key: cord-0516592-fcvyxh3o
authors: Rajabi, Mohammad Sadra; Taghaddos, Hosein; Zahrai, Mehdi
title: Improving Emergency Training for Earthquakes Through Immersive Virtual Environments and Anxiety Tests: A Case Study
date: 2022-05-10
journal: nan
DOI: nan
sha: 62de4903d03af7a5bca63a04452486e289506476
doc_id: 516592
cord_uid: fcvyxh3o

Because of the occurrence of severe and large magnitude earthquakes each year, earthquake-prone countries suffer considerable financial damage and loss of life. Teaching essential safety measures will lead to a generation that can perform basic procedures during an earthquake, which is an important and effective solution in preventing the loss of life in this natural disaster. In recent years, virtual reality technology is a tool that has been used to educate people on safety matters. This paper evaluates the effect of education and premonition on the incorrect decision-making of residents under the stressful conditions of an earthquake. For this purpose, a virtual model has been designed and built from a proposed classroom in a school of the city of Tehran. Accordingly, two educational scenarios, presented in reality and the virtual model respectively, were conducted on a statistical sample of 20 students within the range of 20 to 25 years of age. Within the mentioned sample, the first group of 10 students were taught safety measures in a traditional classroom. The second group of 10 students participated in a virtual classroom. Evaluation tests on safety measures against earthquakes were distributed after two weeks. Furthermore, two self-reporting tests of Depression, anxiety, stress test (DASS), and Beck Anxiety Inventory (BAI) were assigned to the second group to evaluate the effect of foresight under two different scenarios. The results show that educating through virtual reality technology yields a higher performance level relative to the traditional approach to education. Additionally, the ability to detect earthquakes ahead of time is an influential factor in controlling stress and determining the right decisions should the event occur.

In order to mitigate the casualties and financial damages caused by earthquakes, it is important to design and construct buildings that are resistant to natural disasters. Timely reactions and proper evacuation practices following an earthquake are also significant aspects in reducing earthquake-related losses (Alexander, 2012; Bernardini et al., 2016) . For this reason, earthquakeprone countries have issued a list of suggested measures for dealing with dangers during and after an earthquake (Mahdavifar et al., 2009) .Seminars, maneuvers, posters, lectures, and instructive movies have all been suggested as options for conveying these standards to the public. However, these instructive techniques lack depth and emotion and cannot fully convey the message to viewers. As a result, the effectiveness of these teaching initiatives frequently falls below expectations (Chittaro and Ranon, 2009) .

In recent years, new digital technologies such as immersive virtual reality and educational games have been created to address the aforementioned issues (Ott and Freina, 2015) . Immersive Virtual Reality (IVR) is a technique that actively involves a person in a computer-generated virtual world (LaValle, 2017) . More realistic risks and threats may be simulated and presented to participants using this technology, allowing for the creation of desirable scenarios in educational settings. Secondly, Serious Games (SG) are a type of video game in which one of the primary purposes is to educate the player (Wouters et al., 2009) . By utilizing virtual reality technology, serious games can effectively aid in the development of stated educational goals. These technologies have been widely used for educational purposes such as surgery training (Huber et al., 2017) , teaching the repair and maintenance of high-voltage power lines (Ayala García et al., 2016) , and setting pedestrian safety guidelines (Schwebel et al., 2016) . They are also used for engineering purposes, such as programming heavy mobile cranes for construction sites (Kayhani et al., 2018) . However, only a percentage of these tools have been used to teach earthquake safety precautions before, during, and after an earthquake .

The purpose of this study is to demonstrate the capabilities of IVR during earthquake-related safety training and to evaluate how educating, foretelling seismic activity, and simulating earthquakes in IVR affect people's reactions to earthquakes. The intention is to implement virtual reality technology and serious games in an educational way and to increase the amount in which virtual reality is used for safety training. First, IVR, DASS, and BAI are thoroughly examined. The design and development of IVR is then incorporated in a case study based on a proposed high school in Tehran. Participants are divided into two groups and the designed experiment is executed. Finally, the preliminary findings and their implications for safety guidelines and practices on earthquakes are presented.

In the past, schools and offices have taken the initiative to provide earthquake-related safety training, usually in the form of a brief informational session led by a professional. Prior to the 2011 study, unofficial assessments have been conducted in educational settings. Kirikaya and his team explore the impact these teachings have had on students' understanding of earthquakes and related phenomena. According to the findings, almost half of the students are unaware that they are living in a seismically active area (Kirikkaya et al., 2011) .

The San Francisco gulf region in California is one of those seismically active areas. In a study of San Francisco's gulf region by Simpson, residents were taught basic survival and emergency response skills in educational programs in order to be assessed on them. In these programs, CPR, search and rescue techniques, and fire extinguishing lessons were given to locals to prevent earthquake related casualties in the region. According to this research study, there are more than a hundred of these educational programs currently available. Many of them are provided in annual courses that teach and evaluate the aforementioned subjects (Beigi et al., 2022; Moeinifard et al., 2022; Mudiyanselage et al., 2021; Shakerian et al., 2022; Simpson, 2002) .

