Using VR to investigate bystander behavior and the motivational factors in school bullying

https://doi.org/10.1016/j.compedu.2022.104696Get rights and content

Highlights

  • Boys showed lower pro-bullying and outsider behavior in the VR group.

  • There were no differences in empathy or anger between the VR group and the text group.

  • There was no difference in defender self-efficacy between the VR group and the text group.

  • Anger and defender self-efficacy distinguish features of diverse bystander behavior.

Abstract

The main goal of this study is to develop a virtual reality (VR)-based scenario of bullying that is more effective in investigating the motivational factors (i.e., emotion, self-efficacy) of adolescents' bystander behavior than the traditional text method. The second goal is to examine how emotion and defender self-efficacy relate to adolescent bystander behavior. Eighth graders (N = 229) from Taiwan participated in 2 (presentation mode: VR vs. text) x 2 (participant sex: boys vs. girls) x 2 (actor sex: boys vs. girls) experimental trials. The results of MANCOVAs indicated that boys displayed less outsider and pro-bullying behavior in the VR group than in the text group; however, there were no differences in bystander behavior for girls. The subjects in the VR group experienced more fear and excitement, less empathy, and a lower level of defender self-efficacy than those in the text group. The results of regression analyses showed that higher levels of anger and defender self-efficacy significantly predicted positive bystander behavior. Qualitative data revealed that VR provided the subjects with a vivid and realistic bullying experience and enabled them to experience emotions such as anger, empathy, and fear. The findings demonstrate that VR increases students’ antibullying attitudes, justice emotions, and positive bystander behavior. However, VR does not provoke more empathy and self-efficacy, as expected. VR might capture a different dimension of response to bullying than the traditional text method. Nevertheless, VR has promise as an educational tool for developing bystander intervention in high school.

Introduction

Bullying is a prevalent problem in schools in numerous countries (UNESCO, 2019). It is recognized as having long-term negative impacts on the physical, psychosocial, and academic adjustment of children and adolescents (Jennings et al., 2019; Juvonen et al., 2011; Wu et al., 2016). Bystanders are often present during bullying episodes (Hawkins et al., 2001). Paying attention to bystander reactions is essential because bystanders can counteract bullying by intervening, thus affecting bullying prevalence (Nocentini et al., 2020; Salmivalli et al., 2011).

Researchers have investigated the factors associated with bystander behavior in bullying (Forsberg et al., 2014, 2018; Oh & Hazler, 2009; Thornberg et al., 2020). These studies have adopted various research methods, including surveys, interviews, and laboratory experiments. Although these methods can provide critical information on bullying and bystander action, they require subjects to recall previous experiences, imagine their possible behavior in a hypothetical context, or involve triggering corresponding behavior in an artificial laboratory situation. This methodological approach is applied out of context and cannot convey the urgency of a bullying incident or trigger real emotions (Pan & Hamilton, 2018).

Virtual reality (VR) technology creates a computer-based artificial environment that transmits synthetic sensory information to a user in a form that appears real rather than synthetic (Pan & Hamilton, 2018). The sense of immersion, involvement, presence, first-person vision and emotional features in VR environments is highly suitable for studying interpersonal interaction (Pan & Hamilton, 2018). The use of VR to investigate bystander behavior can solve the problems of experiential control, repeatability, and ecological validity. In a VR scenario, a rigorous experimental scenario can be designed according to the research purpose. Even after many experiments, the study can be directly replicated as needed across trials, providing considerable convenience (Pan & Hamilton, 2018).

VR is a powerful medium for addressing social problems such as bullying and aggression (Caceffo et al., 2022). Although VR has been used in educational programs, its application to bullying prevention or intervention remains limited. This study aims to develop a VR-based bullying scenario to investigate how adolescents’ psychosocial factors (e.g., emotion and self-efficacy) are related to bystander behavior in school bullying. We combine self-report and VR, which offer advantages over traditional methodological approaches alone. We expect this design to enable us to present the same situation to different subjects consistently, thus providing a more immersive experience than that presented in a written descriptive scenario. Moreover, compared to prior research that focuses on promoting empathy in facilitating intervention intention (Barreda-Ángeles et al., 2021; Ingram et al., 2019), we include other emotions (e.g., anger, fear, and excitement) and defender self-efficacy and examine the effect of these motivational factors in the VR presentation. The study findings should provide additional information on the essential emotional and motivational factors related to intervening in bullying for practitioners incorporating VR into the design of bystander intervention or programs.

