Advancing Suicide Prevention Through Immersive Technology

Shenghao Chen

1 a

, Yi (Joy) Li

2 b

, Chao Mei

3 c

and Thomas Joiner

Department of Psychology, Florida State University, Tallahassee, U.S.A.

Department of Software Engineering and Game Development, Kennesaw State University, Marietta, U.S.A.

TCL Rayneo XR, Shenzhen 518000, China

Keywords:

Human-Computer Interaction (HCI), Virtual Reality (VR), Suicide Prevention, Simulation, Behavior

Management.

Abstract:

This study presents a novel translational method for suicide prevention research, employing customizable

virtual reality (VR) simulations that mimic real-life situations. We aim to validate the usability of these VR

scenarios as approximations of actual suicide and to rigorously assess their safety. By increasing the variety

of simulated suicide methods and variables, this study enables researchers to effectively isolate the causes of

suicide and thus provide key insights for suicide prevention.

1 INTRODUCTION

Each year, there are more than 700,000 people around

the world who die from suicide (WHO, 2021), leav-

ing devastating impacts on their families, friends, and

communities long after. Suicide remains a major pub-

lic health concern across different regions, with vary-

ing trends and risk factors inﬂuenced by cultural, eco-

nomic, and social contexts. Despite major institu-

tional and scientiﬁc efforts to reduce suicide rates,

such as those outlined by Fox et al.(2020), the preva-

lence of suicidal thoughts and behaviors remains con-

cerningly high in certain populations, including the

United States (CDC, 2024a; Curtin et al., 2021).

An experimental method to investigate the direct

cause of suicide is required to develop effective treat-

ments targeting the speciﬁc cause. We know little

about direct suicide cause since we cannot use sui-

cide as an experimental outcome variable in real life,

where manipulation may make its occurrence more

likely. Suicide scenarios in VR allow us to investi-

gate the causes of suicide by conducting experiments

that manipulate purported causes of suicide and detect

their effects on VR suicide rates. Previous research

has validated shooting and jumping scenarios from

existing VR games as safe approximations of real-

life suicide (Franklin et al., 2019; Huang et al., 2021).

https://orcid.org/0000-0002-5835-1634

https://orcid.org/0000-0002-4209-3722

https://orcid.org/0000-0003-3127-4322

However, these scenarios were limited in the variety

of suicide methods, lack of consistency, and restricted

interactions. This limitation arises because these VR

suicide scenarios were derived from separate, unre-

lated commercial games, each designed with different

mechanics, environments, and objectives that were

not originally intended for suicide research. As a re-

sult, their implementation varies across studies, mak-

ing it difﬁcult to systematically compare ﬁndings or

establish a standardized research framework. In the

current study, we designed a VR Suicide Scenarios

project with a novel VR platform with various build-

able suicide methods, allowing researchers to over-

come previous limitations, test new hypotheses about

speciﬁc causal factors of suicide, and design and de-

liver targeting interventions. This study focused on

the following objectives: 1) Measure the validity of

the VR suicide scenarios by comparing the new sce-

narios with actual suicide and previously validated

VR scenarios; 2) Evaluate the safety of the proposed

VR suicide scenarios.

2 LITERATURE REVIEW

Suicide research has historically struggled to iden-

tify strong predictive risk factors, with most identi-

ﬁed factors offering weak predictions for suicidal be-

haviors (Franklin et al., 2019). Many identiﬁed risk

factors for suicide, such as prior psychiatric hospi-

talization, are correlated with suicide but do not di-

882

Chen, S., Li, Y., Mei, C. and Joiner, T.

Advancing Suicide Prevention Through Immersive Technology.

DOI: 10.5220/0013500300003932

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

In Proceedings of the 17th International Conference on Computer Supported Education (CSEDU 2025) - Volume 1, pages 882-889

ISBN: 978-989-758-746-7; ISSN: 2184-5026

rectly cause it. As a result, interventions targeting

these factors have shown limited efﬁcacy (Fox et al.,

2020). This highlights the need for a shift toward

identifying causal factors with large effects, using ex-

perimental methodologies involving random assign-

ment and controlled manipulation. One such factor is

perceived entrapment — the feeling of being trapped

in an unbearable situation with no escape (O’Connor

and Kirtley, 2018). Unlike correlational risk factors,

causal factors like this can directly inﬂuences suicidal

behavior and can be experimentally tested. Empiri-

cal work supports its causal role: perceived entrap-

ment predicts the ﬁrst onset of suicidal thoughts in

medical students (Wang et al., 2023) and exacerbates

suicide risk in psychiatric populations (Taylor et al.,

2010). Interventions targeting this mechanism, such

as cognitive strategies to reduce entrapment percep-

tions, have demonstrated success in lowering suicidal

ideation (O’Connor and Kirtley, 2018).