The astonishing capabilities of virtual reality technology has motivated researchers to use it for immediate reactions before, during, and after an earthquake. In a study by Gong et al. (2015) , a simulation system for an earthquake was shown in a virtual environment. In this study, users were exposed to a three-dimensional virtual earthquake via a monitor located in front of their faces, which enabled them to experience earthquake simulations in a virtual dormitory environment. This simulation system is capable of establishing communication with a virtual environment through Kinect, a line of motion sensing input devices, and via SIGVerse, a type of simulation environment. The results show that powerful earthquakes can be simulated successfully within virtual reality technology and can be useful for earthquake exercises or drills (GONG et al., 2015) .

In another educational approach, Li et al. (2017) used virtual reality to demonstrate how to survive and await rescue teams while remaining inside of a building during an earthquake. The study aimed to show that safety training programs against earthquakes could be presented through the use of virtual reality. By submerging themselves in a virtual environment, users can learn the necessary skills to stay alert during an earthquake in a very realistic manner (Li et al., 2017) . Similarly, Fang et al. (2018) displayed how serious games in virtual reality could be implemented to teach an evacuation of a building during emergency situations, such as fires and earthquakes. For this purpose, a conceptual framework for effectively designing and implementing virtual reality has been created by systematically reviewing technical literature . Lovreglio et al. (2018) has evaluated the theoretical advantages and disadvantages of using serious games based on virtual reality for assessing people's behavior during an earthquake. The residents of a building should follow regulations to protect themselves during the stressful conditions of an earthquake. Precise and repetitive training courses are required to enforce these rules. In this research study, Lovreglio et al. (2018) discuss the main elements of designing and developing a virtual reality educational system based on serious games for the aforementioned conceptual framework . Additionally, Fang et al. (2020) has cited that strengthening timely and intelligent reactions during an earthquake are useful for decreasing the mortality rate and structural damage in another study on Auckland Hospital (Feng, González, Amor, et al., 2020) .

The New Zealand Civil Defense Agency has provided a safety guideline for earthquakes, which includes 32 measures to enact upon during and after the time of an earthquake ("New Zealand Ministry of Civil Defence & Emergency Management, Working from the same page consistent messages for CDEM", 2015). It includes the appropriate responses across a wide spectrum of scenarios, like instructions for home, the workplace, places away from home, coastal areas, mountainous regions, vehicles, and locations with domestic or farm animals. Research by Fang et. al mainly concentrated on indoor environments to evaluate behavioral responses, which were modeled within virtual reality Feng, González, Amor, et al., 2020) .

Fang et al (2021) has also performed a research study on educating earthquake-related conduct to [school-age] children. The researchers in this study claimed that children are very vulnerable and should be educated to create a more knowledgeable society on the subject of earthquakes. Therefore, in order to successfully prepare children against earthquakes, an effective educational tool must be considered. A virtual reality based educational system for serious gaming was developed in this research study based on problem-based learning (PBL). The educational system's structure consisted of three mechanisms: previous education, instant feedback, and evaluation. These mechanisms were deployed after a completed game to assess how to increase a user's level of proficiency. The results indicated that training children in a virtual reality based serious educational game was the most effective way of improving their self-efficiency due to the performance analysis function within the education system (Feng et al., 2021) .

Ensuring safety throughout an earthquake evacuation process is vital to decreasing the number of deaths and damages sustained to buildings. Despite scientists attempting different ways of predicting earthquakes, Liang et al. (2018) note the impossibility of determining the precise timing and frequency of its occurrence. To accomplish this, the researchers designed a primary sample system in virtual reality. They believe that this sample can provide an immersive virtual environment for participants and evaluate the behavior of participants by simulating the structural and non-structural damages from earthquakes and exercise different emergency scenarios. It can contribute feedback and lay a foundation for improving the necessary behaviors for an earthquake (Liang et al., 2018) .

In another study involving behavioral analysis, Fang et al. (2020) introduced a virtual experiment that evaluated the decisions of a building's residents during and after an earthquake. 83 staff members and patients from Auckland Hospital participated in this study, in which they were required to position themselves on a shaking table to simulate the vibrations of an earthquake while in the virtual environment. To understand the decision-making process, the participants' cognitive abilities were assessed based on verbal protocol analysis. This involved the expression of thoughts orally in order to solve the problem. Through this means of analysis, basic reasoning and other mental processes are transparent for observation (Cornelissen, 2013; Ericsson, 2006; Ericsson and Simon, 1984; Feng, González, Trotter, et al., 2020; Salmon et al., 2013) .

To test the level of compliance when authoritative figures were present, two Auckland Hospital staff members assisted by giving instructions throughout the virtual reality simulation. Most of the participants relinquished their autonomy and acted according to the decisions of the staff members. However, 13 percent of the group elicited a different response by ignoring the staff members and trusting their own instincts. In general, these results show that people will follow instructions given to them, especially from someone in a position of power (Feng, González, Trotter, et al., 2020) .