Bullying is defined as aggressive behavior that involves an imbalance of power status and that is used intentionally and repeatedly over time (Olweus, 2001). Bullying is a group interaction that involves bullies, victims, and bystanders (Olweus, 1994). Bystanders are individuals who are present and witness bullying but are neither the perpetrator nor the victim (Salmivalli, 2010). Bystanders can deter the demoralizing and damaging impacts of bullying by intervening or defending the victim or can reinforce bullying by sustaining or even joining the bullying (Padgett & Notar, 2013; Salmivalli, 2010). Bystanders can respond to bullying incidents in three main ways: defending the victims, remaining outsiders, and assisting or reinforcing the bully (Salmivalli et al., 1996). Bystanders can support victims by taking sides with them, defending them, displaying seemingly antibullying behavior, expressing sympathy toward them, or seeking help from others. In contrast, bystanders might seek to avoid bullying situations or remain uninvolved. Bystanders can be assistors, who join in the bullying by catching or holding the victim, or reinforcers, who provide the bullies with positive feedback (e.g., laughing at the victims), therefore reinforcing the bullying (Jones et al., 2015; Salmivalli, 1999).

Previous research has indicated that bystander behavior has a critical influence on stopping or facilitating school bullying. When bystanders assist or reinforce bullying, it increases not only the perpetrator's aggressiveness and frequency of bullying in the classroom (Salmivalli et al., 2011) but also the negative impact on victims, such as higher anxiety, depression, and lower self-esteem (Salmivalli, 2010). However, when bystanders remain uninvolved, they unknowingly draw boundaries around the victim, encouraging bullying behavior (Padgett & Notar, 2013). In contrast, if bystanders speak out and take appropriate action, it signifies resistance to injustice, which can threaten bullies' status and discourage bullying (Padgett & Notar, 2013). Previous studies have suggested that when bystanders stand up for victims there is at least a 53% chance of effectively stopping the bullying (Hawkins et al., 2001). Recognizing why certain students choose to intervene and stop bullying (e.g., defenders) whereas others take other action (e.g., adopt the role of outsider or reinforcer) would help us better understand the group dynamics underlying bullying and provide helpful information for developing intervention programs.

Defending the victim and reinforcing the bully are two contrasting ways to take sides in bullying situations. The motivational and psychosocial underpinnings of three types of bystander behavior (i.e., defending, reinforcing, and remaining uninvolved) should differ. This study investigates how self-efficacy and emotions affect students’ adoption of these bystander behaviors and examines sex effects on these individual characteristics and bystander behavior.

Self-efficacy is defined as "individuals' belief in their capabilities to successfully organize and execute a particular action required to produce the desired result” (Thornberg & Jungert, 2013). In a bullying situation, helping the victim could involve a risk because it might result in the helper becoming a victim of a hostile perpetrator (Pöyhönen et al., 2012). Thus, persons wishing to help the victim must be confident in their ability to effectively intervene in bullying. If students lack knowledge of or skills related to what to do or fear being victimized, they are less likely to support victims (Thornberg & Jungert, 2013). Bystanders with higher defender self-efficacy are more likely to intervene in bullying (Thornberg & Jungert, 2013).

Research on self-efficacy and the roles of bystanders indicates that defender self-efficacy is positively associated with defending behavior but negatively related to outsider behavior (Thornberg et al., 2020). However, the relationship between defender self-efficacy and probullying behavior (e.g., assisting or reinforcing bullies) is undetermined. Several studies have found that defender self-efficacy is negatively correlated with reinforcing or assisting behavior in bullying (Thornberg et al., 2020), while other studies have found it to be nonsignificant (Pöyhönen et al., 2012). Such conflicting findings might be due to divergent research methods. However, Xie and Ngai (2020) found that defender self-efficacy among Chinese adolescents was not related to defending or outsider behavior but was negatively associated with probullying behavior. Therefore, additional empirical studies on bullying in Asian contexts are necessary.

