EMOTION DETECTION WHILE LISTENING TO QURAN RECITATION USING EEG AND ECG
AMJAD M. R. ALZEERALHOUSEINI
A dissertation submitted in fulfilment of requirement for the degree of Master of Computer Science
Kulliyyah of Information and Communication Technology International Islamic University Malaysia
Emotion modelling and identification has attracted substantial interest from several disciplines including computer science, cognitive science and psychology. Despite the fact that many qualitative studies have been carried out on emotion, quantifying physiological signals remains one of the less-investigated aspects. Therefore, the purpose of this study is to examine various human emotions exhibited by subjects while listening to recitations of Quranic verses based on their perceived meaning or the tone of the verse. This work focuses on understanding and analysing brain and heart activities for two groups: one group understands the language of Al-Quran, while the other group does not. This study attempts to identify the factors (content or intonation) that elicit subjects‘ emotions while listening to Quranic recitations. The study uses two methods to measure subjects‘ physiological signals: the electroencephalogram (EEG) and the electrocardiogram (ECG). The resulting data are used to analyse subjects‘ emotional properties. A solution based on kernel density estimation (KDE) and mel frequency cepstral coefficients (MFCC) is proposed for recognising dynamically developing emotional patterns from EEG and ECG signals.
This work uses the multilayer perceptron (MLP) classifier. This classifier‘s features are based on the affective space model (ASM), which is represented by two factors:
valence and arousal. The experimental setup presented in this work to elicit emotions is based on passive valence/arousal. The EEG and ECG data were collected from 20 Muslim subjects, 10 of whom understood the language of Al-Quran (Arabic), while the remainder did not. In the experiment, visual and auditory stimuli (passive stressors) were used to induce positive and negative emotions. The International Affective Picture System (IAPS) was used to elicit emotions. The results support the use of EEGs as a reliable source for evaluating four basic human emotions. While ECGs can also successfully identify these four basic emotions, the accuracy of emotion extraction from ECG signals is lower than the accuracy from EEG signals.
Additionally, the MFCC algorithm with 12 extracted features resulted in higher accuracy than the KDE algorithm when extracting emotions from signals. The groups who understood Arabic reported lower valence than those who did not, indicating that they were affected by the content or meaning of the recitations.
ABSTRACT IN ARABIC
مولع كلذ في ابم تاصصختلا نم ديدعلا لبق نم يربك لابقإ فطاوعلا ديدتحو ةجذنم تبذج تاساردلا نم ديدعلا نأ نم مغرلا ىلع .سفنلا ملعو ةيفرعلدا مولعلاو ،بوسالحا تراس دق ةيعونلا
َنإف ،كلذلو .ًاقيقتح لقأ بناولجا نم ةدحاو لازت لا ةيجولويزيفلا تاراشلإا سايق َنأ لاا ،ةفطاعلا ىلع ةولات لىإ عامتسلاا ءانثأ في رهظت تيلا ةيناسنلإا فطاوعلا فلتمخ ةسارد وه ةساردلا هذه نم ضرغلا لآا ةجلذ وأ روصتلدا نىعلدا ساسأ ىلع ةينآرقلا تايلآا ةةطنأ ليلتحو مهف ىلع لمعلا اذه زكريو .ةي