VR offers a safe, ethical, and controlled method to

manipulate such factors, enabling researchers to ex-

plore suicide mechanisms beyond the limitations of

observational studies. Franklin et al. (2019) devel-

oped a VR suicide approximation featuring scenar-

ios like jumping from Richie’s Plank Experience and

shooting in Arizona Sunshine. Their ﬁndings indi-

cate that these VR suicide scenarios are realistic and

that factors such as gender, suicidal desire, and prior

suicidality, which predict real-world suicide, also in-

ﬂuence VR suicide rates. This effectively validates

VR as a useful tool for suicide research. Regarding

safety, Huang et al. (2021) conducted two longitudi-

nal studies (one-month and two-year follow-ups) to

assess the psychological safety of VR suicide scenar-

ios. They measured suicidality, capability for sui-

cide, and agitation post-exposure, ﬁnding no signif-

icant increases in distress or suicidal ideation. In

some cases, suicide-related constructs showed small

to moderate reductions, supporting the safety of VR-

based suicide research. These ﬁndings align with

utilitarian ethics, which prioritize maximizing bene-

ﬁts while minimizing harm. The study suggests that

the potential research and intervention beneﬁts of VR

suicide scenarios outweigh minimal risks when ethi-

cal safeguards—such as participant screening, careful

scenario design, and post-experiment debrieﬁng—are

implemented.

While VR suicide scenarios do not replicate real-

life suicide, they provide valuable insights into the

causes of suicide and are crucial in directing future

strategies in suicide prevention. Research studies in-

volving commercial games as VR suicide scenarios

have shown that the anticipation of avoiding future

psychological and physical pain signiﬁcantly inﬂu-

ences VR suicide decision rates (Huang et al., 2020;

Park et al., 2023). These results emphasize the impor-

tance of anticipated consequences over experienced

antecedents like stress, rejection, and physical pain in

determining suicidal behavior. Furthermore, Ribeiro

et al. (Ribeiro et al., 2021) showed that raising doubts

about the desirable outcomes of suicide can effec-

tively reduce VR suicide rates and alter perceptions

regarding the lethality of suicide methods. These cru-

cial ﬁndings further conﬁrm the indispensable role of

VR suicide scenarios; without them, exploring and

experimentally manipulating suicide risk factors in

real-life settings could be dangerous and unethical.

3 OUR CONTRIBUTION

Our VR experience is designed for healthcare pro-

fessionals to study suicide-related decision-making in

the general population. While participants in the VR

scenarios come from the non-clinical population, the

goal is to analyze their behaviors in a controlled en-

vironment, providing insights for suicide prevention

and clinical interventions

Although previous studies using VR suicide sce-

narios have provided valuable insights, they also

present signiﬁcant limitations, which our project aims

to address. First, the two existing scenarios in Richie’s

Plank Experience and Arizona Sunshine, though real-

istic, offer limited variety and fail to adequately rep-

resent the most common methods of suicide in the

United States. For instance, jumping is not a com-

mon method of suicide in the United States. In con-

trast, commonly used methods such as suffocation

and poisoning, as noted by the Centers for Disease

Control and Prevention (CDC, 2024b), were not in-

cluded. Second, both jumping and shooting are highly

lethal methods, offering minimal chances for individ-

uals to reconsider their actions. In reality, many in-

dividuals attempting suicide might opt for less lethal

methods, and many may not complete the act. These

discrepancies highlight the necessity for a wider range

of VR suicide scenarios that more accurately mirror

the methods commonly used in suicide attempts. Our

study responds to this gap by introducing ﬁve VR sui-

cide scenarios— gun shooting, jumping, overdosing,

hanging, and cutting — aligning with the most preva-

lent methods of both suicide death and nonfatal self-

harm according to the U.S. Centers for Disease Con-

trol and Prevention (CDC, 2022).

