=Paper=
{{Paper
|id=None
|storemode=property
|title=Intelligent Tutoring in Informal Settings: Empirical Study
|pdfUrl=https://ceur-ws.org/Vol-991/paper6.pdf
|volume=Vol-991
|dblpUrl=https://dblp.org/rec/conf/ecis/Ivanova13
}}
==Intelligent Tutoring in Informal Settings: Empirical Study==
https://ceur-ws.org/Vol-991/paper6.pdf
INTELLIGENT TUTORING IN INFORMAL SETTINGS:
EMPIRICAL STUDY
Malinka Ivanova, Technical University of Sofia, College of Energy and Electronics,
m_ivanova@tu-sofia.bg
Abstract
Modeling and realization of an intelligent tutor for informal situations in support of formal education
is a new and challenging problem relevant to the learning performance and efficacy. It needs further
exploration regarding the factors that distinguish tutoring and learning in a formal classroom and
informal places. The paper presents the results from an empirical study identifying the environmental
conditions, cognitive states and emotional charge of students that influence learning in informal
places like a cafe, a park, home and public transport. The techniques for attention concentration after
task breaking and during task performance are determined according to the gathered students’
opinion.
Keywords: informal learning, intelligent tutor, environmental conditions, affective state
1 Introduction
One direction for increasing personal effectiveness in learning is applying a strategy of intelligent
tutoring. The research society is looking for a computer tutor combining the advantages of the human
teachers’ capabilities and multi-tasking and multi-modal techniques of machine learning in the form of
web-based and mobile applications. Such computer tutor is described not only as a domain expert and
instructional designer who guides step-by-step and advises a learner on demand but also as a good
psychologist, adapting pedagogies and content to the emotional state. There are several solutions
proposing adaptive tutoring according to the personal cognitive and affective charge of a given
learner, realized by the integration of facial expression methods, sound recognition, text typing mood,
observation of learner’s physical characteristics.
Learning occurs not only during the planned classes according to the curriculum but also very often it
continues in informal settings in cafe places, libraries, home, etc. in the time between classes and after
that. Also, Csanyi et al. found the indicators through empirical study for interface between formal and
informal learning. They agree that "informal learning is not only autonomous activity, but also can be
influenced by teachers or trainers" and that the power of informal learning could be taken only if it is
related to formal learning (Csanyi et al., 2008). Furthermore, Chang et al. report for specially created
places for informal learning in the School of engineering at the University of Melbourne where
students immediately after formal classes could feel comfortable to continue their learning in a
“student-friendly” environment (Chang et al., 2009). The findings after a survey point increase use of
the informal places if they are available for students.
Modelling and realization of an intelligent tutor for informal situations in support of formal education
is a new and challenging problem relevant to the learning performance and efficacy. A motivated
learner is looking for knowledge receiving in informal spaces or outside the university in a time and
place suitable for him. The environmental conditions and the specificity of the places for informal
learning are different than these for lecture or practice classes. There are many variables for disturbing
and interrupting a started learning activity. The conditions and causes for interruptions in an informal
58
learning session and techniques for easy transitions from pauses to learning are still not studied and
algorithmized well. The main factor for improving the learning breakdowns is attention concentration
on the learning content according to research reports. Also, there is evidence that emotions influence
on learner's attention and concentration focusing on a given learning item (Chaffar and Frasson, 2004).
Another problem is related to motivation increasing for learning after interruptions and fast returning
of attention to the current studying point.
The aim of this paper is to identify the variables important for the realization of an intelligent tutor
supporting informal learning through empirical study. The explored variables are grouped in five
categories: (1) influence of environmental factors on attention concentration; (2) attention allocation
techniques, (3) identification of suitable cognitive level of learning in informal places, (4) influence of
emotional charge on learning (5) factors for improvement of motivation for learning.
2 Method
The aim of this empirical study is to identify disturbing and favorable factors that impact on learning
in informal places. For this purpose, survey tools are developed to gather the opinion of students. In
this exploration 22 male and 7 female students are involved. They are in their third year of bachelor
degree study in Computer Science. The survey tools include questions related to:
(1) How the environmental conditions influence on the process of learning? The conditions at four
often used places for informal learning are examined: cafe, home, park and public transport vehicles.
The used scale for evaluation is ranged from 5 – this factor possesses very high influence on learning
to 0 – this factor does not influence on learning.