When witnessing bullying, most students have an emotional reaction. In an interview with groups of primary and high-school students, Chaux (2005) found that 76.6% of the students reported having negative emotions (e.g., bad, fear, sadness, bothered, anger, and shame), 13.3% reported having neutral emotions, and 20% reported having positive emotions. These emotional reactions influence bystanders' decisions regarding whether to intervene.

Empathy, which is the extent to which a person understands and experiences the emotions of others (Barhight et al., 2013), is considered the most critical factor in motivating bystanders to intervene in bullying because it leads them to feel bad for the victims (Thornberg & Jungert, 2013). When bystanders sympathize with victims, they are more likely to help. In contrast, when bystanders experience less empathy toward victims, they tend to stay uninvolved or even join in the bullying. Studies have shown that empathy negatively correlates with probullying behavior (Jolliffe & Farrington, 2004) and with outsider behavior (Yang et al., 2017) but positively correlates with defender behavior (Caravita et al., 2009). However, other studies have found a negative relationship between empathy and defending behavior (Gini et al., 2008; Sutton et al., 1999). These inconsistent findings might be attributed to the empathy dimensions that are measured. Empathy consists of cognitive empathy and affective empathy, and the latter has been found to be more strongly associated with bullying behavior (Pozzoli et al., 2017). The current study measures subjects' affective empathy toward bullying victims.

Anger often occurs in those who disagree with bullying behavior and has thus been considered a moral emotion (Forsberg et al., 2018). When students believe that bullying is wrong and unjust, they feel angry, which further facilitates an intervening motivation (Forsberg et al., 2014, 2018). In contrast, fear might discourage students from intervening because they are afraid of becoming a further target of bullying (Thornberg & Jungert, 2013). Additionally, excitement and joy reflect an audience's interest in watching bullying. The audience's excitement is accompanied by clamor and agitation, which usually reinforces bullies (Thornberg & Jungert, 2013).

Studies on school bullying have indicated sex differences in bystander behavior. Boys are more likely to exhibit probullying behavior than girls, while girls tend to help or defend victims (Pozzoli & Gini, 2013; Trach Hymel, Waterhouse, & Neale, 2010). Relative to girls, boys have a higher tolerance for undesirable behavior, such as bullying, because they are inclined to believe that aggressive behavior is masculine. Boys are also easily influenced by peers, who might hold positive attitudes toward bullying (Pozzoli & Gini, 2013) and are thus more likely to assist or reinforce bullying (Pozzoli & Gini, 2013). Furthermore, there is a high likelihood that although boys know how to intervene, they stay outside the bullying situation to maintain their social status (Jenkins & Nickerson, 2017). In contrast, girls tend to have a more assertive attitude toward antibullying (Desrumaux et al., 2018), a higher sensitivity and awareness of bullying, lower moral disengagement (Thornberg & Jungert, 2013), and a stronger sense of responsibility to intervene (Pozzoli & Gini, 2013).

There are also sex differences in emotional reactions and defender self-efficacy. When witnessing bullying, girls experience more negative emotions and a higher level of empathy than boys (Barhight et al., 2013), which urges them to help victims. Conversely, boys' defender self-efficacy is higher than that of girls (Thornberg & Jungert, 2013), and high defender self-efficacy is positively related to defending behavior (Thornberg et al., 2020). These findings suggest that boys and girls might have different reasons to defend victims, but how they decide to intervene in bullying might be determined by situational factors. For instance, Hawkins et al. (2001) observed that boys and girls on campus are more likely to help a victim when they are of the same sex as the victim. However, Fischer et al. (2006), using an experimental design to examine bystander response in an emergency, found no sex differences in adults with regard to helping victims when encountering an emergency.

Although VR is not a new technology, with the development of a cost-affordable head-mounted display (HMD) for personal VR use, the technology has received increasing attention in recent years and triggered innovative change in education and psychology (Howard & Gutworth, 2020; Pan & Hamilton, 2018). VR presents a three-dimensional environment that emulates real or imagined space with interactive properties and a sense of immersion and presence, placing the user in a virtual world with extraordinarily intense and authentic reactions (Pan & Hamilton, 2018). VR has been used in educational programs to train or develop social skills, such as public speaking (Sulter et al., 2022) and interpersonal communication ability (Hwang et al., 2022), and to enhance learning performance by increasing learner motivation (Lin et al., 2019) and self-efficacy (Lin & Wang, 2021).

