Anxiety and EEG Frontal Theta-Beta Ratio Relationship Analysis Across

Personality Traits During HDR Affective Videos Experience

Majid Riaz and Raffaele Gravina

Department of Informatics, Modeling, Electronics and System Engineering, University of Calabria,Rende, Italy

Keywords:

Electroencephalography, Theta-Beta Ratio, Anxiety, High Dynamic Range, Personality Traits,

Valence-Arousal Plane.

Abstract:

To comprehend the intricate interplay between the frontal regions of the brain, anxiety, and their connection

with individual personality traits holds promising potential for developing contemporary, personality-aware

interventions in anxiety healthcare design. Traditionally, emotional content with low dynamic range has been

employed for anxiety assessment, with personality traits serving as mediators. This paper introduces a novel

approach, examining the inﬂuence of personality traits on the relationship between anxiety and electroen-

cephalography (EEG) frontal Theta-Beta Ratio (TBR) during high dynamic range (HDR) arousal-valence

affective content exposures. Twenty-seven subjects were categorized into ﬁve groups based on big ﬁve per-

sonality scores, further subdivided into high and low personality traits. Correlation analyses were conducted

separately for the right and left frontal regions. Across four HDR video clips, each positioned within a distinct

valence-arousal plane, it was observed that for High Arousal High Valence (HAHV) and High Arousal Low

Valence (HALV), most personality trait groups exhibited a negative correlation between anxiety and frontal

TBR, while a positive correlation was noted for Low Arousal Low Valence (LALV) and Low Arousal High

Valence (LAHV) HDR emotional content. The ﬁndings indicate that the big ﬁve personality traits are the

pivotal intermediate psychological factors affecting alterations in brain activity and anxiety.

1 INTRODUCTION

In the course of daily living, it often happens with

someone to experience anxiety and stress, marked

by various emotional states that signiﬁcantly inﬂu-

ence the overall quality of life. Constant anxious

feeling can cause the sever depression that ultimately

impacts the mental and physical health of an indi-

vidual(Gavrilescu and Vizireanu, 2019). Difference

states of Anxiety and stress are implicated to assess

the onset and progressions of various diseases like

cardiovascular, neurodegenerative and immunologi-

cal disorder(Reiche et al., 2004). Social anxiety dis-

order (SAD) is considered to be one of the major rea-

son for emotional disturbances, cognitive deteriora-

tion and brain disorder(Al-Ezzi et al., 2020). Anxi-

ety constitutes internal dimensions within the intricate

spectrum of human emotions, intricately connected to

both the psychological and physiological characteris-

tics inherent to each individual.

Personality traits are an important but often ne-

glected mediators of emotional well being, physio-

https://orcid.org/0000-0002-2257-0886

logical responses and anxiety levels. There are dif-

ferent psychological dimensions of human personal-

ity but most importance dimensions are those that are

described by ﬁve factors model (FFM) that denotes

the neuroticism (N), extroversion (E), openness (O),

agreeableness (A) and conscientiousness (C) (Mc-

Crae and John, 1992). These personality traits have

been shown to inﬂuence the physiological responses,

stress and cognitive declines among individuals(Yan

et al., 2019). Several researches have been conducted

to investigate the relationship between the personal-

ity traits and chronic diseases i.e., personality traits

and mental disorder(Kotov et al., 2007), personality

traits and anxiety about aging and health(Harris and

Dollinger, 2003; Nik

cevi

c et al., 2021; Bienvenu and

Brandes, 2005). While the connection between per-

sonality and anxiety has been extensively studied, the

underlying mechanism remains unclear. Existing evi-

dence is insufﬁcient to ﬁrmly establish how personal-

ity might serve as a mediator for anxiety triggered by

different audio-visual emotional stimuli.

Physiological signals based anxious levels have

been analyzed during various exposures. For example

Riaz, M. and Gravina, R.

Anxiety and EEG Frontal Theta-Beta Ratio Relationship Analysis Across Personality Traits During HDR Affective Videos Experience.

DOI: 10.5220/0012547100003699

Paper published under CC license (CC BY-NC-ND 4.0)

In Proceedings of the 10th International Conference on Information and Communication Technologies for Ageing Well and e-Health (ICT4AWE 2024), pages 27-36

ISBN: 978-989-758-700-9; ISSN: 2184-4984

in (

Salkevicius et al., 2019), physiological signals like

galvanic skin response (GSR), blood volume pulse

(BVP), skin temperatures have been recorded during

virtual reality therapy exposure to assess the various

anxiety levels. Electroencephalography (EEG) sig-

nals based major depressive disorder (MDD)(Greco

et al., 2021) and human anxiety(Muhammad and Al-

Ahmadi, 2022) is analyzed. In(Li et al., 2023),

during VR scenes have used to record the physi-

ological signals to assess the anxiety and depres-

sion levels. ECG and respiration signals have been

recorded whiles subjects were continuously watching

the video clips and their anxiety was measured (El-

gendi et al., 2022). Numerous cutting-edge studies

have demonstrated that analyzing the electrical ac-

tivity of the brain through EEG offers valuable in-

sights and discernible frequency patterns for under-

standing anxiety states in human. Among numerous

EEG based biomarkers for anxiety assessment, theta-

beta ratio can serve as promising neurophyiological

bio-indicators to better understand the anxiousness of

human.