Furthermore, the existing VR scenarios were pri-

marily designed for entertainment purposes and might

not align with the nuanced requirements of psycho-

logical and behavioral research. These scenarios may

Advancing Suicide Prevention Through Immersive Technology

883

lack the complexity and speciﬁcity needed for a thor-

ough exploration of intricate suicide-related phenom-

ena. Our project goes beyond these enhancements

by developing tailored VR suicide scenarios, created

speciﬁcally for research purposes. These scenarios

will include a private set of environments and situa-

tions, representative of the various contexts in which

suicidal thoughts and behaviors occur. In the develop-

ment of these scenarios, we are collaborating closely

with mental health professionals. This partnership en-

sures that the scenarios accurately represent the psy-

chological and behavioral aspects of suicide, while

also providing researchers with essential, controllable

variables for their studies. This approach is intended

to enhance the generalizability of research ﬁndings

using these scenarios across diverse populations.

4 METHODS

From the literature review and our previous investi-

gation results using existing gaming scenes, we fol-

lowed several key points in designing the current

project. First of all, the VR application prioritized

realism and immersion to closely approximate real-

life situations involving suicide. Unlike some other

VR games that use simple polygons to achieve en-

gagement purposes, the suicide scenarios have to be

as realistic as possible so that the proposed system

can trigger the embodied feelings of the users. Fur-

thermore, we implemented controlled behavior sce-

narios tailored to each potential method of suicide,

such as the available options between shooting one-

self and shooting other objects. Additionally, the sys-

tem offers the ﬂexibility to show or hide tools com-

monly associated with suicide, for example, by al-

lowing a gun to be hidden in a drawer or disabling

it through user choices. This level of detail not only

enhances the realism of the simulation but also allows

for a more comprehensive understanding and analysis

of these critical situations.

4.1 VR Scenario Design

This application is developed as a fully immersive

VR experience, utilizing narrative and interactive el-

ements, unlike conventional VR 360 videos or simu-

lated intervention sessions. The Unity game engine

was used for the development of the VR scenarios.

The primary platform targeted for this application is

Oculus Quest 2 or 3. It features a scenario-based nar-

rative, presented from a ﬁrst-person perspective, and

integrates comprehensive user interactions including

head and hand movements, as well as body tracking,

to ensure a realistic simulation. Users can navigate

the virtual environment – an apartment setting – using

VR control inputs, allowing them to teleport, walk, in-

teract with various objects, and perform speciﬁc ges-

tures. Additionally, they can engage with objects in

multiple ways, such as picking them up or jumping.

The application includes a voiceover feature that sug-

gests acts of self-harm; however, users retain auton-

omy over their actions and can choose to ignore these

suggestions. There are two distinct modes within

the game: a ‘free exploration’ mode intended for

experimental sessions in a laboratory setting, where

a research assistant provides live directions without

in-game voice guidance, and ‘guided scenes’ mode,

where the in-game voice offers navigational cues and

instructions. All scenarios in the application were in-

tended to provide valuable insights into human be-

havior and decision-making under extreme virtual cir-

cumstances, while maintaining a controlled and safe

environment for the player. Researchers outside of the

virtual environment initiate the sequence and choose

different scenes according to the experimental plan,

playing a crucial role in guiding and monitoring the

experience.

4.1.1 Level Design

The main hub is an apartment in a high-rise build-

ing. It has a living room, an open kitchen, a bed-

room, a bathroom, and a balcony linked to the liv-

ing room, all fully furnished and ﬁlled with furniture.

The balcony offers a view of other apartment build-

ings and the street below. The user is the main char-

acter in the game, who resides in the apartment. The

voice is a non-player character who provides guid-

ance and instructions to the player throughout the

game. A ﬂowchart of the scenarios is shown in Fig-

ure 1, screenshots of essential scenarios are shown in

Figure 2 and the following paragraphs describe them

in detail.

Tutorial (Main Hub) - Figure 2a. The player is

introduced to the game mechanics in an immersive

and instructive manner. Upon spawn, the player

ﬁnds themselves in the living room (which serves

as the main hub) of their virtual apartment, where

they are immediately greeted by a voiceover. This

voiceover explains the basic controls and mechanics

of the game, effectively orienting the player to the VR

environment. It provides detailed instructions on how

to move the player character, interact with various ob-

jects within the environment, and access the different

features available in the game. Such hands-on expe-

rience in the tutorial phase is designed to enhance

the player’s comfort and dexterity with the game’s

ERSeGEL 2025 - Workshop on Extended Reality and Serious Games for Education and Learning

884

Figure 1: Flowchart of the scenarios.