(2) How the emotional states of a student contribute to the motivation for learning? The emotions are
divided in positive and negative and their imprint on the task concentration is identified. Students’
vote range from 5 – this emotion is very important for my learning to 0 – this emotion is not
important.
(3) How does personality influence on informal learning? The students’ personality is classified in 6
different groups: group1 includes students with intensive social life, who like to be in the centre of
attention and who in their activities is influenced by positive emotions; group 2 collects quiet,
reserved, self-controlled students who think intensively before every single step, their attention is
concentrated on their inside and specific personal life, they are slightly engaged in social relationships;
group 3 gathers students who are easily affected by the conditions of the environment, they easily can
be discouraged, the typical emotions for them are: anxiety, angry, guilty, envy, depression.; group 4
characterizes very impulsive persons who usually perform risky activities, they do not plan their steps,
they are at enmity with each other, they are with spirited mood.; group 5 are persons who respect
social laws and rules and these rules are a base for their activities; group 6 – it includes students who
cannot categorize them-selves in the above mentioned five groups.
(4) What kinds of techniques are suitable for attention concentration when learning occurs in informal
settings? The techniques are classified in 3 groups: appropriate techniques for attention returning when
a disturbing factor emerges, techniques for attention allocation during the task performance and
techniques for motivation to learn in informal situation.
(5) What is the maximal achieved level of cognition when a student learns in a café, at home, in a
park, in public transport vehicles? Bloom’s taxonomy with its 6 levels: knowledge, comprehension,
application, analysis, synthesis and evaluation is used. This will point the level of task complexity
suitable for serving to students when they decide to learn informally. It is important for motivation
improvement and for induction of positive emotions.
59
3 Influence of environment
At the foundation of this research the main hypothesis is that: the sound level, light intensity,
temperature, dynamics of people stream, furniture comfortableness, weather conditions, GSM ring,
food/coffee smell and secondary computer/mobile applications are among the factors that could
disturb or support learning when it occurs outside the formal classes. In order to prove or reject this
hypothesis the results of students’ votes are summarized in Table 1 in the cases when they use mobile
computers or smart phones for learning in a cafe, at home, in a park or in public transport. The opinion
of male and female students for the interference of the surrounding conditions on a learning process
differs. To identify the most disturbing factors the average values of male and female students are
calculated. The findings point that in a cafe: the loud talk (4.35) and loud music (3.5), intensive stream
of people (3.84), additional computer applications (3.01), GSM ring (2.96), hot room (2.81) and poor
light (2.65) are among the interventional factors with the biggest value. When students are at home,
they break off their task doing because of influence of the above mentioned factors typical for a cafe
plus the feeling of nice sunny weather (2.85), feeling the comfortless of the furniture (table – 2.63,
chair – 2.5) and the low temperatures in the room (2.51).
Factor In a cafe At home In a park In transport
score score score score score score score score score score score score
male female average male female average male female average male female average
Loud music 4.14 2.86 3.5 3.86 3.43 3.65 3.55 2.14 2.85 3.86 3.57 3.72
Quiet music 1.59 1.71 1.61 1.36 2.57 1.97 1.23 1.14 1.19 1.82 2.14 1.98
Loud talk 4.41 4.29 4.35 3.73 4.71 4.22 4.36 4.29 4.33 3.91 4.14 4.03
Quiet talk 2.18 2.71 2.45 2.09 2.57 2.33 1.91 3 2.46 2.18 3.14 2.66
Intense light 2.41 2.29 2.35 2.45 2 2.23 2.41 2.29 2.35 2.5 2.71 2.61
Poor light 2 3.29 2.65 2.14 3.29 2.72 2.23 2.71 2.47 2.23 2 2.12
Hot room/cabin 3.05 2.57 2.81 2.86 2.57 2.72 - - - 4.59 3.86 4.23
Cold room/cabin 2 1.58 1.79 2.59 2.43 2.51 - - - 3.05 3.43 3.24
Intensive stream of 3.68 4 3.84 3.36 4.43 3.9 2.86 4.14 3.5 3.64 4.14 3.89
people
Low stream of 1.82 2.71 2.27 1.4 3.86 2.63 1.36 3 2.18 1.91 2.86 2.39
people
Comfortable 2.55 1.71 2.13 2.86 2.14 2.5 2.55 2.29 2.42 2.45 1.57 2.01
chair/bench
Comfortable table 2.14 2 2.07 2.68 2.57 2.63 - - - - - -
Nice sunny weather 2.14 2.71 2.43 2.41 3.29 2.85 2.91 3.14 3.03 2.36 2.43 2.4
Nasty rainy weather 1.77 2.29 2.03 1.77 1.57 1.67 3.55 4 3.78 2.64 2 2.32