EEG is accessible and cost effective technology to

investigates the neuronal correlates of the theta beta

ratio (TBR) and anxiety. TBR speciﬁes the deviations

in brainwave frequency patterns and these alteration

in neural regions act as indicators of the stress and

anxiety. TBR indicates the cortical and sub-cortical

interactions and numerous studies have shown its cor-

relation with the anxiety and attention control(Wei

et al., 2020). (Poole et al., 2021) states that tempera-

mentally shy children with enhanced theta beta ration

are at greater risk of anxiety. In (Chang and Choi,

2023), alpha beta, alpha theta and theta beta ratios

have been identiﬁed as biomarkers of depression.

Psychological and physical symptoms of the anx-

iety disorders can be decoded through the variety

of audio and video emotional stimuli(Scibelli et al.,

2016). As previously mentioned, decoding induced

anxiety has utilized facial, audio, and visual cues,

yet a comprehensive understanding remains a sub-

ject of ongoing exploration. The emergence of High

Dynamic Range (HDR) multimedia technology has

prompted investigations in healthcare systems within

the realms of medical, computer science, and AI-

related research communities. Offering immersive

and visually captivating content, HDR multimedia

stands out for its capability to encompass the com-

plete range of light and color information inherent

in real-world scenes(Riaz et al., 2023). In past, very

few studies have employed the HDR multimedia for

health related emotion recognition using EEG(Riaz

et al., 2021a) and emotional experience analysis (Riaz

et al., 2021b), quality of experience analysis and per-

ceptual experience analysis using EEG signals(Moon

and Lee, 2015b). Likewise, investigations into the

connectivity among various brain regions have been

conducted to unravel the cognitive processes asso-

ciated with the tone-mapped HDR visual experi-

ence(Moon and Lee, 2015a). To the best of our

knowledge there is no such work that have employed

EEG responses inﬂuenced by HDR emotional stimuli

to evaluate the induced anxiety.

While behavioral and physical factors impacting

anxiety levels have been extensively explored, the

analysis of anxiety levels through the lens of the Big

Five personality traits remains underexamined. Pre-

vious research has physiologically connected person-

ality traits to the emergence of various diseases by

identifying biomarkers. For instance, the role of per-

sonality traits such as extroversion and neuroticism

in managing stress via biofeedback mechanisms like

heart rate has been investigated(Bequet et al., 2022).

The ﬁndings of(Olofsson et al., 2008), uncovered

that the variation in arousal and valence levels pro-

vides crucial insights into the neural underpinnings of

anxiety triggered by emotional stimuli. While prior

research has explored the relationship between psy-

chological factors and anxiety through the dimensions

of valence and arousal emotions, the extent to which

speciﬁc personality traits serve as stronger mediators

in this connection has not been explicitly and clearly

delineated.

In this study, we have addressed the challenges

given for assessing the anxiety assessment by relat-

ing to psychological contributor of human person-

ality through their neural responses. Previous re-

search have not neurologically correlated the person-

ality traits with the anxiety through such strong emo-

tional content. To the best of our knowledge, this

study marks the ﬁrst attempt to utilize affectiv HDR

valence and arousal stimuli in investigating the cor-

relation between personality traits and anxiety levels.

This exploration involves the analysis of anxiety re-

sponses and EEG reactions elicited during exposure

to novel visual stimuli.

1.1 Motivation

Anxiety related problems cost the health and over-

all quality of life of an individual with various de-

mographic details. To investigate the dynamic in-

terplay between the personality traits, microbiologi-

cal responses during immersive and visually appeal-

ing multimedia encounters could be of paramount

importance for health related research communities.

Although, previous researches have determined the

personality and anxiety link separately, but integrat-

ICT4AWE 2024 - 10th International Conference on Information and Communication Technologies for Ageing Well and e-Health

ing it with HDR stimuli and brain responses remains

largely under-explored area of research. Examin-

ing the connection between personality traits, partic-

ularly theta-beta EEG frontal correlates, and the ex-

pression of anxiety, this research aims to shed light

on both affective computing and the practical impli-

cations for subjective well-being and mental health.

The key discoveries from this study provide valuable

insights for crafting personality-aware interventions

aimed at controlling anxiety. This involves the cre-

ation of interventions that are unassuming, practical,

and emotionally engaging in the multimedia context,

contributing to the development of customized and ef-

fective strategies.