ةغللا مهفت لا ةيناثلا ةعولمجا ينح في ،نآرقلا ةغل مهفت لىولأا ةعوملمجا :ينتعوملمج بلقلاو غامدلا ءانثأ في فطاوعلا يرثت تيلا )ديوجتلا وأ ىوتلمحا( لماوعلا لىإ فرعتلا ةساردلا هذه لواتحو .ةيبرعلا نآرق تاولات لىإ عامتسلاا ( :ةيجولويزيفلا تاراشلإا سايقل ينتقيرط ةساردلا مدختستو .ةي
( ريدقت ساسأ ىلع للحا حترقُيو ،ةيفطاعلا صئاصلخا ليلحتل ةتجانلا تانايبلا مادختسا متيو .)ECG
( ةاون ةفاثك ( تلاماعم ليم دّدرتو )KDE
ليلتح قيرط نع تافصلاو صئاصلخا جارختسلا )MFCC
او ةيغامدلا تاجولدا ( تاقبةلا ةدّدعتم تلابقتسلدا فنصلدا لمعلا اذه مدختسي .ةيبلقل
( يفطاعلا ءاضفلا جذونم لىإ فنصلدا اذه تازيم دنتستو ؤفاكتلا :ينلماعب لثتد يذلاو ،)ASM
ىلع ةيبايجلإا و ةيبلسلا رعاطلدا جارختسا لىإ لمعلا اذه في بييرجتلا دادعلإا اذه دنتسيو .ةراثلإاو ا / ؤفاكتلا تانايبلا عجم ّتم دقو .ةراثتسلا
،ينملسلدا نم اصخش 02
تم ،ةبرجتلا فيو .ةيبرعلا ةغللا مهفت لا ىرخلأا ةعوملمجا ينح في ،)بيرع( نآرقلا ةغل مهفي مهنم 02 بلسلاو ةيبايجلإا رعاطلدا ىلع ثحلل )ةيبلسلا تاطوغضلا( ةيعمسلاو ةيرصبلا تايرثلدا مادختسا تمو .ةي
( ةيلود ةيفطاع روص ماظن مادختسا ًاضيأ مادختسا جئاتن معدتو .فطاوعلا ىلع لوصحلل )IAPS
بلقلا مادختسا نكيم ينح في ،ةيساسلأا ةيناسنلإا رعاطلدا ةعبرلأا مييقتل قوثوم ردصمك خلدا مسر لدا جارختسا ّدعُي .ةقدو حاجنب عبرلأا ةيساسلأا فطاوعلا هذه ىلع فرعتلا في اضيأ تاراشإ نم رعاط
تاراشإ نم فطاوعلا جارختسا نم ًةَقد لقأ بلقلا طيةتخ ةيمزراوخ تدأ ،كلذ لىإ ةفاضلإاب ،EEG
ةيمزراوخ نم ىلعأ ةقد في جرختسلدا تازيم نم ةزَيم 00 نم رعاطلدا جارختسا دنعKDE
.تاراشلإا ةيبرعلا ةغللا مهفت تيلا تاعوملمجا ّنأ لىإ ةراشلإا ردتجو
لا تيلا تاعوملمجا نم رثكأ ترثات
.تاءارقلا نىعم وأ ىوتحبم ترثأت انهأ لىإ يرطي امم ،ةيبرعلا ةغللا مهفت
I certify that I have supervised and read this study and that in my opinion; it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Master of Computer Science.
Imad Fakhri Al Shaikhli Supervisor
I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Master of Computer Science.
Normaziah Abdul Aziz Internal Examiner
Tahir Ahmad External Examiner
This dissertation was submitted to the Department of Computer Science and is accepted as a fulfilment of the requirement for the degree of Master of Computer Science.
Normi Sham Awang Abu Bakar Head, Department of Computer Science.
This dissertation was submitted to the Kulliyyah of Information and Communication Technology and is accepted as a fulfilment of the requirement for the degree of Master of Computer Science.
Abdul Wahab Bin Abdul Rahman Dean, Kulliyyah of Information and Communication Technology
I hereby declare that this dissertation is the result of my own investigation, except where otherwise stated. I also declare that it has not been previously or concurrently submitted as a whole for any other degrees at IIUM or other institutions.
Amjad M.R. Alzeeralhouseini
Signature………....………. Date …….……….
INTERNATIONAL ISLAMIC UNIVERSITY MALAYSIA
DECLARATION OF COPYRIGHT AND AFFIRMATION OF FAIR USE OF UNPUBLISHED RESEARCH
EMOTION DETECTION WHILE LISTENING TO QURAN RECITATION USING EEG AND ECG
I declare that the copyright holder of this dissertation are jointly owned by the student and IIUM.