GSM ring 2.91 3 2.96 2.5 3 2.75 2.23 2.86 2.55 2.73 2.71 2.72
Strong food/coffee 2.18 1.43 1.81 2.27 2.29 2.28 2.09 1.29 1.69 2.45 1.29 1.87
smell
Week food/coffee 1.23 1.14 1.19 1.45 1.14 1.26 1.23 1.14 1.19 1.23 1.86 1.55
smell
Computer/mobile 2.73 3.29 3.01 3.36 3.43 3.36 2.41 3.29 2.85 2.5 3.29 2.9
applications
Stops frequency - - - - - - - - - 2.55 3.14 2.85
Table 1. Students’ vote about the influence of environmental conditions on their learning
60
During the time in the park students’ attention is drawn aside from the loud sound (talk – 4.33 and
music – 2.85), the weather conditions (nasty rainy weather – 3.78 and nice sunny weather – 3.03),
intensive stream of people (3.5), additional computer/mobile applications (2.85) and GSM ring (2.55).
When students travel via a public transport vehicle they feel intervention of temperatures in the cabin
(hot cabin – 4.23 and cold cabin – 3.24), loud sound (loud talk – 4.03 and loud music – 3.71),
intensive stream of people (3.89), additional computer applications (2.9), stops frequency (2.85), GSM
ring (2.72) and nice sunny weather (2.4). The repeated factors in the four described cases for informal
learning: in a café, at home, in a park and in public transport are shown on Figure 1. As it can be seen
the students are very sensitive to the loud sound, intensive stream of people, additional
computer/mobile applications installed locally on their devices and GSM ring. These factors should be
taken into consideration when an intelligent tutor is designed in support of learning in informal
situations.
Figure 1.The common factors disturbing learning in a cafe, at home, in a park and in public transport
4 Techniques for attention concentration and motivation
Different interventions could interrupt the students’ task doing or problem solving and researchers are
looking for suitable techniques that can easily and fast return the students’ attention on the working
item. Roda and Nabeth discuss four levels for attention focusing in an online learning environment: (1)
perceptual level - techniques for facilitating the access to the important parts of a web page, for
selection the needed information as well as techniques for attention focusing after interruption; (2)
deliberative level - tools for performing control on tasks and tasks priorities, techniques for motivation
and attention returning; (3) operational level - techniques and tools for context keeping after
interruptions, for information filtering, for reducing the level of complexity, (4) meta-cognitive level -
tools for self-diagnostic, techniques for attention allocation (Roda and Nabeth, 2008). Techniques,
tools and strategies for attention focusing suitable for an application at the above described four-level
model are extracted from the book “Human attention in digital environments” edited by Roda (2011)
and this information is analyzed and visualized on Figure 2.
For the purposes of this exploration the techniques for attention concentration are divided to: (1)
techniques for attention returning when a task is interrupted and the student gives up to continue at this
moment and (2) techniques facilitating task performance and attention allocation. According to the
students’ opinion when a factor breaks a task it would be very useful for software to remember the last
used item and to select it in different color or in another way (77% of all surveyed students). When a
student decides to return to this task it will be easy to continue from the selected item. The students
also appreciate help (72.5%) and advice receiving (64.5%) how to continue with the problem solving,
hint how far it gets last time (58%) and summary describing the work to this moment (53.5%). From
the survey it becomes clear that students do not like software interventions in the form of reminder
saying that they must continue (only 35% of all students give their vote for this technique). Among the
61
techniques enhancing task performance with the highest rate are: availability of instructions for task
performance (90.5% of all voted students), structuring the tasks according to different criteria (53.5%),
receiving appropriate advice (62.5%) and help (58%), a kind reminder about the goal of this task
(60.6%), and decreasing the task’s complexity when it is needed (53.5%). The students do not
appreciate enough the power of techniques like: tracking of affective states (face and facial expression
recognition) (49%), eye tracking (9%), and hands tracking (26%), although there are several
developed intelligent tutors reporting promising results. They integrate methods for affective state
recognition (Whitehill et al., 2008; Alexander et al., 2006; Ntombikayise and Robinson, 2011 )
and hand gestures tracking (Bustos et al., 2011; Abbasi et al, 2008).