1.2 Contributions

In the present study, we employed personality traits

as mediators, serving as intermediary factors that im-

pact the underlying connection between anxiety and

the spectrum of brain activity, as indicated by the

TBR observed in EEG recordings from both the right

and left regions of the brain during exposure to novel

HDR emotional content. Our primary contributions

are delineated as follows:

1. We introduced the novel application of the HDR

valence arousal stimuli to investigate the rela-

tionship between personality traits and anxiety

level by expanding the scope of experiment i.e.,

paradigm shift from emotional research to anxi-

ety and personality related research

2. We analyzed the anxiety manifestation and EEG

frontal channels bands extracted in response to the

HDR visual experience to explore a complex rela-

tionship

3. Finally, we introduced a conceptual framework

that integrates personality traits with anxiety and

TBR derived from frontal EEG recordings on both

the right and left sides, collected in response to

HDR valence arousal experiences. The statisti-

cal analysis revealed a notable role of personality

traits as mediators for inﬂuencing anxiety.

Remainder of the paper is organized as fallows. De-

tailed proposed methodology for anxiety and person-

ality traits analysis is discussed in section II. Section

III provides a comprehensive analysis of the statisti-

cal results while section IV is about the discussion.

Final section V concludes the paper and presents fu-

tures recommendations.

2 PROPOSED METHODOLOGY

Fig 1 illustrates the step-by-step process employed to

investigate the connection between anxiety and per-

sonality traits via physiological reactions. The pro-

cedure encompasses the assessment of the Big Five

personality traits through a questionnaire, followed

by stimuli selection, EEG signal acquisition, and the

completion of the STAI form. Subsequently, the pro-

cess involves the extraction of power spectral density

bands and statistical analysis. The speciﬁcs of each

step are delineated below.

2.1 HDR Valence Arousal Stimuli

Selection

Total four video clips i.e., one musical clip and 3 from

Hollywood movies in HDR10 version with 4K reso-

lution and 10 bits color depth were selected for the

experiment. Before ﬁnalizing the stimuli, subjective

pilot study was conducted in lab to rate the valence

and arousal level against each video using 9-point self

assessment manikin(SAM) scale(Peacock and Wong,

1990). To validate the valence and arousal ratings, we

mapped the ratings given by the viewers to valence

arousal (V-A) space and each video was successfully

mapped into one of four quadrant of V-A space. Va-

lence scale is between negative(low valence) to pos-

itive emotions (High valence) while arousal scale is

between calm or relaxation (low arousal)to excite-

ment state (high arousal). In High arousal high va-

lence quadrant (HAHV) a musical clip (LG Jazz Mu-

sic) while in high arousal low valence (HALV) a hor-

ror clip from movie nun version 2018 were mapped.

Similarly, in 3rd quadrant of V-A emotional dimen-

sion that is low low valence (LALV), a clip from the

movie Gladiator version(2000) while in low arousal

high valence (LAHV) a clip from john vick version

(2014) were mapped. Duration of each clip was 90

seconds.

2.2 Subjective Self Assessment

2.2.1 Subjective Personality Assessment

To measure personalized personality traits, we em-

ployed a questionnaire, speciﬁcally the Big Five Per-

sonality Test, which evaluates the structure of person-

ality(Butt et al., 2020). This questionnaire gauges

ﬁve personality dimensions based on user responses

to each of its 50 questions. Users rate their agreement

on a scale of 1 to 5, where 1 signiﬁes disagreement, 2

indicates slight disagreement, 3 represents neutrality,

4 signiﬁes slight agreement, and 5 denotes agreement.

Anxiety and EEG Frontal Theta-Beta Ratio Relationship Analysis Across Personality Traits During HDR Affective Videos Experience

Figure 1: Proposed methodology followed for relationship analysis between personality traits, EEG frontal TBR and anxiety.

The questionnaire’s 50 questions are categorized into

ﬁve groups, aligning with the ﬁve personality traits

(neuroticism, extroversion, openness to experience,

agreeableness, and conscientiousness). Each person-

ality trait’s overall score ranges from 0 to 40. To fur-

ther categorize traits as either low or high, we calcu-

lated the mean value for each personality trait score,

labeling traits with scores below the mean as low traits

and vice versa.

2.2.2 Subjective Anxiety Assessment

To measure the personalized anxiety levels, State-

Trait-Anxiety-Inventory (STAI) questionnaire was

employed(Mokhtari et al., 2023). STAI tool reﬂects

the anxiety levels aroused in response to various phe-

nomena. This tools comprises 40 questions and users

have to mark the anxiety perceived on a scale of 1 to 4

where 1 = almost never, 2 = sometimes,3 = often and

4 = almost always. Finally anxiety scores were cal-

culated for each personality traits based on the rating

given by users during visual experience.

2.3 Data Sensing Mechanism

2.3.1 Test Subjects

A total of 27 individuals took part in the research,

comprising 17 males and 10 females with ages rang-

ing from 22 to 30 years. All participants were in

good health and willingly participated in the exper-

iment. Prior to the commencement of the study, it

was ensured that none of the volunteers had previ-

ously participated in any anxiety-related studies, and

it was their ﬁrst exposure to videos in HDR format.