Copyright © 2016 Amjad M.R. Alzeeralhouseini and International Islamic University Malaysia. All rights reserved.
No part of this unpublished research may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording or otherwise without prior written permission of the copyright holder except as provided below
1. Any material contained in or derived from this unpublished research may be used by others in their writing with due acknowledgement.
2. IIUM or its library will have the right to make and transmit copies (print or electronic) for institutional and academic purposes.
3. The IIUM library will have the right to make, store in a retrieved system and supply copies of this unpublished research if requested by other universities and research libraries.
By signing this form, I acknowledged that I have read and understand the IIUM Intellectual Property Right and Commercialization policy.
Affirmed by Amjad M.R. Alzeeralhouseini
This dissertation is dedicated to my beloved parents
First and above all, all praise to Allah, the almighty the compassionate the merciful for providing me this opportunity and granting me the capability to proceed successfully.
This thesis appears in its current form due to the assistance of and guidance of several people. I would therefore like to offer my sincere thanks to all of them.
It is my utmost pleasure to dedicate this work to my dear parents, to my father Talal and my mother Intisar, who granted me the gift of their unwavering belief in my ability to accomplish this goal: thank you for your support and patience.
I would like to express my deepest gratitude towards Professor Imad Fakhri Taha for his continuous support, encouragement and leadership. He continually conveyed a spirit of adventure in regard to research and scholarship, and for that, I will be forever grateful.
I would like to thank my brothers Abdul Raheem and Majdi for their continuous support and patience.
I would like to thank my aunties and my uncles Nawal, Tagred, Riyad, Alia for their doa.
I would like to thank my uncle Amir Al Tamimi for his continuous support.
I would like to thank the dean of KICT Professor Abdul Wahab Bin Abdul Rahman for his support.
I would like to thank Associate Professor Dr Normaziah binit Abdul Aziz for her motherly care.
I would like to thank Dr Normi sham for her help and support.
Also, my deepest appreciation for my examiner Professor Tahir Ahmad who provided constructive feedback.
I would like to thank my friends Ibraheem Ahmaro, Fardous Eljadi, Khamis Al arabi for their support.
I would like to thank Sr.Narieta, Sr.Kamsiah, Sr. Shahran and Sr.Pauziah for their continuous help.
I wish to express my appreciation and thanks to those who provided their time, effort and support for this project. To the members of my dissertation committee, faculty staffs and students and everyone who help directly or indirectly thank you.
TABLE OF CONTENTS
Abstract ... ii
Abstract in Arabic ... iii
Approval Page ... iv
Declaration ... v
Copyright ... vi
Dedication ... vii
Acknowledgements ... viii
List of Tables ... xii
List of Figures ... xiv
CHAPTER ONE: INTRODUCTION ... 1
1.1Introduction ... 1
1.2Problem Statement ... 5
1.3Research Objectives ... 6
1.4Research Questions ... 6
1.5Research Hypotheses ... 7
1.6Significance of the Study ... 7
1.7Background ... 8
CHAPTER TWO: LITERATURE REVIEW ... 10
2.1Introduction ... 10
2.2Psychological Health and the Quran ... 11
2.2.1 Identification of Human Emotions ... 12
2.3Emotion ... 15
2.3.1 Emotion and Cognition ... 16
2.3.2 Psychological Models of Human Emotions ... 17
188.8.131.52The Theory of Basic Emotion ... 17
2.4Electroencephalogram ... 18
2.4.1 Methods for Recording EEG Signals ... 18