filtering, rating
information selection
visualization tracking eye movement graphic display
Perceptional level
info graphs
information comprehension abstract info
info summary
show hidden info
metadata use
hide distractive info
group perception
sending of an email, message in a chat box, the
displaying of an item in the home page of a portal, the
display of a blinking icon, or the intervention of an
presentation prominence notifications artificial character
control of tasks priority
ordering of the item flicker
Deliberative
level
increase motivation for focus finding reminds help for encouragements
assessment of actions
context restoring tasks fragmentation in subtasks notifications interruptions
management,
interruption for
Operational
filtering incoming info watch lists, notifications performance facilitation
– time, content
level
reducing the required level of vigilance reducing the task complexity
order of activities time for each activity
visualization
statistics – number and nature of interruptions
Meta-cognitive level
time between message receiving and reflection
results comparison with others
self-diagnosis
discovering patterns of behavior
guidance for improvement of attention management
Figure 2. Four levels for attention management in digital environment according to (Roda and
Nabeth, 2008) and (Roda, 2011)
62
The tutoring system M-Ecolab uses a mechanism for adaptation of the learning resources according to
the individual student’s motivational state (Rebolledo-Mendez et al., 2011). The degree of motivation
is identified automatically through measurement of the students’ participation in solving problems, the
level of tasks difficulties, and according to the requested help during a task doing. The effect of the
applied motivational techniques on learning are evaluated through statistical tests that show
“significantly higher scores in the domain knowledge” achieved by students.
Among the directions for motivation improvement the students think that the following techniques
could contribute: time limiting for every single task (70.5% of all responders), comparison of learning
achievements among all participant in a lesson (62.5%), creation of a report summarizing the students’
current progress (56%), usage of messages notifying the current tasks’ status and student’s progress
(53%). The techniques like creation of a rating list (42%) and usage of messages for encouragement
(37%) are not rated very high by students.
5 Cognitive states
The Bloom’s taxonomy with its six levels for achievement of given cognitive skills is used to
determine the types of tasks suitable for serving to students when they are placed in informal settings.
According to the learning objectives the tasks are classified into 6 groups: group A includes tasks for
existing knowledge recalling, group B contains tasks for new knowledge understanding, group C –
tasks that require applying old and new knowledge in problem solving, group D – tasks that are
connected to the analytical students’ skills, group E consists of tasks that require skills for knowledge
combination and creation of new solution, group F includes tasks for evaluation and judgement of
knowledge, concepts, solutions. According to the responders the best learning tasks in a cafe should be
focused on existing knowledge recalling, comprehension of new concepts and applying this
knowledge to solve a problem (tasks from A, B and C groups). In a park and in transport the suitable
tasks are not so complex tasks - they require repetition of existing knowledge and understanding new
concepts (tasks from groups A and B). Home proposes the best conditions for informal learning and
students could solve complex problems including tasks at the levels of analysis, synthesis and
evaluation (tasks from all groups).
6 Affective states
Nowadays, ITSs are improved through modules for recognition of the student’ affective states and
possibilities for the induction of suitable emotions for learning. Several of them can predict the
emotional state of a student, dynamics in his behaviour and in this way different strategy for
motivation could be applied and the learning path could be optimized.
Rishi (2009) discusses the influence of emotional state on attention in task doing – negative emotions
(anger, anxiety, or distress) do not allow focusing on the learning item or moving the attention to the
new one and in this way the learning performance is decreased. Positive emotions like joy and pride
could facilitate thinking and learning. The author proposes a rule-based dynamic method for ensuring
the best emotional conditions for learning, including detection of emotions and provoking suitable
affective state for performance improvement.