2.3.2 Experimental Material

In block 3 of the Fig 1, sensor used for the sen-

sory data acquisition have been shown. EEG signals

were captured utilizing the commercially accessible

5-channel Emotive Insight Headset. The Emotive In-

sight is a wireless headset with ﬁve channels (AF4,

T7, PZ, T8, AF3) and two reference electrodes de-

signed to capture brain activity and translate it into

meaningful information(He et al., 2023). The Emotiv

Insight recorded brain waves at a sampling rate of 128

samples per second. It features a 16-bit analog-to-

digital converter (ADC) activated during signal trans-

mission. The headset comprises ﬁve electrode posi-

tions and two reference electrodes situated at the left

mastoid bone. EEG signals were recorded from the

subjects’ scalp using the EMOTIVE Xavier test bench

v.3.1.21.

To present the valence and arousal emotional con-

tent, a 55-inch TCL 55R16 UHD resolution LED-

LCD screen with Full-Array Local Dimming technol-

ogy was utilized. The display adhered to UHD pre-

mium speciﬁcations(Association et al., 2013) and had

the capability to reach a peak brightness exceeding

1000 cd/m².

ICT4AWE 2024 - 10th International Conference on Information and Communication Technologies for Ageing Well and e-Health

Figure 2: Experimental protocol followed for relationship analysis between personality traits, EEG frontal TBR and anxiety.

2.3.3 Experimental Protocol

Detailed experimental procedure followed for the data

acquisition is shown in the Fig 2. First of all, subjects

were provided with the Big ﬁve personality test ques-

tionnaire a day before the subject was subjected to

the experiment. Each of the subjects have to watch 4

HDR valence and arousal clips in a sequence i.e., ﬁrst

clip was presented and EEG signals were recorded.

after watching the ﬁrst clip, subject have to rate is anx-

iety level on the scale printed on the STAI question-

naire. Following the ﬁrst trial, next clip was played

and same procedure was followed. So for one subject

there were 4 trials and time consumed for one experi-

ment was almost 30-35 minutes.

2.4 Signals Preprocessing

To analyze the EEG recordings for better anxiety

analysis, acquired physiological signals were prepos-

sessed before PSD extraction. EEG signals recorded

through electrodes of Emotive have a DC offset in

their values which shall be discarded before do-

ing analysis either in time domain or frequency do-

main(Saeed et al., 2020). For DC offset removal,

mean value of data from each channel was subtracted

from sample values of that channel. Additionally,

sensors used for recording is highly sensitive and have

chances to be interfered from external environment

resulting into high signal to noise ratio (SNR) which

may distort acquired data.

2.5 PSD Extraction

Frequency domain features of EEG signals were ex-

tracted by computing power spectrum density (PSD)

using Welch’s method with window length of 128

samples per seconds and without overlapping. An

open source MATLAB toll Brainstorm(Tadel et al.,

2011) was used for frequency domain analysis.

Changes in the Power of EEG signals were averaged

over ﬁve frequency bands of Delta (2-4) HZ, Theta (5-

9) HZ, Alpha (8-12) HZ, Beta (15-29) HZ, Gamma

(30-59) HZ for each channel. Therefore, total of 25

EEG frequency bands (5 electrodes × 5 Bands) were

extracted from each EEG signal. In this study we only

used the frontal theta beta ratio i.e., theta beta ratio for

AF3 and AF4 channels as recently studies have shown

that frequency bands speciﬁcally theta and beta waves

in the frontal regions shows the relationship between

the anxiety manifestations and symptoms reported by

the individuals.

2.6 Statistical Analysis

A statistical method was employed to investigate the

connection between personality traits and anxiety lev-

els triggered by High Dynamic Range (HDR) emo-

tional stimuli related to arousal and valence. Ini-

tially, the theta and beta ratio was computed for the

AF4 and AF3 channels of the Emotiv Insight head-

set. Subsequently, the correlation was determined be-

tween the Theta/Beta Ratio (TBR) calculated for AF4

channels and anxiety ratings provided by subjects ex-

Anxiety and EEG Frontal Theta-Beta Ratio Relationship Analysis Across Personality Traits During HDR Affective Videos Experience

hibiting varying personality traits in response to af-

fective HDR clips. Subjects were categorized into

two groups for each personality trait based on their

self-reported ratings, such as low neuroticism (LN)

and high neuroticism (HN), introversion and extrover-

sion, low openness (LO) and high openness (HO), low

agreeableness (LA) and high agreeableness (HA), and

ﬁnally, low conscientiousness (LC) and high consci-

entiousness (HC) for the respective personality traits.

The same procedure was replicated for the AF3 chan-

nel for the statistical analysis.

3 EXPERIMENTAL RESULTS

This section presents the outcomes of a correlation-

based statistical analysis examining the relationship

between anxiety and personality using EEG frontal

TBR. We conduct distinct analyses for both the right

(AF4 electrodes) and left (AF3 electrodes) frontal re-

gions of the brain. Among the 27 participants, 15 ex-

hibited high levels of neuroticism, while 12 demon-

strated lower neurotic tendencies. Additionally, 13

participants were identiﬁed as extroverted, contrast-

ing with 14 who displayed introverted traits. In terms

of openness to experience, 18 participants exhibited

a higher inclination, while 9 exhibited a lower incli-

nation. For agreeableness, 24 participants displayed

higher levels, with 3 displaying lower levels. Regard-

ing conscientiousness, 18 participants exhibited high

conscientiousness, whereas 9 displayed lower consci-

entiousness. The hierarchical distribution depicted in

Fig. 3 illustrates how 27 participants were categorized

into groups with high and low traits, according to their

questionnaire scores.