184.108.40.206Functional Magnetic Resonance Imaging FMRI ... 19
220.127.116.11Positron Emission Tomography PET ... 22
18.104.22.168Electroencephalogram EEG ... 22
2.4.2 EEG Data Collection ... 24
2.4.3 EEG Analysis ... 25
22.214.171.124Types of Waves in EEG ... 26
126.96.36.199Quantitative EEG ... 27
2.5Electrocardiogram ... 27
2.5.1 Cardiovascular Reactivity to Emotion ... 30
2.6Affective Computing: ECG ... 33
2.7Experimental Setup ... 34
CHAPTER THREE: PROPOSED METHODOLOGY ... 36
3.1Introduction ... 36
3.2Data Collection ... 38
3.3Pre-Processing ... 39
3.3.1 Signal Filter ... 39
3.4Feature Extraction ... 40
3.4.1 Mel Frequency Cepstral Coefficient (MFCC) ... 40
3.4.2 Kernel Density Estimate (KDE) ... 42
3.5Classification ... 43
3.5.1 K-Fold Cross Validation ... 43
3.5.2 Multilayer Perceptron (MLP) ... 44
CHAPTER FOUR: EXPERIMENTAL PROTOCOL ... 47
4.1Introduction ... 47
4.2Capturing Emotion Responses ... 47
4.3Preliminary Experiment to Capture Subjects‘ Emotions ... 48
4.3.1 Experimental Setup... 48
4.3.2 Electrode Placement ... 49
4.3.3 Experiment Design ... 50
4.3.4 Preliminary Emotion Detection ... 52
188.8.131.52EEG Valence and Arousal Analysis Using KDE (Memory Test) ... 52
184.108.40.206EEG Valence and Arousal Analysis Using MFCC (Memory Test) ... 56
220.127.116.11ECG Valence and Arousal Analysis Using KDE (Memory Test) ... 59
18.104.22.168ECG Valence and Arousal Analysis Using MFCC (Memory Test) ... 62
4.4Primary Experiment on Subject‘s Emotion ... 66
4.4.1 Experimental Setup... 66
4.4.2 Stimuli 69 4.4.3 Participants ... 70
4.4.4 Experimental Setup for Training and Testing Data ... 70
4.5Chapter Summary ... 75
CHAPTER FIVE: RESULTS AND DISCUSSION ... 76
5.1Introduction ... 76
5.2Results ... 76
5.2.1 Valence and Arousal Analysis of Emotions ... 77
5.2.2 Valence Arousal Analysis of Quranic Recitation ... 86
5.2.3 Self-Assessment Manikin (SAM) Results ... 90
5.3Chapter Summary ... 93
CHAPTER SIX: CONCLUSION AND FUTURE WORK ... 94
6.1Introduction ... 94
6.2Summary of Results ... 95
6.3Significance of Study ... 97
6.3.1 Significance of Study towards Human Life ... 97
6.4Contribution of Study ... 97
6.4.1 Contribution to Computer Science Field ... 97
6.5Chapter Summary ... 98
REFERENCES ... 99 APPENDIX: CONSENT FORM ... 104
LIST OF TABLES
Table 2.1 Taxonomic Table for Research Studies Related to the Quran &
Mental Health 12
Table 2.2 Types of Brain Waves 13
Table 3.1 MLP Network Parameters 46
Table 4.1 Experiment Time Frame Protocol 52
Table 4.2 EEG confusion Matrix for Male participants using the KDE 54 Table 4.3 EEG confusion Matrix for female participants using KDE 55 Table 4.4 EEG confusion Matrix for Male Participants using MFCC 57 Table 4.5 EEG confusion Matrix for female participants using MFCC 58 Table 4.6 ECG confusion Matrix for Male Participants using KDE 60 Table 4.7 ECG confusion Matrix for Female Participants using KDE 61 Table 4.8 ECG confusion Matrix for male Participants using MFCC 63 Table 4.9 ECG confusion Matrix for female Participants using MFCC 64
Table 4.10 Emotion Distribution for 3 Subjects 65
Table 4.11 Experiment Research Protocol 69
Table 4.12 Analysis types for participants‘ emotions 72
Table 4.13 Five Fold Homogenous Test using the MFCC 73
Table 4.14 Five Fold Homogenous Test using the KDE 74
Table 5.1 Accuracy of Emotion Detection based on the MFCC using the
Homogenous Test 83
Table 5.2 Accuracy of Emotion Detection based on the KDE using the
Homogenous Test 84
Table 5.3 Accuracy of Emotion Detection based on the MFCC using the
Heterogeneous Test 85
Table 5.4 Valence Arousal before and after listening to the Quran 86
Table 5.5 Valence Arousal Analysis for All Groups 87
Table 5.6 Valence Arousal Analysis for Group A 88
Table 5.7 Valence Arousal Analysis for Group B 88