Some of the most common psychological disorders in modern times are depression, anxiety, and stress. Depression is a psychological disorder that causes hopelessness and loss of interest in formerly enjoyable activities. Most people feel depressed and disconsolate. After all, it is the body's natural reaction to problems in life, resulting in a lack of enthusiasm for relevant people and interests. However, when this sensation of severe despondency, hopelessness, and disheartenment lasts more than a few days or weeks, the person is diagnosed with depression (Jones, 1995; Selye, 1991) . Anxiety acts in a similar manner, and a person can be diagnosed with it when the condition interferes with his or her daily life. It is an intense feeling of worry, fear, and uncertainty originating from an unknown source. Anxiety is natural from time to time, but chronic and severe anxiety is problematic and unusual. The average individual with anxiety can be constantly afraid, troubled, and stressed (Sonnentag and Frese, 2003) .

As for stress, its symptoms are related to anxiety and it feels like a sensation of pressure. It can be overwhelming when a person is not adjusted to handling certain levels of duress. When undergoing stress, a person's breathing rate and heartbeat increases, which gives him or her more energy. Stress is common and even beneficial in smaller quantities because of the motivation it gives. Some believe that stress helps them and that life loses its meaning without it. However, stress can have negative consequences when it impacts a person's daily life and overpowers their ability to function properly. This is called bad or ablative (decreasing) stress (Pearlin et al., 1981; Selye, 1991; Sonnentag and Frese, 2003) . These three psychological disorders (i.e, depression, anxiety, and stress) are interconnected, which is why psychologists will examine them together when formulating a diagnosis (Sonnentag and Frese, 2003) .

In order to treat these psychological disorders, they must be diagnosed first. There are different ways to detect psychological disorders. The most reliable way is to visit a psychiatrist or psychologist. Nevertheless, some people will feel that their problems cannot be severe and will search for a more convenient way of diagnosis. Using accredited psychological tests is an easier way of verifying if someone has a psychological disorder without consulting a medical professional. One of the best accredited tests, designed by Lavibound (Lovibond, 1983) , was named the Depression, Anxiety, and Stress scale (DASS). It has been proven trustworthy because of its accuracy during many trials (Ham et al., 2017; Lovibond, 1983; Lovibond and Lovibond, 1993; Pearlin et al., 1981; Selye, 1991; Sonnentag and Frese, 2003) .

The depression, anxiety, and stress scale (DASS) has two different forms. The main form of DASS has 42 questions that measure each psychological component every 14 questions, while the shortened form contains 21 questions that measure the psychological factors every 7 questions (Lovibond, 1983) . In a study by Tellegen et al. (Tellegen and Ben-Porath, 1992) , the team of researchers evaluated the DASS scale for the factors of depression, anxiety, and stress. The results of this study showed that 68% of the variance on all scales were evaluated by these three factors. The special values for stress, depression, and anxiety were 9.07, 2.89, and 1.23 respectively, and the alpha coefficient for these factors were 0.97, 0.92, and 0.95 respectively. When Tellegen et al. (Tellegen and Ben-Porath, 1992) calculated the correlation between all of the factors, it was revealed that there was a correlation coefficient of 0.48 between depression and stress, 0.53 between anxiety and stress, and 0.28 between anxiety and depression.

In 2005, Henry and Crawford (Henry and Crawford, 2005) studied the shortened form of the DASS scale for the validity of its structure and accounted for the factors of depression, anxiety, and stress.Henry and Crawford reported the final coefficients for all of the factors, which were equal to 0.88, 0.82, 0.9 and 0.93 respectively. Because of the relationships between depression, anxiety, and stress, and especially the relationship between anxiety and depression, in several studies in Iran it has been chosen to use the shortened form of DASS.

In 2007, Samani and Jokar (Samani and Joukar, 2007) also evaluated the validity and reliability of the DASS scale. They discovered that the reliability of depression, anxiety, and stress were equivalent to 0.8, 0.76, and 0.77 respectively, and that the Cronbach alpha for those same variables were equivalent to 0.81, 0.74, and 0.78 respectively. In the same year, Asghari et al. (Asghari et al., 2008) conducted another research study involving the same subject but on nonclinical samples. The goal was to determine the reliability and validity of the scale within the greater context of society. They performed the Beck Anxiety Inventory (BAI) and the Four-Dimensional Symptom Questionnaire (4DSQ) on a sample of 420 adults in order to test the effectiveness of the depression, anxiety, and stress scale. The evaluation of internal similarity and test-retest coefficients confirmed the validity of the DASS scale. The exploratory factor analysis showed that 14 phrases were attributed to stress while depression and anxiety had 2 phrases attributed to them each. The 2 phrases for depression and anxiety respectively impacted all three of the psychological aspects greatly. When these four phrases were removed, the analysis also confirmed the exploratory factor of the DASS structure and its exceptional operation. Additionally, the correlation calculation method of the sample scores in the Beck depression scale and the four orders anxiety questionnaire confirmed the reliability of the DASS structure. The concurrent credibility of the DASS scale was confirmed through comparing the scores of a subsidiary sample taken from the general population (315 people) with a counterpart group of patients who had psychological disorders (130 people). It can be concluded based on the results of this research study that the depression, anxiety, and stress scale meets the required conditions for psychological research or clinical settings in Iran (Asghari et al., 2008) .