VR has substantial potential for resolving issue that have concerned researchers in education and psychology, including experimental control, reproducibility, and ecological validity (Pan & Hamilton, 2018). VR can well control a social interaction experiment by manipulating variables individually. It also facilitates the controlled staging of situations that are difficult to establish in real life while enabling natural behavior to be triggered (Kozlov & Johansen, 2010). In a VR scenario, subjects can experience triggered reactions that are similar to real-life situations but in a safe context; such an experiment can be implemented repeatedly as needed across trials (Pan & Hamilton, 2018), which increases the validity and reliability of such studies.

VR has been used to test the bystander effect and examine the effectiveness of VR in triggering users' reactions when witnessing an emergency event (e.g., interpersonal violence or a help-needed situation) in a VR scenario (Rovira et al., 2021; Rovira & Slater, 2022). These studies have indicated that subjects’ experiences in VR are similar to those that are elicited in real-life situations. Related studies have also found that the key elements that motivate subjects to intervene are negative emotions (e.g., anger, discomfort, empathy) (Buckler et al., 2019), the social identity of the others who are present (Rovira et al., 2021; Rovira & Slater, 2022; Slater et al., 2013), the number of the bystanders present (Rovira et al., 2021; Rovira & Slater, 2022), a sense of personal responsibility (Buckler et al., 2019), and possessing the skills needed to intervene Buckler et al. (2019).

VR has also been used in bullying prevention or intervention programs. Such intervention programs aim to enhance students' empathy for victims since bystanders often justify their lack of empathy by blaming the victim (Stassen Berger, 2007). Promoting empathy toward victims appears to be essential. For example, Ingram et al. (2019) conducted a pseudorandomized trial of a virtual reality-enhanced bullying prevention program among middle-school students in the U.S. They found that compared with the control group, students in the VR group displayed a higher level of change in empathy, which then increased their willingness to intervene. Additionally, Barreda-Ángeles et al. (2021) used a within-subject experimental design involving high-school students to investigate whether the VR presentation mode was better at eliciting more empathetic emotions toward the victims than a traditional computer presentation. Their results supported the effectiveness of VR in provoking realistic emotional reactions to bullying.

In a review of the application of VR in bystander behavior in interpersonal violence, Xue et al. (2021) indicated that to create an effective bullying simulation the researcher should design virtual avatars and VR scenarios appropriately, such as by customizing the characteristics of the avatar (i.e., sex, grade, age, appearance), the experience (i.e., scripts and scenarios), and the environment (i.e., classroom and hallway) (Oyekoya et al., 2021).

Prior research has indicated that VR facilitates users' adoption of the victim's perspective, fosters empathy toward the victim, and helps change users' attitudes and behaviors. However, these studies did not closely examine whether participant and actor sexes make a difference (Xue et al., 2021). Moreover, we believe that effective VR intervention should consider various motivational factors that enhance positive bystander behavior. Therefore, we included other emotions (i.e., anger, fear, and excitement) in the study in addition to empathy, which is usually the focus emotion in previous studies (Barreda-Ángeles et al., 2021; Ingram et al., 2019).

In addition, the current study aimed to understand how VR experiences might change bystanders' efficacy beliefs with respect to intervening in a bullying situation. According to the cognitive and affective model of immersive learning (CAMIL), the technological factors of VR, including immersion, control, and representational fidelity, enhance users' self-efficacy and motivation through feelings of presence and agency (Makransky & Petersen, 2021). Presence, a feeling of being there, enables users to sense the virtual experience as real, and such novel, intense experiences trigger their immediate interest, promoting their intrinsic motivation and resulting in increased self-efficacy. Agency, a feeling of perceived control with respect to taking action in the environment, increases one's estimate of self-efficacy (Makransky & Petersen, 2021). Thus, we suppose that students in a VR environment might experience higher defender self-efficacy.

Furthermore, we customized the VR avatars and VR environment to fit our target group (8th graders) and their school environment (junior high school) because design appropriateness impacts the effectiveness of VR presentation (Oyekoya et al., 2021). Our research questions are as follows:

  • 1.