3.1 Personality Traits-Wise Statistical

Analysis for the EEG AF4 TBR and

Anxiety

3.1.1 Higher Personality Traits-Wise Results

Anxiety from a personality perspective during expo-

sure to novel valence and arousal-related HDR con-

tent is essential, given the inﬂuence of personality on

both brain function and the expression of anxiety. In

this section, we explore the correlation between anx-

iety and EEG TBR speciﬁcally for the AF4 channel.

The correlation outcomes for individuals with higher

personality traits are presented in Table 1, where cor-

relation coefﬁcients range from -1 to 1. A correla-

tion of -1 indicates a perfect negative linear relation-

ship, signifying reciprocal variation i.e., if one vari-

ables increases other variable tends to decrease pro-

portionally. A correlation coefﬁcient of 1 signiﬁes a

complete positive linear connection, suggesting that

as TBR increases, anxiety aslo tends to increase, and

vice versa—a directly proportional relationship. On

the other hand, a correlation coefﬁcient of 0 indicates

the absence of any correlation. Notably, a robust neg-

ative correlation emerged between TBR and anxiety

in response to High Arousal High Valence (HAHV)

and High Arousal Low Valence (HALV) stimuli for

individuals with high-order big ﬁve personality traits.

This implies that an increase in AF4 TBR corresponds

to a relative increase in anxiety levels among individ-

uals with high personality traits exposed to HAHV

and HALV HDR stimuli. Conversely, a positive cor-

relation between TBR and anxiety was observed for

Low Arousal Low Valence (LALV) and Low Arousal

High Valence (LAHV) stimuli across all high-order

personality traits, except for the High Conscientious-

ness (HC) group, which exhibited a negative corre-

lation for the LALV HDR clip. Positive correlations

indicate a reciprocal relationship: an increase in AF4

TBR corresponds to a decrease in anxiety, and vice

versa.

3.1.2 Lower Personality Traits-Wise Results

With the exception of LA, the remaining personality

traits, namely LN, I, LO, LC, demonstrated a nega-

tive correlation between TBR and anxiety in response

to HAHV stimuli. Conversely, for High HALV af-

fective HDR content, LN and LO exhibited a posi-

tive correlation, while the others displayed a negative

correlation. Interestingly, all lower personality traits

displayed a positive correlation between anxiety and

TBR for LALV and LAHV, except for LO, which re-

vealed a negative correlation during the viewing of

the LALV HDR video clip. A summarized overview

of the relationship between anxiety and TBR is pre-

sented in Table 2.

Table 1: Summary of EEG AF4 TBR and anxiety corre-

lation across 15 HN, 13E, 18 HO, 24 HA, 18 HC higher

personality traits groups.

HDR Affective Clips

Traits TBR HAHV HALV LALV LAHV

Anxiety Anxiety Anxiety Anxiety

HN θ/β -0.38 -0.40 0.20 0.42

E θ/β -0.57 -0.27 0.02 0.37

HO θ/β -0.25 -0.30 0.31 0.15

HA θ/β -0.31 -0.20 0.15 0.25

HC θ/β -0.43 -0.20 0.00 0.21

ICT4AWE 2024 - 10th International Conference on Information and Communication Technologies for Ageing Well and e-Health

Participants (27)

Neuroticism

High

Low

Extroversion

Extroverted

Introverted

Openness

High

Low

Agreeableness

High

Low

Conscientiousness

High

Low

Figure 3: Placement of 27 subjects into Big ﬁve personality traits groups

Figure 3: Placement of 27 subjects into Big ﬁve personality traits groups.

Table 2: Summary of EEG AF4 TBR and anxiety correla-

tion across 12 LN, 14 I, 9 LO, 3 LA, 9 LC lower personality

traits groups.

HDR Affective Clips

Traits TBR HAHV HALV LALV LAHV

Anxiety Anxiety Anxiety Anxiety

LN θ/β -0.25 0.02 0.07 0.19

I θ/β -0.19 -0.17 0.29 0.06

LO θ/β -0.55 0.02 -0.56 0.53

LA θ/β 0.80 -0.30 0.16 0.84

LC θ/β -0.10 -0.25 0.45 0.19

Table 3: Summary of EEG AF4 TBR and anxiety corre-

lation across 15 HN, 13E, 18 HO, 24 HA, 18 HC higher

personality traits groups.