In the depression, anxiety, and stress (DASS) evaluation test, all three factors are witnessed in a person. The individual percentages of depression, anxiety, and stress can be observed in the test results (Lovibond and Lovibond, 1993 ).

Assessing anxiety symptoms is important for diagnosing and treating anxiety. However, there are numerous scales proposing different perspectives, which leads to a lack of standardization (Costello and Comrey, 1967; Endler et al., 1991; Zung, 1971 ). These scales most likely contain problems regarding their conception and their psychological properties (Dobson, 1985; Mendels, 1972) . Becket et al. ) took these problems into consideration and introduced the Beck anxiety inventory (questionnaire) (BAI), which specifically measures the intensity of patients' anxiety symptoms.

To define the Beck anxiety questionnaire, it is a self-reporting questionnaire that measures the intensity level of anxiety in teenagers and adults. Conducted studies prove that this questionnaire is highly credible. Its internal consistency coefficient (Alpha coefficient) is equal to 0.92, its reliability after one week using the test-retest method is equal to 0.75, and the correlation of its clauses varies from 0.30 to 0.76. Five correlations between content, concurrent, structure, diagnosis, and factorial are measured in this test, which all show the efficiency of BAI in measuring anxiety intensity . The psychological components of this test have also been investigated. Gharaie et al. (Gharaei et al., 2000) reported its reliability coefficient through the test-retest method after two weeks. Kaviani and Mousavi (Kaviani H and Mousavi A, 2008) also assessed the psychological components of BAI in the Iranian population. They reported that the coefficients were almost 0.72, with the retest reliability coefficient being equal to 0.83 after one month and the Cronbach alpha being equal to 0.92.

Regarding its contents, the Beck anxiety questionnaire has 21 questions with four options each. Each question describes one of the common symptoms of anxiety (mental, physiological, and fear). Based on the participant's response, each answer will be ranked on a scale of 0 to 3. Therefore, the total score of this questionnaire ranges from 0 to 63.

In recent studies, researchers have evaluated human behavior and thought processes by simulating earthquakes in virtual public areas (Feng et al., , 2021 Feng, González, Trotter, et al., 2020; Lovreglio et al., 2018) . According to Feng et al (missing citation?) ., comprehending decision-making processes can explain people's reactions to different conditions in an earthquake. This information can assist in the development of new guidelines and proper evacuation methods after an earthquake. Furthermore, it can provide important data for enacting earthquake-related safety policies and improve safety training on this natural disaster. Although these efforts have advanced the quality of education on safety procedures, more research is required on how certain psychological factors are affected by earthquakes. This study aims to thoroughly research this subject and close the research gap by using self-reporting psychological tests. It also discusses the effect of previous alerts for the occurrence of the earthquake on stress and the performance of people's decision-making. Finally, it evaluates whether or not virtual reality has improved the quality of people's educational training.

A virtual reality model of a classroom in Tehran was constructed in Unity software at the beginning of this study along with the virtual educational scenarios in virtual environment. After the virtual education model was completed, the real-life educational model was constructed precisely based on virtual education due to the suggested statistical sample of two types of education, virtual and real. Finally, data from the DASS, BAI tests, and assessment tests for earthquake safety information were collected and evaluated.

For the purposes of this research, the Oculus Rift, a type of virtual reality glass, was used to display the virtual reality environments. An image of this equipment is given in the space below. 

A virtual reality model, designed after a hypothetical class in Tehran, was made by a threedimensional unity software. Unity is well known for its physics software and its animation options, which have been used to create realistic scenarios and life-like environments. In order to conduct a case study on students, different elements, parts, and assets have to be included within the modeled classroom environment on the Unity software. To define Unity assets, they are cases that can be used in games or projects. They may be created from a file that has been composed of Unity software like a three-dimensional model, an audio file, an image file, or any other type of file that Unity supports.

To create a simulated environment that is similar to reality, objects, elements, and textures from the physical classroom environment in Tehran must be incorporated visually and graphically.Textures are video files or movies that exist in models or serious game elements and provide visual effects. The properties and capabilities of Unity must be closer to reality so the model can be more similar to how the classroom is in real life.

Within the physical classroom environment, the floor is made from ceramic while half of the wall is composed of a combination of plaster and stone. Classrooms in Tehran are commonly comprised of these construction materials. Similarly, the model contains ceramic, plaster, and wall for a more realistic texture. The six benches used in the model are exactly alike to the wooden benches used in Iranian schools. The body and frame of these benches are metal, and the seats and table of the bench are made of wood. Just like Iranian benches, a metal box is located under the bench surface. Some books, notebooks, pens or pencils have been placed in front of the students. Each student's backpack has been modeled beside the bench or above the ground.