(a) Main hub.

(b) Shooting scene.

(d) Overdosing scene.

(e) Hanging scene.

(f) Cutting scene.

Figure 2: Screenshots of the scenarios.

interactive elements, setting a solid foundation for

their subsequent VR experience. All scenarios have

a timer; if the user chooses not to do self-harm, they

will be sent back to the main hub scene after a certain

amount of time.

Shooting - Figure 2b. The player receives instruc-

tion, which instructs them to go to the bedroom and

retrieve a handgun. This is followed by instructions

to shoot at a neutral object in the environment, serv-

ing as an introduction to the use of the ﬁrearm within

the virtual space. Subsequently, the voice presents

the player with a critical decision: to either “point

the gun between your eyes and pull the trigger, ef-

fectively shooting yourself in the face,” or to opt for

shooting another neutral object. This decision point is

central to the scenario, testing the player’s response to

extreme virtual situations. Should the player choose

the suicide option, the game is programmed to dis-

play a death scene before automatically resetting the

player back to the living room. This reset serves both

as a narrative conclusion to the chosen path and as

a mechanism to allow the player to re-engage with

the scenario or explore other aspects of the VR expe-

rience. This scenario is carefully crafted to provide

insights into player decision-making and responses

within high-stress environments.

Jumping - Figure 2c. The scenario begins with in-

structions guiding the player to move out onto the

balcony of the virtual apartment. The player is con-

fronted with a signiﬁcant choice: to either “step for-

ward and fall to their death” or to retreat safely back

into the apartment. In the event that the player opts

for the suicide option, the VR application is designed

to depict a death scene, providing a stark visual con-

clusion to the choice made. Following this, the player

is automatically reset to the living room, allowing for

a reevaluation of their decision or the opportunity to

explore other scenarios within the VR environment.

Overdosing - Figure 2d. The scenario begins with

an in-game voice guiding the player to locate the

medicine cabinet in the kitchen of the virtual environ-

ment. Once found, the player is directed to pour two

pills into their hand. The player is confronted with a

signiﬁcant choice: to either “take the whole bottle of

medicine,” implying an act of self-harm, or to adhere

to the prescribed dosage of just two pills. If the player

opts for the overdose, the VR system allows for an

important safety mechanism: the player can abort the

scenario at any time by calling an ambulance. This

feature adds a layer of realism and choice, reﬂecting

real-world emergency responses. If the player does

not call an ambulance after choosing the overdose

option, the game is programmed to display a death

scene, followed by a reset of the player back to the

main hub. If the player calls an ambulance, the screen

transitions to white, and the player “wakes up” on the

couch in the apartment, simulating a rescue and re-

covery.

Hanging - Figure 2e. The scenario begins with a

voice directing the player to locate a noose hidden

within a cabinet. The player is then instructed to

place a stool under a ceiling ﬁxture and attach the

Advancing Suicide Prevention Through Immersive Technology

885

noose to it. The player is confronted with a signiﬁ-

cant choice: placing the noose around their neck and

stepping off the stool, implying a simulated act of sui-

cide, or deciding against it and simply walking away

from the stool. If the player chooses to simulate sui-

cide, the VR application is programmed to display a

death scene. This is followed by a reset of the player

back to the living room.

Cutting - Figure 2f. The scenario begins with an

in-game voice instructing the player to locate a sharp

knife in the kitchen of the virtual environment. This

introduction to the tool is immediately followed by a

directive to cut an orange on a chopping board, es-

tablishing a mundane and safe use for the knife. The

player then faces a signiﬁcant choice: either simulate

self-harm by cutting their wrist or engage in a harm-

less activity by cutting a loaf of bread on the counter.

Moreover, the player can choose to abort this sce-

nario at any time by calling an ambulance. Should the

player opt for the simulated self-harm option and not

call an ambulance, the VR application is designed to

depict a death scene, followed by an automatic reset

of the player back to the living room.

Death Scene. All scenarios either end with exceed-

ing the time limit or a death scene. In the death scene,

when a user triggers a death scene, the immersive ex-

perience takes a poignant turn. As the participant’s

virtual perspective slowly darkens, the environment

fades away, symbolizing the retreat of life itself. Con-

currently, a heartbeat sound underscores this transi-

tion—starting with a strong, rhythmic pulse that grad-

ually shifts to a heavy, dissonant beat before ﬁnally

dwindling into a weak echo. This orchestrated blend

of visual and auditory cues is designed to deeply en-

gage users, allowing them to viscerally experience the

ﬁnal moments in a controlled, virtual setting.