Table 5.8 Valence Arousal Analysis for Group C 89
Table 5.9 Valence Arousal Analysis for Group D 89
Table 5.10 Valence Arousal Analysis of All subjects using ECG signals 90 Table 5.11 Valence Arousal Analysis of SAM for Group A 91 Table 5.12 Valence Arousal Analysis of SAM for Group B 91 Table 5.13 Valence Arousal Analysis of SAM for Group C 92 Table 5.14 Valence Arousal Analysis of SAM for Group D 92
LIST OF FIGURES
Figure 2.1 The Two-Dimensional Emotion Model 13
Figure 2.2 Electrode Placements for EEG 14
Figure 2.3 Typical ECG Signal 14
Figure 2.4 Electrodes Placement for ECG 15
Figure 2.5 The QRS Complex of ECG 28
Figure 3.1 Proposed Methodology for Data Analysis 37
Figure 3.2 Comparison between filtered signal (left) and raw signal (right) 40 Figure 3.3 Block Diagram for the MFCC algorithm (Han et al., 2006) 41 Figure 3.4 12 Features extracted by the MFCC for one channel 41 Figure 3.5 Block Diagram for the KDE algorithm (Han et al., 2006) 42 Figure 3.6 Schematic Representation of fivefold cross validation 44 Figure 3.7 The MLP Network with one hidden layer (Orbach, 1962; Orhan,
Hekim, & Ozer, 2011) 45
Figure 4.1 Block Diagram for Analysing Raw EEG Signal 49
Figure 4.2 Block diagram for ECG analysis 50
Figure 4.3 Standard 10-20 Measurement System 67
Figure 4.4 Standard Lead II Placement 67
Figure 4.5 Five required stages of data to obtain results 71 Figure 4.6 Testing data block diagram for Quranic recitation tasks 75 Figure 5.1 Accuracy of Emotions based on the KDE using Memory Test for
EEG signal 78
Figure 5.2 Accuracy of Emotion Detection based on the KDE using the
Memory Test for ECG signals 78
Figure 5.3 Accuracy of Emotion Detection based on the MFCC using the
Memory Test for EEG signals 79
Figure 5.4 Accuracy of Emotion Detection based on the KDE using the
Homogenous Test for EEG signals 80
Figure 5.5 Accuracy of Emotion Detection based on the MFCC using the
Homogenous Test for EEG signals 80
Figure 5.6 Accuracy of Emotion Detection based on the KDE using
Homogenous Test for ECG signals 82
Figure 5.7 Accuracy of Emotion Detection based on the KDE using the
Heterogeneous Test for ECG signals 82
CHAPTER ONE INTRODUCTION
The goal of this research was to investigate the relationship between human emotions and Quranic recitations using Electrocardiogram (ECG) and Electroencephalogram (EEG) measurements. To accomplish this goal, we developed techniques based on the Affective Space Model (ASM) which enables psychological feelings to become manifest as valence (V) and arousal (A). Emotions are associated with every activity in everyday life and play a major part in non-verbal communication. Feelings may also affect the areas of the brain that regulate thinking, reasoning, and decision making. Deep emotions produce specific physical reactions such as increasing the heart rate, face flushing, increasing the rate of respiration, and increasing hypertension—all are moving effects. Numerous distressing elements can affect human attention (Li, Chai, Kaixiang, Wahab, & Abut, 2009; Othman, Wahab, Karim, Dzulkifli, & Alshaikli, 2013). In previous studies, researchers have proven that negative (and sometimes positive) psychological states can be altered. Negative emotions are individual reactions to stressful situations, while positive emotions are individual reactions to pleasing and calm situations. Therefore, being able to quantify emotions is vital to understanding individual conduct. This study aims to deliver an emotion-detection system that uses pattern recognition and classification methods, constituting a step toward developing a man-machine interface that can respond to human non-verbal cues such as individual desires, feelings and emotions,.