The students within the virtual model have supported the creation of a realistic classroom environment. All of them have been designed and modeled individually. As for their attributes, they have been given a blue uniform which is commonly seen within schools in Tehran. All aspects of their appearances are similar to those of Iranian students in real life. The student's pants, belts, and shoes have been designed based on current Iranian fashion trends. In total, there are six students in the virtual classroom including the user. The teacher is a dummy taller than the students and is equipped with gray hair, a professional suit, and a handbag. To create a natural classroom environment, these items have been hung on the wall: pictures of famous people in Iran, a map of the world, the periodic table, a whiteboard, a clock, and a wallpaper made by a student commemorating the anniversary of the Iranian Revolution. Other elements in the model include three heaters on the classroom's wall, two bookshelves, garbage bins, two windows with a view of the city of Tehran, a teacher's desk and table, a fluorescent light bulb, and other light bulbs. There is also a short platform in front of the class for teachers and presenters, with two standing bookshelves on top of the mentioned platform.

Two different scenarios have been designed and implemented in the virtual classroom model. In the first scenario, the user awaits the teacher's entrance into the classroom while he or she is seated on a bench. Other student's voices can be heard throughout the classroom at this time, and they are located at other benches where notebooks, books, and pencils have been designated for them. The students model realistic movements such as looking at the box under the table or looking around the classroom so the virtual model can simulate reality. The user can freely observe any area of the classroom, such as the teacher's desk, benches, walls, windows, bookshelves, and other students, by moving around his or her head.

Underneath the virtual classroom, an engine room has been attached. At the end of the first scenario, a student falls into the engine room due to incorrect positioning. In one of the figures, a visual from this scenario has been depicted. After the user looks at the door, the teacher enters the classroom with a bag in his hand. The classroom representative orders the students to stand out of courtesy and respect for the teacher. The user stands too and observes the classroom from a higher point of view. During this time, the teacher walks towards his desk, places the bag down, and tells the students to sit.

Next, the teacher gives instructions for the lesson: "Hello. Class, you may be seated now. If you remember the lesson from last week, we were discussing the reasons why an earthquake occurs. Today, we will talk about preparing for the occurrence of an earthquake while in school. All of you should not exit the door simultaneously. Instead, the best plan of action is to locate the safest areas in the classroom and take shelter there until the building stops shaking. An example of a safe spot is underneath the benches. Hide under the benches and hold the legs tightly so it stops vibrating. If you are in the library, workshop, or laboratory and cannot escape because of obstruction, keep a safe distance from the shelves and look for shelter. If you are in the schoolyard, keep away from the school building." These instructions are based on the guidelines of several organizations, namely the Center for Disease Control and Prevention (CDC) (48) in the United States, the New Zealand Civil Defense Organization (49), and the United States Geological Survey (USGS) (50).

While the teacher is lecturing, the other students listen eagerly. The user can move his or her head around to look at the students or the teacher. For a more realistic experience, the teacher incorporates body language into his mannerisms. For example, the teacher moves both of his hands forward when he instructs the students to sit down and points to the section below the desks when he discusses where to hide in the classroom.

When the teacher reaches the last sentence, an earthquake erupts. The user will witness and feel the vibrations within the virtual environment. During this scene, the students will start screaming and the teacher will inform the students that they need to take shelter immediately. After the initial shock subsides, some of the students will begin to defy orders and act differently. The simulation has been designed in this manner to illustrate the consequences of safe and unsafe actions. Because of this, the user will be able to distinguish which types of behaviors are appropriate during an earthquake; this will be elaborated on in the following sections.

As the scene continues, other visual techniques are employed to add to the effect of realism. Several books fall down at the front of the classroom and the fluorescent light bulb above the teacher's head becomes detached from its wire. The teacher moves behind his desk while ordering students to stay calm and take shelter. Unfortunately, one of the bookshelves falls on top of the teacher, who screams out of pain.

Dust begins to cloud the atmosphere and the students diversify in action. One of the students at the front of the classroom takes shelter under the door frame. Another student hides underneath a bench in the front row and puts his hands over his head. Immediately after the student does this, several chunks of the ceiling land on top of the bench. His quick and timely reactions show that the right choices can save a person's life.

As the building continues to collapse, two students at the back of the classroom take shelter under their benches. A student on the right side of the user, however, stands up from his bench and runs toward the front of the classroom. Plaster, soil, and bricks fall on top of the student and bury him, and the student stops moving. Various construction materials falling down from the roof cause the floor to crack due to the weight of their impact. Consequently, the teacher warns students that the roof has collapsed and that students should not approach the area.

The user then stands up and proceeds to the middle of the classroom where the damage in the floor is located. The structure of the floor has lost its strength and can no longer support any weight, so the user falls through and lands in the engine room in the lower story of the building. The scene ends after this incident.

Once again, the simulation highlights the consequences of not taking shelter properly. The user is allowed to experiment with his or her choices and thus becomes acquainted with the right and wrong decisions during an earthquake. Through visual and auditory learning, the user learns how to react properly whether they are in a classroom, library, or laboratory.