4.2 Experimental Design

For this study, we will recruit undergraduate college

students from a non-clinical sample, primarily from

psychology courses at Florida State University. These

students will participate in the study in exchange for

course credit. Students will receive course credits

for their participation. We will implement a within-

subjects approach, meaning each participant will be

exposed to all ﬁve VR suicide scenarios. The order of

these scenarios will be counterbalanced using a bal-

anced Latin square method to ensure no order effects

bias the results.

For each scenario, participants will be faced with a

decision: to follow through with the simulated suicide

option or to choose a safer alternative action. This

study has no manipulation to encourage participants

to choose one option over the other, assessing the sui-

cide completion rate at a neutral condition.

Our primary objective is to assess the readiness

of the newly developed VR suicide scenarios by ex-

amining their ﬁdelity in replicating real-life suicide

situations and comparing them to previously vali-

dated VR suicide scenarios using commercial games.

This study is designed to validate the system’s effec-

tiveness as a research tool for suicide-related experi-

ments, not to investigate the underlying causes of sui-

cide. Once the system is validated, it will serve as a

foundation for future research to explore how speciﬁc

variables might inﬂuence participants’ likelihood of

choosing the suicide option.

4.2.1 Hypotheses

First, we hypothesize that participants will perceive

the scenarios as realistic and directly relevant to ac-

tual suicide contexts. To test this, participants will be

asked to provide realism and suicide-relevance ratings

for each scenario after completion.

Second, we anticipate that the rates of virtual sui-

cide completion will be low, reﬂecting the low preva-

lence of suicidal behaviors observed in the general

population. We expect the completion rates to be sim-

ilar to those observed in neutral conditions of pre-

viously validated VR suicide scenarios (around 5%)

(Franklin et al., 2019). This will be tested by calcu-

lating the percentages of participants who opt for each

behavioral outcome within the scenarios.

Third, we expect that factors commonly associ-

ated with real-life suicide will predict participants’

decisions to complete the VR suicide scenarios.

Speciﬁcally, we will examine male sex, suicidal de-

sire, agitation, prior suicidality, and other psycholog-

ical risk factors. These variables will be assessed us-

ing pre-experiment self-report questionnaires and val-

idated psychometric measures.

Fourth, we predict that the justiﬁcations provided

by participants for choosing the safe alternative or for

aborting the virtual suicide attempt will parallel the

reasons often cited in actual suicide cases. To ex-

amine this, we will collect the reasons that partici-

pants follow through with the simulated suicide op-

tion or choose the safe alternative. This aspect of the

study aims to deepen our understanding of the cogni-

tive and emotional processes underlying such critical

decisions.

Finally, we hypothesize the responses elicited by

the newly developed scenarios to closely align with

those from previously established ones. We will draw

comparisons of all the above metrics among the sce-

ERSeGEL 2025 - Workshop on Extended Reality and Serious Games for Education and Learning

886

narios. This will help us determine the similarity in

how participants react and make decisions across dif-

ferent VR suicide scenarios.

Finally, we hypothesize that the responses elicited

by the newly developed VR suicide scenarios will

align with those from previously established scenar-

ios. To evaluate this, we will compare key behavioral

(e.g., VR suicide completion rates) and psychological

(e.g., distress levels, realism ratings) metrics within

our study conditions and against data from previous

VR suicide studies where available (Franklin et al.,

2019; Huang et al., 2021).

4.3 Measures and Assessments

Participants will complete a series of validated psy-

chometric measures at baseline, after each VR sce-

nario, and at the end of the study. Baseline assess-

ments include demographics, suicidal ideation (Beck

and Steer, 1991), depressive symptoms (Beck et al.,

1996), agitation (Ribeiro et al., 2011), risk-taking ten-

dencies (Blais and Weber, 2006), fearlessness about

death (Ribeiro et al., 2014), moral reasoning (Fleis-

chmann et al., 2019), self-injurious behaviors (Nock

et al., 2007), interpersonal needs (Bryan, 2010), sub-

stance use (WHO ASSIST Working Group, 2002),

and current suicidal desire. After each VR sce-

nario, participants will rate realism and suicide rel-

evance and provide responses on the reasons for en-

gaging, not engaging, or aborting VR Suicide Scales

(Franklin et al., 2019). At the end of the study, sui-

cidal desire and BSS will be re-administered. Partici-

pants meeting suicide risk criteria (e.g., BSS ≥ 5, his-

tory of multiple attempts, or increased suicidal desire)

will undergo a clinical risk assessment (Chu et al.,

2017) by a trained doctoral student, following estab-

lished VR suicide research protocols (Franklin et al.,

2019; Park et al., 2023). All study team members will

receive IRB-approved risk management training be-

fore conducting assessments.