Electroencephalography (EEG) measures the amplitude and frequency of the electrical pulses produced by human brains. It has advantages in carrying out experiments
because it is non-invasive, straightforward, fast, and low-cost. Moreover, EEGs are not harmful, distressing, or time-consuming for participants. Therefore, EEGs have become a desirable technique for examining the brain‘s reactions to emotional stimuli. In this research, a common neural network was applied to EEG and ECG signals‘ frequency domain data, with the intent of establishing a suitable selection of parameters. These parameters were derived by conducting several experiments on different subjects. Two of the parameters are the overlapping rate and sampling length. The selection of parameters is normally much better and more insightful when based on practical experience. Finally, experiments were carried out to reveal the significant characteristics that can be used to classify four basic human emotions (happiness, calmness, sadness, and fear).
The initial step in modelling any phenomenon is to gather data. The experimental design is a vital step in acquiring meaningful data; therefore, we had to institute methods that could effectively stimulate emotions in a lab environment, where we could record and gather meaningful data. In quantifying psychological actions, we are restricted to the study of observable expressions such as vocal traits, gestures, and facial expressions. These strategies are well-known in Human Computer Interaction because they make use of the same hints that people depend upon to identify and determine emotional states. Furthermore, the majority of people exhibit similar expressions as a result of identical psychological stimuli, which enables researchers to objectively notate emotions (Agrafioti, Hatzinakos, Member, & Anderson, 2012).
The main disadvantage to applying attitudinal strategies for emotion recognition is the scepticism on the part of participants who either knowingly control their emotional expressions or are simply normally suppressive. For example, even though facial expressions may be examined to discover emotions, there is no certainty that a person will convey the
related cue, regardless of whether they are experiencing a specific emotion. This has critical ramifications in certain applications, such as surveillance.
An intriguing alternative option to revealing emotional modalities is to examine the essential bio signals produced by the body. Methods for collecting bio signals include the electromyogram (EMG), electroencephalogram (EEG), galvanic skin response (GSR), electrocardiogram (ECG), heart rate (HR), blood volume pressure (BVP) and heart rate variability (HRV). These bio signals have typically been employed in medical diagnostics, but there is substantial evidence that they are responsive to and may relay details about psychological states (Anttonen & Surakka, 2005; Ekman, 1999; Hönig, Batliner, & Nöth, 2007; Jones & Troen, 2007; Mandryk, Inkpen, & Calvert, 2006; Picard, 2003). One advantage of recognizing emotions and feelings using physiological signals is that these are unconscious responses and therefore are difficult for participants to conceal. Furthermore, during the time that the electrodes are connected to a person, the signals are captured constantly, permitting regular analysis of the subject‘s psychological state. This is not the case with verbal features, for instance, which may be recorded only while the person is conversing.
In addition to the open theoretical issues, there are practical challenges as well. The experimental setup used here is much more sophisticated than those used in previous behavioural emotion investigations, where the data gathering involves showing that the participants demonstrated emotions. For EEG and ECG testing more complex techniques are required to generate genuine feelings in a lab environment. Additionally, interpreting physiological signals is highly subjective and, therefore, there is a risk of complexity in developing the ground truth. An additional practical obstacle pertains to signal acquisition.
The data gathering process for EEGs and ECGs is more invasive than that for conduct modalities because they require electrodes that must be in contact with a participant‘s skin
throughout a recording period. Therefore, it is crucial to reduce the amount of information needed for this particular task—i.e., for recognition to depend on as few signals as possible.
Stress and psychological disorders are some of the most common problems in our daily lives; stress treatment has been the subject of numerous studies. Some studies suggest that the Quran can offer a great deal of help in treating stress and psychological disorders.