The second scenario begins right after the first one ends. In this scenario, the user can observe safe and unsafe locations within the previous setting. Safe locations, such as the space underneath the benches or the corners in the back of the class, are marked by a green cube. Unsafe locations, such as the area in front of the bookshelves, the door frame, or the center of the classroom, are labeled with a red cube. Photos from the second scenario are presented down below. It should be mentioned that the user's perspective in this scenario is in front of the classroom and besides the teacher's desk.

In order to evaluate how effective virtual reality is on learning, a questionnaire was designed to ask the subjects about the learning objectives. With the assessments, the extent of this impact will be known. The questionnaire contained 5 questions, which are shown in Table 1 . List the unsafe locations in the classroom during the earthquake.

4

What type of safety measure should we take during an earthquake at a library, workshop, or laboratory? 5

What type of safety measures should we take in the school's yard during an earthquake?

The research study consisted of two groups of 10 male students, all aged 20 to 25. The first group was educated on earthquake related safety procedures in a physical environment that adhered to COVID-19 protocols. To keep the physical and virtual models consistent, the topics corresponded to the first scenario in the virtual reality classroom; each sentence spoken in the physical classroom followed what the teacher said in the simulation by 10 seconds or less. Two weeks after the initial training date, the first group returned to complete their assessment tests.

The second group was similar to the first group in terms of size, background, and age. The students were asked to finish the DASS and BAI questionnaires in an undisturbed environment. Next, they played through the two virtual scenarios discussed in the previous sections on an Oculus Rift. It should be noted that none of the students had used an Oculus Rift before, and that for safety purposes, all COVID-19 safety protocols were implemented at this time. After completing both scenarios, they were asked to resubmit the DASS and BAI questionnaires. Just like the first group, the students in the second group completed their assessment tests two weeks after the initial date. Unlike the first group, however, the second group was divided into 2 subsections of 5 students. The first subsection was informed of the earthquake within the first scenario while the second subsection was informed of the first scenario without an earthquake. The results of the assessment tests between the two subsections will be analyzed in the next section.

As mentioned before, the goal of the study is to compare the two groups against each other and to determine which method of education, traditional or virtual, produces a better learning curve. The first group attended a physical classroom with COVID-19 safety measures in effect, and they were taught how to navigate the dangers of an earthquake in the safest way possible. For the sake of consistency, the first group's curriculum mirrored the one in the virtual reality model. The second group experienced the same training as the first group with the exception of some differences. The second group received their training in a virtual classroom, and half of those students were not informed of an earthquake within the first scenario. All of the students were tested in one day.

Because virtual reality is an immersive environment, the participants of the second group were given a revolving chair to enhance their experience and to allow their heads to move 360 degrees. 

Two weeks after the initial training sessions, both of the groups took a 10 minute assessment test based off of the questions listed in Table 1 . The next section explains the results of the first group, who were taught traditionally in a physical classroom. When comparing the answers of the first and second groups, the second group's answers were more accurate and corresponded with the first and second scenarios in virtual reality more significantly. Some inaccurate responses can be viewed in Table 2 as participants in the first group drew on outside knowledge. For example, the first question asks about the safest place in a classroom during an earthquake. All of the students in the second group answered that it was under a desk or a bench, but only two students in the first group gave the same answer. The remainder of the students responded with answers like "the door frame of the entrance" or "the schoolyard", both of which are incorrect. For the rest of the questions, the students in the second group gave accurate behavioral responses such as "staying calm and quiet" and named other safe locations in a classroom like "the corner of the classroom". In contrast, some of the students in the first group left their questions with no response. After this analysis, it can be concluded that the second group conducted the test in a more thorough and accurate manner.

To reiterate, the second group was exposed to the two different scenarios within the virtual reality model. 5 students knew of the earthquake in the first scenario, and the other 5 students were not informed. Hence, they assumed that it would be a normal educational setting in virtual reality.

As mentioned before, the DASS and BAI tests were taken before and after the experiment, and it was taken again two weeks after the initial date of training. Tables 4 and 5 list the results of the DASS tests before and after both scenarios. Figure 6 is a graph analyzing Tables 4 and 5 more precisely. Because anxiety is a long-term type of stress and lies beyond the scope of this study, only short-term stress from the DASS test was entered into Figure 6 . The category of depression was also not considered and those results were ignored as well. Every interval in Figure 6 has a left-hand side and a right-hand side column, which represent the levels of stress before and after the experiment respectively. The stress levels of the second group and its subsections are depicted in Figure 7 . Not surprisingly, the subsection that was informed of the earthquake in the first scenario experienced less of an increase in their stress levels than the subsection that was uninformed of the earthquake. For the sake of concision, the second subsection will refer to the five students who were uninformed.