4.4 Data Analysis

To evaluate differences across the ﬁve VR suicide sce-

narios, we will use repeated-measures ANOVA to test

for signiﬁcant variations in behavioral (e.g., VR sui-

cide completion rates) and psychological (e.g., dis-

tress, realism ratings) metrics. If signiﬁcant effects

are detected, post hoc Bonferroni corrections will be

applied to control for multiple comparisons.

For comparisons with previous studies (Franklin

et al., 2019; Huang et al., 2021), we will use indepen-

dent t-tests for direct mean comparisons and mixed-

effects models to account for individual differences

across studies. Effect sizes, such as Cohen’s d for

between-group comparisons and partial eta squared

for within-subject effects, will be reported to contex-

tualize ﬁndings.

All statistical analyses will be conducted in R,

with an alpha threshold of 0.05 for signiﬁcance test-

ing.

4.5 Safety Evaluation

This study was reviewed and approved by the

Florida State University Institutional Review Board

(STUDY00004909).

To ensure the safety of participants exposed to the

new VR suicide scenarios, our safety evaluation plan

involves a combination of immediate, short-term, and

long-term assessments.

Immediately after the study, we will adhere to

the safety protocols established in previous VR sui-

cide scenario studies (Franklin et al., 2019; Huang

et al., 2020; Ribeiro et al., 2021). All participants

will undergo a positive mood induction by watching a

VR animation The Rose and I. Then, all participants

will receive a suicide risk assessment conducted by a

trained clinical psychology graduate student under the

supervision of the study’s faculty advisor, a licensed

clinical psychologist. The assessment will utilize an

empirically informed risk assessment methodology,

based on guidelines by Chu et al. (2015) and Joiner

et al. (1999). Corresponding steps will be taken to

mitigate risk based on their risk levels. These mea-

sures will be applied irrespective of the participant’s

decision to engage in VR suicide scenarios or their

assigned study conditions. Following this, our plan

includes a short-term follow-up survey within a week

of participation that repeats the same questions as at

baseline about the presence or absence of suicidal

thoughts and behaviors. This will help us detect any

delayed psychological effects, focusing on changes

in mood, mental health, and any occurrence of suici-

dal thoughts or behaviors since their experience with

the VR scenarios. We will also repeat the survey at

long-term follow-ups at one month and three months

post-participation to monitor any sustained or late-

emerging effects on participants’ mental health. A

trained clinical psychology graduate student will con-

duct a suicide risk assessment for participants with

elevated suicide risk and take corresponding steps to

mitigate their risk levels.

In addition to these follow-ups, we will compare

the data obtained from the new scenarios with that

from the established VR suicide scenarios. This com-

parative analysis will help us determine if the new

scenarios pose additional risks or have different psy-

Advancing Suicide Prevention Through Immersive Technology

887

chological impacts on participants.

Throughout the study, we will provide partici-

pants with access to mental health support and re-

sources to address any distress or heightened suici-

dal ideation following the VR experience. The safety

evaluation plan, including follow-up protocols and

support mechanisms, will undergo ethical review and

approval by our institutional review board.

By implementing this comprehensive plan, we

aim to ensure the safety of the new VR suicide sce-

narios, maintaining them at a standard comparable to

the established ones. This approach is integral to re-

sponsibly exploring participant responses to intense

suicide stimuli in a virtual environment.

5 DISCUSSION

In order to get feedback for the proposed VR appli-

cations, we conducted a preliminary playtesting, in

which seven college students from different majors

without any mental conditions participated. They pro-

vided valuable insights into the VR experience. We

summarized a few key points of the feedback.

First of all, users generally found the VR applica-

tion to be realistic and immersive. They felt as if they

were actually present within the scene. This suggests

the VR environment and scenarios are effectively de-

signed to create a convincing and engaging experi-

ence.

