

interest in this context. Up to now, most of the used

algorithms for this task are based on single

objective, however, in this case, it is necessary to

find a set of key players which can perform well

with respect to multiple objectives of interest (see,

e.g., Gunasekara et al., 2015). There is much room

for improvement in this area.

4 CONCLUSIONS AND FUTURE

WORK

NeuroK e-learning framework has been presented. It

is an innovative neurodidactics-based platform.

NeuroK offers teachers the essential tools to follow,

guide and evaluate students and, most important, to

motivate them achieving an effective long-term

learning.

It is not a MOOC or a traditional e-learning

framework focused on contents. NeuroK is, instead,

a social network focused on the students which lets

them learn by doing activities in a collaborative way

and not in a memorizing-based approach. It has

innovative features such as gamification, learning

analytics, peer reviewing, content analysis, social

network interface, travel to the past, and a

collaborative learning perspective, among others.

NeuroK has already been tested in relevant

institutions with users such as Universidad Rey Juan

Carlos (Spain), Universidad Libre de Música de

Guadalajara (Mexico), Niuco, Banco Santander or

Catenon Multinational. More tests, including

learning analytics data analysis, will be driven at the

University of Extremadura. Some experimental

results to try to demonstrate the efficacy of this

approach will be obtained during the next months.

Now, Cátedra ASPgems (University of

Extremadura) leads the R&D actions in the fields of

Learning Analytics, Data Mining and Social

Learning. The future work includes researching and

implementing new methodologies, including Big

Data techniques to analyze massive amounts of data

coming from the platform. The results will allow to

understand what happens during the educational

process and to predict what will occur in the future.

This will give to the teachers a deeper knowledge of

the students behaviour (what they know, what they

need and what they are going to do next), allowing

to act in advance by improving the learning process

and completion rate.

CSEDU 2017 - 9th International Conference on Computer Supported Education

554

ACKNOWLEDGEMENTS

This research has been supported by GR15106 and

GR15011 projects (Gobierno de Extremadura,

Spain), Cátedra ASPgems (Universidad de

Extremadura and ASPgems SL, Spain), and

European Regional Development Funds (European

Union).

REFERENCES

Anastasia, C. 2016. The Social Framework of Learning

via Neurodidactics. Creative Education, 7(15), 2175.

Anshari, M., Alas, Y., & Guan, L. S. 2016. Developing

online learning resources: Big data, social networks,

and cloud computing to support pervasive knowledge.

Education and Information Technologies, 21(6), 1663-

1677.

Bacon, D. R., & Stewart, K. A. 2006. How fast do

students forget what they learn in consumer behavior?

A longitudinal study. Journal of Marketing Education,

28(3), 181-192.

Buckingham Shum, S.; Ferguson, R. 2012. Social

Learning Analytics. Educational Technology &

Society, 15 (3), 3-26.

Dalsgaard, C. 2006. Social software: E-learning beyond

learning management systems. European Journal of

Open, Distance and E-Learning, 9(2).

De-Marcos, L., Garcia-Lopez, E., & Garcia-Cabot, A.

2016. On the effectiveness of game-like and social

approaches in learning: Comparing educational

gaming, gamification & social networking. Computers

& Education, 95, 99-113.

De Smet, C., Valcke, M., Schellens, T., De Wever, B., &

Vanderlinde, R. 2016. A Qualitative Study on

Learning and Teaching With Learning Paths in a

Learning Management System. JSSE-Journal of

Social Science Education, 15(1), 27-37.

Di Gesù, F., & Seminara, A. 2012. Neurodidáctica y la

implicación de emociones en el aprendizaje. Lynx:

Panorámica de estudios lingüísticos, 11, 5-39.

Edelenbosch, R., Kupper, F., Krabbendam, L., & Broerse,

J. E. 2015. Brain-Based Learning and Educational

Neuroscience: Boundary Work. Mind, Brain, and

Education, 9(1), 40-49.

Elsenbaumer, S. (2013). Neurodidactics in Practice. A

Practical Approach to Introducing Informatics into a

Primary School in a Brain-based Way. Diploma

Thesis, Alpen-Adria-Universität Klagenfurt.

Fallon, C., & Brown, S. 2016. E-learning standards: a

guide to purchasing, developing, and deploying

standards-conformant e-learning. CRC Press.

Gunasekara, R. C.; Mehrotra, K.; Mohan, C. K. (2015)

Multi-Objective Optimization to Identify Key Players

in Large Social Networks. Social Network Analysis

and Mining, 5 (1), 21:1-21:20.

Harandi, S. R. 2015. Effects of e-learning on Students’

Motivation. Procedia-Social and Behavioral Sciences,

181, 423-430.

Hattie, J. 2009. Visible Learning: A Synthesis of Over 800

Meta-Analyses Relating to Achievement. Taylor &

Francis, London, New York.

Haythornthwaite, C., Andrews, R., Fransman, J., &

Meyers, E. M. 2016. The SAGE Handbook of E-

learning Research, 2e. Sage.

Hermann, U. 2009. Neurodidaktik. Grundlagen und

Vorschläge für gehirngerechtes Lernen (2. überarb.

Aufl.).

Johnson, L., Adams Becker, S., Estrada, V., Freeman, A.

2014. NMC Horizon Report: 2014 Higher Education

Edition. Austin, Texas: The New Media Consortium.

Jordan, K. 2015. Massive open online course completion

rates revisited: Assessment, length and attrition. The

International Review of Research in Open and

Distributed Learning, 16(3), 341-358.

Masud, M. 2016. Collaborative e-learning systems using

semantic data interoperability. Computers in Human

Behavior, 61, 127-135.

Monteiro, S., Lencastre, J. A., Osório, A. J., & Silva, B. D.

D. 2016. Reducing attrition and dropout in e-learning:

the development of a course design model. In

ICERI2016 Proceedings 9th International Conference

of Education, Research and Innovation (pp. 2440-

2446). IATED Academy.

Resendes, M.; Scardamalia, M.; Bereiter, C.; Chen, B.;

Halewood, C. 2015. Group-level formative feedback

and metadiscourse. International Journal of

Computer-Supported Collaborative Learning, 10 (3),

309-336.

Rivas, J. C. 2009. Neurodidáctica y estimulación del

potencial innovador para la competitividad en el tercer

milenio. Educación y desarrollo social, 3(2), 28-35.

Roth, G. 2004. Warum sind Lehren und Lernen so

schwierig? Zeitschrift fur Paedagogik, 50(4), 496-506.

Sabitzer, B. 2011. Neurodidactics–a new stimulus in ICT

and computer science education. INTED, 5881-5889.

Saleh, H. H.; Kadhim, N. J. 2016 Extractive Multi-

Document Text Summarization Using Multi-Objective

Evolutionary Algorithm Based Model. Iraqi Journal

of Science, 57 (1C), 728-741.

Sousa, D. A. 2014. Neurociencia educativa: Mente,

cerebro y educación. Narcea Ediciones.

Waelti, P., Dickinson, A. & Schultz, W. 2001. Dopamine

responses comply with basic assumptions of formal

learning theory, Nature 412, 43-48.

Westerhoff, N. 2010. La neurodidáctica a examen. Mente

y cerebro, (44), 34-40.

NeuroK: A Collaborative e-Learning Platform based on Pedagogical Principles from Neuroscience

555