HDR Affective Clips

Traits TBR HAHV HALV LALV LAHV

Anxiety Anxiety Anxiety Anxiety

HN θ/β -0.40 -0.48 0.06 0.55

E θ/β -0.59 -0.34 -0.23 0.18

HO θ/β -0.29 -0.31 0.34 -0.1

HA θ/β -0.32 -0.27 0.02 0.34

HC θ/β -0.45 -0.29 -0.18 0.26

3.2 Personality Traits-Wise Statistical

Analysis for the EEG AF3 TBR and

Anxiety

In this section, we present the correlation results be-

tween anxiety and the TBR of left prefrontal lobe of

the brain that is acquired through AF3 channel of the

sensors.

3.2.1 Higher Personality Traits-Wise Results

For AF3, we found some interesting results. All

higher-order personality traits exhibited the nevagtive

correlations between TBR and anxiety when exposed

to the HAHV and HALV. However in case of the

LALV, E and HC indicated negative correlation while

for LAHV only HO exhibited the negative correla-

tion between the TBR and the anxiety. A summarized

Table 4: Summary of EEG AF3 TBR and anxiety correla-

tion across 12 LN, 14 I, 9 LO, 3 LA, 9 LC lower personality

traits groups.

HDR Affective Clips

Traits TBR HAHV HALV LALV LAHV

Anxiety Anxiety Anxiety Anxiety

LN θ/β -0.19 0.02 0.13 0.14

I θ/β -0.15 -0.01 0.20 0.21

LO θ/β -0.55 -0.13 -0.50 0.61

LA θ/β 0.80 0.11 0.17 0.69

LC θ/β -0.02 -0.21 0.32 0.25

overview for the AF3 TBR and anxiety relationship

for higher personality traits have been shown in the

Table 3.

3.2.2 Lower Personality Traits-Wise Results

Table 4 provides a condensed overview of the cor-

relation between EEG AF3 TBR and anxiety con-

cerning lower-order personality traits. In the context

of HAHV HDR content, there is a positive correla-

tion between TBR and anxiety for LA. Similarly, dur-

ing HALV exposure, LN and LA exhibited a posi-

tive correlation. For LALV, all lower personality trait

groups, except LO, displayed a positive correlation,

while for LAHV, all lower personality trait groups in-

dicated a positive correlation. Notably, LA consis-

tently showed positive correlations across all four va-

lence and arousal-related HDR content.

4 DISCUSSION

With the exception of uncovering the connection be-

tween anxiety and TBR in response to HDR valence-

arousal emotional content from the right and left pre-

frontal lobes of the brain, this study made notable ob-

servations in this context. We provided separate re-

sults for each personality trait while considering both

frontal regions of the brain. The following section of-

fers a comprehensive discussion of the statistical anal-

ysis First of all, subjects were divided into ﬁve groups

Anxiety and EEG Frontal Theta-Beta Ratio Relationship Analysis Across Personality Traits During HDR Affective Videos Experience

based on the big ﬁve personality traits. Each group

was further subdivided into low/high trait group based

on the big ﬁve personality scores. In order to assess

the impacts of personality traits on brain activities and

anxiety either for low of high group, we check the cor-

relation for right and right prefrontal region of brain

separately. As recent studies have proved that these

cortex of human brain are crucial regions and contain

neurophysiological markers of anxiety and depressive

disorder(Shanok and Jones, 2023). Right prefrontal

activity is covered through the AF4 channel while

left prefrontal lobe of the brain is captured through

the AF3 channels of the 5-channel EMOTIV Insight

headset. From each channels we calculated the PSD

and frequency spectrum. It is important to note that

both anxiety and EEG waves were measured while

subjects were watching a novel valence and arousal

related content in HDR version.

The correlation outcomes indicated that both high

and low personality traits impact the association be-

tween anxiety and EEG TBR differently in both brain

regions. Among the high personality traits group,

HAHV and HALV HDR clips elicited anxiety and

brain activity in a similar manner, with TBR and anx-

iety showing a negative correlation. Conversely, for

LALV and LAHV, there is a positive correlation, ex-

cept for the HC group, which demonstrated a nega-

tive correlation for LALV, indicating almost no corre-

lation (correlation coefﬁcient = -0.005). For the low

personality traits group, exposure to HAHV triggered

anxiety and brain activity in opposite direction as in-

dicated by the negative correlation, except for the LA

group (correlation coefﬁcient = 0.802). During LALV

and LAHV emotional experiences, anxiety and TBR

exhibited an direct relationship, except for LO, where

the correlation coefﬁcient was -0.56. In summary,

it can be inferred that the right and left EEG frontal

TBR and anxiety during LALV and LAHV emotional

experiences showed positive correlation for both per-

sonality traits groups, except for the HC group during

LALV and the LO group during LALV.

We delved into the detailed effects of higher and

lower personality trait groups on the left prefrontal re-

gion and anxiety in response to the same emotional

content. The high personality traits group experi-

enced negative correlations between TBR and anxi-

ety while viewing HAHV and HALV HDR emotional

content. During LALV, only E and HC showed a in-

verse direction of relationship, while during LAHV,

only the HO group experienced a negative direction

of relationship between the left frontal TBR and anx-

iety. Lastly, we investigated how lower personality

traits inﬂuence the left prefrontal cortex of the brain

and anxiety during valence-arousal related HDR mul-

timedia content. Some noteworthy and intriguing out-

comes emerged, indicating that the LA personality

traits groups exhibited the direct relationship between

the left front TBR and anxiety during all four trials.