In Figure 8 , the grades of natural, weak, medium, and severe correspond to the colors light blue, dark blue, purple, and red respectively. Before the experiment, two students were in the "natural" grade, two students were in the "weak" grade, and one student was in the "medium" grade for the second subsection. After the experiment, the second subsection had two students in the "medium" grade and three samples in the "severe" grade. While the grading had increased for all students, the increase was considerably greater among the second subsection as opposed to the first. In figure 8 , the left and right columns in every interval represent the stress levels before and after the experiment respectively, and the colors light blue, dark blue, purple, and red correspond to the grades of "least/nothing," "weak," "medium," and "severe." The results of the BAI are similar to the DASS test: both of these tests report an increase in all cases and a higher increase for the second subsection. The difference shows how the ability to foretell an earthquake can decrease the levels of stress in individuals during an earthquake. Before the experiment, the grades for the first subsection were "least/nothing" for three students and "weak" for two students. After the experiment, these grades changed to three in "medium," one in "weak," and one in "severe" for the first subsection. The second subsection scored three grades of "least/nothing" and two grades of "weak" before the experiment, and increased to three grades of "medium" and two grades of "severe" after the experiment. Simply stated, this means that the second subsection experienced a greater increase than the first subsection after the experiment. A value has been assigned to each sample. Figure 9 . The increased values of stress before and after the experiment in the BAI test Figure 9 describes the increased stress levels from the BAI test in the second group as a whole. The graph can also be interpreted as a greater increase in stress levels for the second subsection. The average amount of increase for the first and second groups are equivalent to 13 and 21.8 respectively.

According to recent studies in the field of cognitive science (especially pertaining to stress and decision-making), stress is found at the behavioral and neural levels. In other words, regions of the brain that regulate reasoning and decision-making are influenced by changes in stress (51). These researchers have reached a common consensus that individuals who are stressed or are in stressful situations will make more harmful and impulsive decisions. A severe increment in an individual's stress level will negatively impact his or her ability to rationalize and make proper decisions (51-53), which can often be seen in circumstances with high risk or reward (54, 55). As examined in this study, virtual reality technology provides a basis for evaluating the way premonition affects stress and decision-making. Furthermore, this study explores the most successful ways to relay safety measures against earthquakes and the remarkable capabilities of virtual reality technology. To summarize, it has been revealed that educating others through virtual reality is much more effective than the traditional method of education, and that individuals who are unaware of an earthquake occurring beforehand are proportionally more stressed than individuals who are aware. Alert systems that notify people of earthquakes in advance have been implemented in contemporary countries like Japan. This can minimize the amount of stress during an earthquake and govern the decisions people will make in preparing for one.

The results of this thesis can be summarized as follows:

• Using virtual reality technology and applying a virtual environment to educate safety measures against earthquakes can be a more effective way of teaching compared to the traditional way in this field. • By simulating the conditions of an earthquake in virtual reality, people can experience a rare natural disaster like a large magnitude earthquake. This will assist them in regulating their stress and behaviors in the event of an actual earthquake. • Installing alarms throughout a city and implementing a national alert system can increase the probability that people will act swiftly and safely in response to an earthquake. • Virtual reality imitates real life through its sensory feedback and immersive environment.

The second group that participated in the virtual reality scenarios experienced a general increase in stress, which proves that virtual reality is more than capable of creating realistic scenarios.

Because this study was conducted during the COVID-19 pandemic, only a certain number of participants could be recruited to follow guidelines pertaining to research. If the number of samples increases in a future study, it should lead to a more precise and comprehensive conclusion. Moreover, the lack of an earthquake shaking table and the confined spaces of the research laboratory contributed to the difficulties of creating a more realistic environment. Should a shaking table be present in a future study, it may assist in obtaining more extensive and accurate results.

Because not many studies are available on the topic of earthquake related safety measures, there are diverse paths researchers can take to broaden the scope of this field. Some suggested research topics are listed below:

• Evaluating how age, gender, and education level can affect the ability to learn and make decisions in regards to earthquakes and earthquake related safety measures • Educating how to escape fire breakouts after an earthquake by using virtual reality simulations • Investigating into decision-making and stress management processes during an earthquake by using brain and heart tests like EEG and HRV • Examining other technological advancements like augmented reality as a way to educate safety measures against earthquakes

In recent years, the importance of safety measures against earthquakes has increased significantly. Earthquake prone countries have been educating the public on the appropriate measures against earthquakes. It should be the duty of governments, businesses, and policymakers to assist in training current and future generations on how to safely navigate natural and manmade disasters like floods, earthquakes, fires, etc. Although Iran has implemented educational programs in offices and schools to raise public awareness, it has failed to impact Iranian society as a whole. These programs are widely ignored, which has regrettably led to financial damages and casualties when earthquakes have occurred. While earthquake engineers have written various guidelines on constructing structurally resistant buildings, they have not considered how educating safety measures to the public could potentially save lives during an earthquake. Posters, movies, and conferences on this matter are small steps towards increasing awareness, but not introducing people to the simulations of an earthquake will leave them unprepared mentally and emotionally when encountering a real one. Consequently, people will be more likely to make rash and harmful decisions due to unprecedented levels of stress.

Virtual reality receives attention from researchers in many fields, and it has already been used in other seismic research studies. Virtual reality technology enhances an individual's senses, which creates very realistic scenarios when simulating earthquakes. The knowledge gained from experiencing an earthquake in virtual reality can assist a person in making the right choices and managing their stress in the case of a real earthquake. In being aware of safety measures against earthquakes, individuals have the power to control their stress and to make the best decisions for themselves.