Secondly, the project’s portrayal of suicide scenar-

ios, such as the shooting and jumping options, was

found to be emotionally impactful. Users reported

feeling genuinely scared in these scenarios, suggest-

ing a strong emotional engagement with the content

as we expected.

All testers have reported that the death scene

leaves a profound and lasting impression, prompt-

ing them to reexamine their own perceptions of life

despite the actions occurring in a virtual environ-

ment. The carefully designed interplay of visual dark-

ness and the diminishing heartbeat sound creates an

emotionally charged experience that resonates deeply.

This immersive encounter not only underscores the

gravity of loss but also challenges participants to con-

front the reality of mortality, blurring the lines be-

tween virtual simulation and real-life introspection.

As an important note, users suggested the inclu-

sion of in-game surveys to avoid breaking immersion

by removing the headset, indicating a preference for

a seamless experience. This aligns with our original

plan that has not yet been implemented into the pro-

totype. Furthermore, improvements such as character

customization and enhanced interactivity within the

environment were also suggested. We will integrate

the feedback into the upcoming development and en-

hance the user experiences for better study results.

6 CONCLUSION AND FUTURE

WORK

In conclusion, our proposed VR experience will sup-

port the development of a translational approach to

studying suicide, validate its reasonableness with ac-

tual suicide, and examine its safety. The developed

VR suicide scenarios will increase the variety of sui-

cide methods and manipulating variables researchers

can use to design studies to isolate the causes of sui-

cide. Only after the causes of suicide, rather than cor-

relates, are identiﬁed, can researchers develop effec-

tive treatments targeting the causes.

For instance, prior research using VR scenarios

highlighted that the anticipation of avoiding future

psychological and physical pain inﬂuenced partici-

pants’ choice of the suicide option. With our new

VR scenarios, future research could examine how

different suicide methods inﬂuence completion rates.

Speciﬁcally, we hypothesize that methods perceived

as more distressing (e.g., shooting, hanging) may

have lower completion rates due to increased aver-

sion, while methods that allow for aborted attempts

(e.g., overdose) may result in more hesitancy and

lower ﬁnal completion rates compared to irreversible

methods (e.g., jumping, shooting). Studying these

differences in a controlled VR environment could pro-

vide valuable insights into the decision-making pro-

cesses behind suicidal behavior, ultimately informing

prevention strategies.

We will keep improving the prototype and conduct

the proposed study and evaluation. Future develop-

ment of the platform can even incorporate treatment

strategies and test their effectiveness in laboratories.

REFERENCES

Beck, A. T. and Steer, R. A. (1991). Manual for the beck

scale for suicide ideation. San Antonio, TX: Psycho-

logical Corporation, 63.

Beck, A. T., Steer, R. A., and Brown, G. (1996). Beck de-

pression inventory–ii. Psychological assessment.

Blais, A.-R. and Weber, E. U. (2006). A domain-speciﬁc

risk-taking (dospert) scale for adultpopulations. Judg-

ment and Decision making, 1(1):33–47.

Bryan, C. J. (2010). The clinical utility of a brief measure

of perceived burdensomeness and thwarted belonging-

ness for the detection of suicidal military personnel.

Journal of Clinical Psychology, 67(10):981–992.

ERSeGEL 2025 - Workshop on Extended Reality and Serious Games for Education and Learning

888

CDC (2022). WISQARS Leading Causes of Nonfatal In-

jury.

CDC (2024a). Fatal Injury Reports, National, Regional and

State, 1981 - 2020.

CDC (2024b). Suicide Data and Statistics | Suicide Preven-

tion | CDC.

Chu, C., Buchman-Schmitt, J. M., Stanley, I. H., Hom,

M. A., Tucker, R. P., Hagan, C. R., Rogers, M. L.,

Podlogar, M. C., Chiurliza, B., Ringer, F. B.,

Michaels, M. S., Patros, C. H. G., and Joiner, T. E.

(2017). The interpersonal theory of suicide: A

systematic review and meta-analysis of a decade

of cross-national research. Psychological Bulletin,

143(12):1313–1345.

Chu, C., Klein, K. M., Buchman-Schmitt, J. M., Hom,

M. A., Hagan, C. R., and Joiner, T. E. (2015). Rou-

tinized Assessment of Suicide Risk in Clinical Prac-

tice: An Empirically Informed Update. Journal of

Clinical Psychology, 71(12):1186–1200.

Curtin, S. C., Hedegaard, H., and Ahmad, F. B. (2021).