While duirng HAHV and HALV exposures, inverser

relationship was notices across all lower personality

traits groups except LN during HALV (correlation co-

efﬁcient = 0.02 ). Similarly, for the LALV and LAHV

HDR experiences, all groups showed an direct rela-

tionship between the left frontal TBR and anxiety, ex-

cept for LO (correlation coefﬁcient = -0.50).

5 CONCLUSION

In this investigation, we explored the impact of differ-

ent big ﬁve personality traits groups on EEG frontal

Theta-Beta Ratio (TBR) and anxiety responses to

novel HDR valence and arousal affective content.

Key ﬁndings unveiled distinct correlation patterns for

both higher and lower personality traits groups, shed-

ding light on how various dimensions of an individ-

ual’s personality inﬂuence emotional and neural re-

sponses. These preliminary outcomes offer valuable

insights for developing personalized interventions for

anxiety control that consider the interplay between

personality and brain activity. Furthermore, this study

contributes to ﬁelds like affective computing for men-

tal health, providing a nuanced understanding of the

intricate relationships between personality traits, anx-

iety, and neural activity. Overall, this research lays

the groundwork for future investigations that can be

enhanced by incorporating additional demographic,

psychological, and physiological details.

ACKNOWLEDGEMENTS

We acknowledge co-funding from Next Generation

EU, in the context of the National Recovery and

Resilience Plan, Investment PE8 – Project Age-It:

“Ageing Well in an Ageing Society”. This resource

was co-ﬁnanced by the Next Generation EU [DM

1557 11.10.2022]. The views and opinions expressed

are only those of the authors and do not necessar-

ily reﬂect those of the European Union or the Eu-

ropean Commission. Neither the European Union

nor the European Commission can be held responsi-

ble for them. This work has been also partially sup-

ported by the Italian MUR, PRIN 2022 Project “CO-

COWEARS” (A framework for COntinuum COm-

puting WEARable Systems), n. 2022T2XNJE, CUP:

H53D23003640006

ICT4AWE 2024 - 10th International Conference on Information and Communication Technologies for Ageing Well and e-Health

REFERENCES

Al-Ezzi, A., Kamel, N., Faye, I., and Ebenezer, E. G. M.

(2020). Eeg frontal theta-beta ratio and frontal mid-

line theta for the assessment of social anxiety disorder.

In 2020 10th IEEE International Conference on Con-

trol System, Computing and Engineering (ICCSCE),

pages 107–112. IEEE.

Association, W. M. et al. (2013). Declaration of

helsinki, ethical principles for scientiﬁc requirements

and research protocols. Bull World Health Organ,

79(4):373.

Bequet, A. J., Hidalgo-Munoz, A. R., Moreau, F., Quick, J.,

and Jallais, C. (2022). Subtle interactions for distress

regulation: Efﬁciency of a haptic wearable accord-

ing to personality. International Journal of Human-

Computer Studies, 168:102923.

Bienvenu, O. J. and Brandes, M. (2005). The interface of

personality traits and anxiety disorders. Primary Psy-

chiatry.

Butt, A. R., Arsalan, A., and Majid, M. (2020). Multimodal

personality trait recognition using wearable sensors in

response to public speaking. IEEE Sensors Journal,

20(12):6532–6541.

Chang, J. and Choi, Y. (2023). Depression diagnosis based

on electroencephalography power ratios. Brain and

Behavior, 13(8):e3173.

Elgendi, M., Galli, V., Ahmadizadeh, C., and Menon, C.

(2022). Dataset of psychological scales and physio-

logical signals collected for anxiety assessment using

a portable device. Data, 7(9):132.

Gavrilescu, M. and Vizireanu, N. (2019). Predicting depres-

sion, anxiety, and stress levels from videos using the

facial action coding system. Sensors, 19(17):3693.

Greco, C., Matarazzo, O., Cordasco, G., Vinciarelli, A.,

Callejas, Z., and Esposito, A. (2021). Discrimina-

tive power of eeg-based biomarkers in major depres-

sive disorder: A systematic review. IEEE Access,

9:112850–112870.

Harris, L. A. and Dollinger, S. M. C. (2003). Individual dif-

ferences in personality traits and anxiety about aging.

Personality and Individual Differences, 34(2):187–

194.

He, C., Chen, Y.-Y., Phang, C.-R., Stevenson, C., Chen,

I.-P., Jung, T.-P., and Ko, L.-W. (2023). Diversity and

suitability of the state-of-the-art wearable and wireless

eeg systems review. IEEE Journal of Biomedical and

Health Informatics.

Kotov, R., Watson, D., Robles, J. P., and Schmidt, N. B.

(2007). Personality traits and anxiety symptoms: The

multilevel trait predictor model. Behaviour research

and therapy, 45(7):1485–1503.