In conclusion, this study has evaluated the effect of education and foresight on rational decisions during an earthquake. In order to enact this experiment, virtual reality simulations were introduced to realistically model earthquakes and self-reporting assessment tests were conducted to assess the comprehension levels after learning about earthquake related safety measures. It can be reasonably inferred that virtual reality is a successful educational tool and that it should be distributed on a wider scale for earthquake training purposes. Finally, predicting large earthquakes beforehand and implementing digital and/or physical national alert systems can significantly help to guide people's decision-making processes during an earthquake.

The writers would like to acknowledge TECNOSA R&D Center and National Brain Mapping Laboratory for their sincere support. They are also thankful to Dr. Reza Rostami, Dr. Mojtaba Noghabaei, Ms. Neda Mohammadi, Ms. Emmy Ly and those who participated in this study and supported the research project.

Author Contributions: All authors have read and agreed to the published version of this manuscript.

Funding: This research received no external funding.

Institutional Review Board Statement: Ethical review and approval were waived for this study due to the absence of Personal Identifiable Information (PII) in the data collected by a member of the research team. Informed Consent Statement: Informed consent was obtained from all subjects involved in the study.

The data presented in this study are available on request from the corresponding author. The data are not publicly available due to privacy issues.

The authors declare no conflict of interest.

What can we do about earthquakes?

A preliminary validation of the d epression, anxiety and stress scales (DASS) in non-clinical sample

Virtual reality training system for maintenance and operation of high-voltage o verhead power lines

An inventory for measuring clinical anx iety: Psychometric properties

Beck anxiety inventory (BAI)

Station Reallocation and Rebalancin g Strategy for Bike-Sharing Systems: A Case Study of Washington DC

Towards a 'behavioural design' approach for seismic risk reduction strategies of buildings and their environment

Serious games for training occupants of a building in person al fire safety skills

How can they do it? A structured approach to modelling performance variability

Scales For Measuring Depression and Anxiety

The relationship between anxiety and depression

Endler Multidimensional Anxiety Scales (EMAS)

Protocol analysis and expert thought: Concurrent verbalizations of thinkin g during experts' performance on representative tasks

Protocol Anal ysis: Verbal Reports as Data

Immersive virt ual reality serious games for evacuation training and research: A systematic literature revie w

An immersive virtual reality serious game to enhance earthquake behavioral responses and post-earthquake evacuation preparedness in buildings

Instructional mec hanisms in immersive virtual reality serious games: Earthquake emergency training for chil dren

How people make decisions during earthquakes and post-earthquake evacuation: Using Verbal Protocol Analysis in Immersive Virtual Reality

Attribution style in patients with anxiety and d epression comorbidity

A Novel Earthquake Educatio n System Based on Virtual Reality

Discrimination of multiple stress levels in virtual reali ty environments using heart rate variability

The short-form version of the Depression Anxiety Stress Scales (DASS-21): Construct validity and normative data in a large non-clinical sample

New dimen sions in surgical training: immersive virtual reality laparoscopic simulation exhilarates surgi cal staff

Foreshocks, aftershocks, and earthquake probabilities: accounting for the La nders earthquake

Psychometric properties of the Persian version of Beck Anxie ty Inventory in an Iranian population age and sex classes

Utilization of Virtual Reali ty Visualizations on Heavy Mobile Crane Planning for Modular Construction

Earthquake training is gaining importa nce: the views of 4th and 5th year students on Earthquake

Virtual Reality/University of Illinois

Earthquake Safety Training through Vi rtual Drills

Development of a VR prototype for enhancing earthquake evacuee safety

The nature and measurement of anxiety, stress and depression

Manual for the depression anxiety stress scales (DAS S)

Prototyping virtual reality serious games for building earthquake preparedness: The Au ckland City Hospital case study

Appropriate and correct reactions du ring earthquakes: 'drop, cover and hold on' or 'triangle of life

The Relationship Between Depression and Anxiety

Lost Vibration Test Data Recovery Using Con volutional Neural Network: A Case Study

Automated Workers' Ergonomic Risk Assessment in Manual Material Handling Using sEMG Wearable Sensors and Machine Learning

New Zealand Ministry of Civil Defence & Emergency Management, Working from the same pa ge consistent messages for CDEM

A LITERATURE REVIEW ON IMMERSIVE VIRTUAL REALITY IN EDU CATION: STATE OF THE ART AND PERSPECTIVES

The Stress Process

Compatible cognition amongst road user s: The compatibility of driver, motorcyclist, and cyclist situation awareness

A study on the reliability and validity of the short form of the depression anxiety stress scale (DASS-21)

Community-base d pedestrian safety training in virtual reality: A pragmatic trial

History and Present Status of the Stress Concept

Hybrid Simulation-based Resou rce Planning and Constructability Analysis of RCC Pavement Projects

Earthquake Drills and Simulations in Community-based Training and Pre paredness Programmes

Stress in Organizations

The new uniform T scores for the MMPI--2: Rationale, derivation, and appraisal

Current practices in serious ga me research: A review from a learning outcomes perspective

A Rating Instrument For Anxiety Disorders