Many Muslim practitioners use the Quran in their daily lives as a means to reduce stress. The Quran has been—and still is—used in healing human psychological disorders. Moreover, Muslims have used the Quran throughout the ages to heal and treat people suffering from stress and psychological disorders. The source of these psychological disorders and stresses is lack of human motivation, corrupt belief systems, and social and political events. Therefore, the purpose of the Quran is to establish and maintain balance between different aspects of human life (Ebrahimi, 2011).
More studies are needed to integrate readings of the Quran with other medical treatments. The Quran can help to reduce human suffering and stress; listening to the Quran can be used to develop good mental health and achieve greater peace (Mahjoob, Nejati, Hosseini, & Bakhshani, 2014). Quran memorization affects internal factors and can improve people‘s mental health; it can also be an effective resource in coping with modern stresses and challenges (Kimiaee, Khademian, & Farhadi, 2012).
The verses of the Quran have different levels of impact than other verses. There are verses that tell stories about the prophet, the hereafter, treatment, and so on. Several studies have been conducted on the effects of listening to Quranic recitations as a way to reduce daily stress and bring peace and tranquillity to the heart (Atarodi, Mottaghi, & Atarodi, 2012;
Hamidi, Bagherzadeh, & Gafarzadeh, 2010; Khan et al., 2010; Kimiaee et al., 2012; Mahjoob et al., 2014; Mottaghi, Esmaili, & Rohani, 2011; Siahpoosh, 2012; Zulkurnaini, Kadir, Murat,
& Isa, 2012). However, these studies have not specified the selection criteria for the verses of
the Quran used in their experiments. Therefore, one aim of this study is to identify specific verses of the Quran that have an emotional impact on humans.
1.2 PROBLEM STATEMENT
Topics associated with the effect of the Quran on the human brain have been a subject of interest to several researchers. Al-Quran words are believed by Muslims to be the words of God, which have distinctive characteristics (Khan et al., 2010; Siahpoosh, 2012).
In spite of the large amount of research on the relationship between religious perceptions and psychological variables (including anxiety and stress), an analysis of studies and sources conducted thus far indicate that—unfortunately—very little scientific research is available concerning the psychological effects of the Quran (Atarodi et al., 2012; Mottaghi et al., 2011).
Until now, there has been no scientific investigation comparing those who understand the language of the Quran (Arabic) and those who do not while listening to Quranic recitations as measured by EEG and ECG signals. This research aims to investigate and compare the brain wave and heart wave patterns for four groups (A, B, C, and D). Groups A and B each consist of 5 males who do or do not understand Arabic, respectively, while Groups C and D consist of 5 females who do or do not understand Arabic, respectively.
In previous studies, the criteria for selecting subjects were often random or unspecified. These studies did not note whether the participants were able to understand Arabic. Additionally, arbitrary selections of Quran verses and reciters were adopted in these studies; therefore, they did not take the meaning of the selected Quran verses into consideration (Abdullah & Omar, 2011; Zulkurnaini et al., 2012).
This research attempts to identify emotions elicited in the four groups described above and compare them. Additionally, this study uses a specific set of Quran verses and an
experienced reciter. This study explores the possibilities for using Quranic recitations as a means of reducing stress and anxiety through listening therapy.
1.3 RESEARCH OBJECTIVES
This study is focused on understanding the relationship between Quranic recitations and their emotional effects on different groups. The basis for deriving the subjects‘
emotional profiles is the affective space model (ASM), which primarily relies on two factors: valence and arousal. Two feature extraction methods, the mel-frequency cepstral coefficient (MFCC) and the kernel density estimate (KDE) algorithms were used with the multilayer perceptron (MLP) classifier to produce emotion classifiers.