Li, M., Zhang, W., Hu, B., Kang, J., Wang, Y., and Lu,

S. (2023). Automatic assessment of depression and

anxiety through encoding pupil-wave from hci in vr

scenes. ACM Transactions on Multimedia Computing,

Communications and Applications, 20(2):1–22.

McCrae, R. R. and John, O. P. (1992). An introduction to

the ﬁve-factor model and its applications. Journal of

personality, 60(2):175–215.

Mokhtari, R., Ajorpaz, N. M., Golitaleb, M., et al. (2023).

The effects of rosa damascene aromatherapy on anxi-

ety and sleep quality in burn patients: A randomized

clinical trial. Burns, 49(4):973–979.

Moon, S.-E. and Lee, J.-S. (2015a). Eeg connectivity analy-

sis in perception of tone-mapped high dynamic range

videos. In Proceedings of the 23rd ACM international

conference on Multimedia, pages 987–990.

Moon, S.-E. and Lee, J.-S. (2015b). Perceptual experience

analysis for tone-mapped hdr videos based on eeg and

peripheral physiological signals. IEEE Transactions

on Autonomous Mental Development, 7(3):236–247.

Muhammad, F. and Al-Ahmadi, S. (2022). Human

state anxiety classiﬁcation framework using eeg sig-

nals in response to exposure therapy. Plos one,

17(3):e0265679.

Nik

cevi

c, A. V., Marino, C., Kolubinski, D. C., Leach, D.,

and Spada, M. M. (2021). Modelling the contribution

of the big ﬁve personality traits, health anxiety, and

covid-19 psychological distress to generalised anxi-

ety and depressive symptoms during the covid-19 pan-

demic. Journal of affective disorders, 279:578–584.

Olofsson, J. K., Nordin, S., Sequeira, H., and Polich, J.

(2008). Affective picture processing: an integra-

tive review of erp ﬁndings. Biological psychology,

77(3):247–265.

Peacock, E. J. and Wong, P. T. (1990). The stress appraisal

measure (sam): A multidimensional approach to cog-

nitive appraisal. Stress medicine, 6(3):227–236.

Poole, K. L., Hassan, R., and Schmidt, L. A. (2021).

Temperamental shyness, frontal eeg theta/beta ratio,

and social anxiety in children. Child Development,

92(5):2006–2019.

Reiche, E. M. V., Nunes, S. O. V., and Morimoto, H. K.

(2004). Stress, depression, the immune system, and

cancer. The lancet oncology, 5(10):617–625.

Riaz, M., Majid, M., and Mir, J. (2021a). Emotion recogni-

tion using electroencephalography in response to high

dynamic range videos. In 2021 International Confer-

ence on Information Technology (ICIT), pages 565–

570. IEEE.

Riaz, M., Majid, M., and Mir, J. (2021b). Emotional

experience analysis in response to hdr and sdr con-

tent. In 2021 13th International Conference on Qual-

ity of Multimedia Experience (QoMEX), pages 121–

124. IEEE.

Riaz, M., Majid, M., and Mir, J. (2023). High dynamic

range multimedia: better affective agent for human

emotional experience. Multimedia Tools and Appli-

cations, pages 1–16.

Saeed, S. M. U., Anwar, S. M., Khalid, H., Majid, M., and

Bagci, U. (2020). Eeg based classiﬁcation of long-

term stress using psychological labeling. Sensors,

20(7):1886.

Salkevicius, J., Dama

sevi

cius, R., Maskeliunas, R., and

Laukien

e, I. (2019). Anxiety level recognition for

virtual reality therapy system using physiological sig-

nals. Electronics, 8(9):1039.

Scibelli, F., Troncone, A., Likforman-Sulem, L., Vincia-

relli, A., and Esposito, A. (2016). How major depres-

Anxiety and EEG Frontal Theta-Beta Ratio Relationship Analysis Across Personality Traits During HDR Affective Videos Experience

sive disorder affects the ability to decode multimodal

dynamic emotional stimuli. Frontiers in ICT, 3:16.

Shanok, N. A. and Jones, N. A. (2023). Eeg asymmetry

characteristics in relation to childhood anxiety sub-

types: A dimensional approach. Clinical EEG and

Neuroscience, page 15500594221150213.

Tadel, F., Baillet, S., Mosher, J. C., Pantazis, D., and Leahy,

R. M. (2011). Brainstorm: a user-friendly application

for meg/eeg analysis. Computational intelligence and

neuroscience, 2011:1–13.

Wei, H., Chang, L., Huang, Q., and Zhou, R. (2020). Re-

lation between spontaneous electroencephalographic

theta/beta power ratio and test anxiety. Neuroscience

Letters, 737:135323.

Yan, L., Wang, Y., Ding, C., Liu, M., Yan, F., and Guo,

K. (2019). Correlation among behavior, personality,

and electroencephalography revealed by a simulated

driving experiment. Frontiers in psychology, 10:1524.

ICT4AWE 2024 - 10th International Conference on Information and Communication Technologies for Ageing Well and e-Health