Chaffar and Frasson (2004) present a system ESTEL (Emotional State towards Efficient Learning
system) that has features to predict the optimal emotional state for learning according to the learner’s
personality. It can induce the appropriate emotions to improve the processes memorization and
comprehension through applying different techniques like guided imagery, music and images. The
learners’ personality is divided into four groups: (1) extraverts are active and communicative persons
who could easily be influenced by positive emotions; (2) neuroticism is typical for people who are
easily affected by the conditions of the surrounding environment and who are easily discouraged; (3)
psychoticism are impulsive and hostile people; (4) lie group includes sociable persons with respect to
63
the societal laws. The authors show the connection between a learner's personality and optimal
emotional state. As it can be seen the common affective states that could play a catalizator role for
learning are positive emotions like: joy, confident, pride, anxious, self-gratification. The authors
propose a six module architecture including: emotion manager - responsible for emotions monitoring,
distribution and tasks synchronization; emotion identifier - recognizes and predicts the emotional state
of the learner according to his color preferences; personality identifier - identify the personality of a
learner through a questionnaire and communicates with the emotion manager module; optimal
emotion extractor determines the optimal emotional state according to the learner personality using a
set of rules; emotion inducer - induces the suitable emotions for learning; learning appraiser evaluates
the learner’ s performance in his current affective state through a pre-test and evaluates the
performance in his optimal emotional state through a post-test. The Naive Bayes classifier is applied
for optimal emotional state prediction in correspondence of the learner’s personality.
du Boulay researches the factors that could support the design of motivational modules in ITSs
(Boulay, 2011). The author explores three negative emotions: frustration, anxiety and boredom and
searches for suitable pedagogical strategies according to the three motivational states: values
(personal, social and cultural background of a student that stimulates his participation in a learning
process), expectancies (expectations of a student for performing learning) and feelings (the emotion
emerging from the previous experience).
Another study addresses the influence of positive emotions on learning performance and facilitation of
the cognitive process (Um et al., 2007). The findings show significant impact of positive emotions on
cognition and learning. The authors agree that instructional design and instructional learning objects
can be used for induction of a positive mood, to improve students’ experience, satisfaction and
performance.
For the purposes of this study a set of emotions – positive and negative is proposed to students to cast
their vote about the influence of emotions on learning in an informal situation. As it can be seen on
Figure 3 and Figure 4 the positive as well as the negative emotions have the power to drive learning
and to motivate or not the students. The positive emotions with the highest scores that support
meaningful learning are: joy, happiness, enthusiasm, and confidence. Among negative emotions that
influence learning with the highest scores are: angry, perturbed, anxiety and hopelessness.
The students’ personality is explored too according to self-report of responders. 31.9% of male
students and 28.6% of female students self-describe as quiet, reserved, self-controlled students; they
prefer to learn alone, concentrating all their attention on the learning item. 22.7% of male students and
42.8% of female students are classified in the group of social, active and communicative persons; they
are easily influenced from the positive emotions.
Figure 3. Students’ vote about the influence of Figure 4. Students’ vote about the influence of
positive emotions negative emotions
64
The rest of the students (the smaller part of the responders) define them-selves as students who are
easily affected by the environmental conditions (13.7% of male students and 14.3% of female
students); as very impulsive persons who usually perform risky activities (13.7% of male students and
14.3% of female students); as persons who follow social rules in their activities performance (9% of
male students) and as students who combine features of the above mentioned personality groups –
showing one or other feature according to the conditions and situations (9% of male students).
The results from the empirical study and findings from literature point that an intelligent tutor with an
intention to support informal learning has to integrate techniques for induction of suitable emotions
predisposing students to continue their work on tasks and achieving the planed outcomes.
7 Conclusions
The paper presents the results from an empirical study with focus on the main factors that impact on
learning efficacy when students are placed in informal settings according to their vote. The findings
show that there are many different factors that disturb and interrupt learning. Several of them are so
strong and they could lead to task breaking for a long time or refusal of further problem solving and
task doing. Therefore, suitable techniques for motivation and emotional charge have to be selected
very precisely and used for realization of an intelligent tutor. This study stresses strongly the main
factors that should be taken into consideration when learning occurs in an informal situation and they
are found in environmental conditions, the level of cognition, the emotional state, attention
concentration and motivation for learning. A model is developed to summarize variables that influence
learning in informal situations (Figure 5).