Provisional numbers and rates of suicide by month

and demographic characteristics: United states, 2020.

NVSS-Vital Statistics Rapid Release.

Fleischmann, A., Lammers, J., Conway, P., and Galinsky,

A. D. (2019). Paradoxical effects of power on moral

thinking: Why power both increases and decreases

deontological and utilitarian moral decisions. Social

Psychological and Personality Science, 10(1):110–

120.

Fox, K. R., Huang, X., Guzm

an, E. M., Funsch, K. M., Cha,

C. B., Ribeiro, J. D., and Franklin, J. C. (2020). Inter-

ventions for suicide and self-injury: A meta-analysis

of randomized controlled trials across nearly 50 years

of research. Psychological Bulletin, 146(12):1117–

1145.

Franklin, J. C., Huang, X., and Bastidas, D. (2019). Vir-

tual reality suicide: Development of a translational

approach for studying suicide causes. Behaviour Re-

search and Therapy, 120:103360.

Huang, X., Funsch, K. M., Park, E. C., Conway, P.,

Franklin, J. C., and Ribeiro, J. D. (2021). Longitu-

dinal studies support the safety and ethics of virtual

reality suicide as a research method. Scientiﬁc Re-

ports, 11(1):9653.

Huang, X., Funsch, K. M., Park, E. C., and Franklin,

J. C. (2020). Anticipated consequences as the pri-

mary causes of suicidal behavior: Evidence from a

laboratory study. Behaviour Research and Therapy,

134:103726.

Joiner Jr., T. E., Walker, R. L., Rudd, M. D., and Jobes,

D. A. (1999). Scientizing and routinizing the assess-

ment of suicidality in outpatient practice. Professional

Psychology: Research and Practice, 30(5):447–453.

Nock, M. K., Holmberg, E. B., Photos, V. I., and Michel,

B. D. (2007). Self-injurious thoughts and behav-

iors interview: Development, reliability, and validity

in an adolescent sample. Psychological Assessment,

19(3):309–317.

O’Connor, R. C. and Kirtley, O. J. (2018). The integrated

motivational–volitional model of suicidal behaviour.

Philosophical Transactions of the Royal Society B: Bi-

ological Sciences, 373(1754):20170268.

Park, E. C., Harris, L. M., Sigel, A. N., Huang, X., Chen,

S., and Ribeiro, J. D. (2023). Is physical pain causally

related to suicidal behavior: An experimental test. Be-

haviour Research and Therapy, 165:104321.

Ribeiro, J. D., Bender, T. W., Selby, E. A., Hames, J. L.,

and Joiner, T. E. (2011). Development and validation

of a brief self-report measure of agitation: The brief

agitation measure. Journal of Personality Assessment,

93(6):597–604.

Ribeiro, J. D., Linthicum, K. P., Harris, L. M., Bryen, C. P.,

and Broshek, C. E. (2021). Raising doubt about the

anticipated consequences of suicidal behavior: Evi-

dence for a new approach from laboratory and real-

world experiments. Behaviour Research and Therapy,

147:103971.

Ribeiro, J. D., Witte, T. K., Van Orden, K. A., Selby, E. A.,

Gordon, K. H., Bender, T. W., and Joiner Jr, T. E.

(2014). Fearlessness about death: the psychometric

properties and construct validity of the revision to the

acquired capability for suicide scale. Psychological

assessment, 26(1):115.

Taylor, P. J., Gooding, P. A., Wood, A. M., Johnson, J.,

Pratt, D., and Tarrier, N. (2010). Defeat and entrap-

ment in schizophrenia: The relationship with suicidal

ideation and positive psychotic symptoms. Psychiatry

Research, 178(2):244–248.

Wang, S., Wei, T., Zhu, R., Li, S., Liu, X., Cai, Y., and

Gong, R. (2023). Perceived entrapment predicts ﬁrst-

onset suicidal ideation: A longitudinal study among

medical students in china. Frontiers in Public Health,

10.

WHO (2021). Suicide Worldwide In 2019: Global Health

Estimates. World Health Organization, Geneva, 1st ed

edition.

WHO ASSIST Working Group (2002). The alcohol, smok-

ing and substance involvement screening test (assist):

development, reliability and feasibility. Addiction,

97(9):1183–1194.

APPENDIX

A link to the game trailer can be found below.

https://youtu.be/GtSm41xPKxQ

Advancing Suicide Prevention Through Immersive Technology

889