The objectives of this study are the following:
1. To determine the emotional effects of listening to the Quran for the four study groups.
2. To compare the degree of accuracy of emotions extracted using EEG signals with those extracted using ECG signals.
3. To compare the performance of extracting emotions using psychological methods through computational methods.
4. To use computational methods to profile subjects‘ emotions while listening to recitations of the Quran.
1.4 RESEARCH QUESTIONS
This study attempts to answer the following questions:
1. What are the emotional effects of listening to the Quran on the four groups?
2. How accurate are EEG and ECG with respect to extracting emotions?
3. How well does extracting emotions using psychological methods compare to extracting emotions using computational methods?
4. What is the relationship between precursor emotions and listening to recitations of the Quran?
1.5 RESEARCH HYPOTHESES
This study examines four basic emotions: calmness, happiness, fear and sadness. The detection of these emotions is carried out using EEG and ECG signals. The extraction of these emotions is important for understanding the relationship between listening to recitations of the Quran and its effects on listeners. The hypotheses are as follows:
a) Listening to Quranic recitations will bring calmness and tranquillity to listeners.
b) The content or the meaning of the Quran verses have a higher impact on groups who understand their meaning than groups who do not.
c) There is no difference in emotional responses between male listeners and female listeners.
d) The international affective picture system will elicit subjects‘ emotions.
1.6 SIGNIFICANCE OF THE STUDY
The research is expected to make contributions to both human well-being and the discipline of computer science, particularly in understanding human brain and heart signals when listening to recitations of the Quran. Therefore, studying human emotions and their relationship to Quranic recitations may help people to release stress. Furthermore, it can help psychologists, health practitioners and therapists to understand the effects of listening to the Quran. The computational techniques
employed in this research may help to interpret individuals‘ emotions using EEG and ECG signals. The feed-forward artificial neural network multilayer perceptron (MLP), fast Fourier transforms (FFT), the mel frequency cepstral coefficient (MFCC), the kernel density estimate (KDE) and averaging methods are adopted to identify the relationships between human emotions and Quranic recitations. The analysis techniques introduced in this research are based on the Affective Space Model (ASM) and may help other computer scientists improve their methodologies and understanding in other disciplines. Affective computation has been gaining interest among researchers. Affective computation is a branch of artificial intelligence that may help in developing computer systems that can understand human emotions and thus enhance user experiences.
The impact of this research is that it will bring a new approach to meditation based on listening to Quranic recitations that can be applied in different disciplines to heal and treat people. Such an approach can help practitioners analyse patients more affordably. Moreover, the collected data will be analysed and documented to scientifically indicate that listening to Quranic recitations is useful as a medicinal approach.
Human beings are constantly seeking a superior power to fill the spiritual gaps in their lives; a superior power is necessary for humankind to feel safe and protected, and this superior power is usually referred to as Allah. Human beings seek help and refuge from Allah to enlighten their paths via communications such as praying and praising (Atarodi et al., 2012).
Muslims believe that Allah sent down the Quran to his prophet Muhammad S.A.W.
through Gabriel to spread guidance and peace across the land, and that the Quran contains the
words of Allah. Muslims believe that listening to the words of God can bring tranquility and peace to their hearts and minds. The Quran helps maintain a balance between various aspects of human life (Ebrahimi, 2011). In Islamic medicine, the Quran has been often used to promote health and treat people who suffer from stress and psychological disorders (Mahjoob et al., 2014).
One study concluded that listening to recitations of the Quran can promote mental health (Mahjoob et al., 2014), (Zulkurnaini et al., 2012). This statement can be reinforced by saying that one way to gain mental and spiritual relaxation is by listening to Quranic recitations. This is further emphasized by the results from an experiment in (Khan et al., 2010), which suggested that listening to the Quran can reduce stress and tension if adopted on a regular daily basis.
The study by (Siahpoosh, 2012) states that the Quran contains many verses that emphasize promoting physical and spiritual health. Another found that the Quran acts as a medicine for those who suffer from materialism and unconditional surrender to lust (Hamidi et al., 2010). Finally, listening to Quranic recitations reduces anxiety in athletes before competitions (Mottaghi et al., 2011).