Intelligent tutor for informal learning
Environment Attention returning Motivation Cognitive states Affective states
Loud talk Last item improvement Home – Positive
Intensive remembering Time limiting remember, emotions – joy,
people stream Help/advise for a task understand, happiness,
Loud music receiving Comparison of apply, analyze, enthusiasm,
Computer apps Summary of the achievements synthesize, confidence
GSM ring work to this Progress report evaluate Negative
moment Notification Cafe - emotions –
messages remember, angry,
Attention at task understand, perturbed,
performing apply anxiety,
Instructions Park and hopelesness
Tasks structuring transport -
Help/advice/ remember, Personality
reminder understand Social, active
and
communicative
Quite, self-
controlled
Easily affected
by
environmental
conditions
Follow social
rules
Get risky
activities
Combination
of others
Figure 5. Factors influencing learning in informal situations
65
References
Csanyi, G., Jerlich, J., Pohl, M., Reichl, F. (2008 ) Formal and informal technology enhanced
learning for initial and continuing engineering education. Presented at IACEE 11th World
Conference on Continuing Engineering Education, Atlanta, USA,
http://smartech.gatech.edu/jspui/bitstream/1853/24401/1/202.pdf
Chang, R., Stern, L., Sondergaard, H., R. Hadgraft (2009) Places for learning engineering: A
preliminary report on informal learning spaces, Proceedings of the Research in Engineering
Education Symposium Palm Cove, QLD,
http://rees2009.pbworks.com/f/rees2009_submission_86.pdf
Chaffar, S., Frasson, C. (2004) Inducing optimal emotional state for learning in intelligent tutoring
systems. 7th International Conference, ITS 2004, Maceio, Alagoas, Brazil, Springer Lecture Notes
in Computer Science 3220, Springer, 45-54.
Rishi (2009) Intellectual Intelligent Tutoring System: The ITS with Emotions, International Journal of
Engineering and Technology Vol. 1, No. 1, April, 2009, pp 1793-8236
Roda and Nabeth, Attention Management in Organizations: Four Levels of Support in Information
Systems. (2008) In A. Bounfour (Ed.), Organisational capital: Modelling, measuring,
contextualising: Routledge - advanced research series in management, 214-233.
Roda (2011) Human attention in digital environments. Cambridge University Press, 2011
Whitehill, J., Bartlett, M., Movellan, J. (2008) Automatic Facial Expression Recognition for
Intelligent Tutoring Systems. Proceedings of the CVPR Workshop on Human Communicative
Behavior Analysis, http://mplab.ucsd.edu/~jake/its08.pdf
Alexander, S.T., Sarrafzadeh, V. A., Hill, S. (2006) Easy with Eve: A Functional Affective Tutoring
System. Workshop on Motivational and affective issues in Intelligent Tutoring Systems, Intelligent
Tutoring Systems 2006, Jhongli, Taiwan
Banda, N., Robinson, P. (2001) Multimodal, Affect Recognition in Intelligent Tutoring Systems.
ACII’11 Proceedings of the 4th international conference on Affective computing and intelligent
interaction 2011- Volume Part II , 200-207.
Bustos, D. M., Chua, G. L., Cruz, R. T., Santos, J. M., Suarez, M. T. (2011) Gesture-Based Affect
Modeling for Intelligent Tutoring Systems. Artificial Intelligence in Education Lecture Notes in
Computer Science Volume 6738, 426-428 .
Abbasi, A. R., Nitin V. Afzulpurkar and Uno, T. (2008) Exploring Un-Intentional Body Gestures for
Affective System Design. Affective Computing, book edited by Jimmy Or, ISBN 978-3-902613-
23-3, http://cdn.intechopen.com/pdfs/5193/InTech-
Exploring_un_intentional_body_gestures_for_affective_system_design.pdf
du Boulay, B. (2011) Towards a Motivationally-Intelligent Pedagogy: How should an intelligent tutor
respond to the unmotivated or the emotivated? In New Perspectives on Affect and Learning
Technologies, R. Calvo and S. D'Mello, Eds. Springer, New York, 41-52.
Rebolledo-Mendez, G., Luckin, R., du Boulay, B. (2011) Designing Adaptive Motivational
Scaffolding for a Tutoring System. In Calvo, R.A. & D’Mello, S. (Eds.) New Perspectives on
Affect and Learning Technologies, Springer: New York,
http://www.sussex.ac.uk/Users/bend/papers/calvo2.pdf
Um, E.R., Song, H., Plass, J. (2007) The Effect of Positive Emotions on Multimedia Learning. In C.
Montgomerie & J. Seale (Eds.), Proceedings of World Conference on Educational Multimedia,
Hypermedia and Telecommunications 2007, Chesapeake, VA: AACE, 4176-4185.
66