    How effective is the VR presentation of a bullying scenario in eliciting emotion, efficacy beliefs, and defending bystander behavior compared to the traditional text presentation when considering the sexes of the subjects and the actors?

  • 2.

    How do emotions (i.e., empathy, anger, fear, and excitement) and efficacy (e.g., defender self-efficacy) predict bystander behaviors when controlling for presentation mode, the sexes of the subjects and the actors, and previous bullying experience?

We conducted a 2 (VR mode vs. text mode) x 2 (participant sex) x 2 (actor sex) between-subject experimental trial. The main effects and the interaction effect of the presentation mode, participant sex, and actor sex were assessed to answer the first question. We hypothesized that the subjects in the VR group would have more immersive and profound experiences of emotions, defender self-efficacy, and bystander behavior than those in the text group and that these differences would vary with the sexes of the subjects and the actors. Regarding the second question, we hypothesized that the emotions of anger and empathy and defender self-efficacy would positively predict bystanders’ defending behavior but negatively predict probullying or outsider behaviors.

Section snippets

Experimental design

This study adopted a 2 (participant sex: male vs. female) x 2 (actor sex: male vs. female) x 2 (presentation mode: VR vs. text) between-subjects design, in which three independent variables, with two levels each, were manipulated. Each participant experienced only one condition. After presenting the scenario, the subjects were asked to respond to a series of questions about emotions, defender self-efficacy, and bystander response.

Subjects

In a 2 × 2 x 2 experimental design, the minimum sample size to

Bystander behavior tendency

The results of the three-way MANCOVA indicated a significant main effect for mode of presentation (Pillai's Trace = 0.091, F (3,217) = 7.23, p < .001) (Table 1). However, the main effects for participant sex (Pillai's Trace = 0.013, F (3,217) = 0.97, p = .41) and actor sex (Pillai's Trace = 0.02, F (3,217) = 1.62, p = .19) were not significant. There was a significant two-way interaction between mode of presentation and participant sex (Pillai's Trace = 0.06, F (3,217) = 4.42, p < .01).

General discussion

This study used VR to investigate bullying bystander behavior and its influencing factors. The following sections discuss the findings and provide insight regarding further application of VR in bullying bystander research.

Conclusion and limitations

This study used VR to compare individuals' emotions, defender self-efficacy, and bystander behaviors during bullying episodes to those of a text group. Thus, the study contributes to two areas of VR application: bystander effects and bullying intervention research. The study indicates that VR is a useful tool for studying bystander behavior in high-school students. VR increased students' awareness of the severity and urgency of bullying, enabled them to witness how the victim suffered, and

Credit author statement

Yih-Lan Liu: Conceptualization, Methodology, Investigation, Resources, Writing – original draft, Project administration, Funding acquisition. Chia-Yun Chang: Project administration, Data curation, Formal analysis. Cheng-Yan Wang: HLM data analysis and explanation.

Acknowledgments

The authors are extremely grateful to all participants in our study. We also thank our students C. Y. Wang, Y.T. Chang, and T. G, Jian for their indispensable help in data collection. This work was supported by the Ministry of Science and Technology of the Republic of China under Contract No. MOST 108-2511-H-009-012 -

References (57)

  • C. Salmivalli

    Bullying and the peer group: A review

    Aggression and Violent Behavior

    (2010)
  • K. Stassen Berger

    Update on bullying at school: Science forgotten?

    Developmental Review

    (2007)
  • R.E. Sulter et al.

    SpeakApp-Kids! Virtual reality training to reduce fear of public speaking in children-A proof of concept

    Computers & Education

    (2022)
  • R. Thornberg et al.

    Bystander behavior in bullying situations: Basic moral sensitivity, moral disengagement, and defender self-efficacy

    Journal of Adolescence

    (2013)
  • H. Xie et al.

    Participant roles of peer bystanders in school bullying situations: Evidence from Wuhan, China

    Children and Youth Services Review

    (2020)
  • L.R. Barhight et al.

    Children's physiological and emotional reactions to witnessing bullying predict bystander intervention

    Child Development

    (2013)
  • D.G. Buckler et al.

    Observing the stages of bystander intervention in virtual reality simulation

    World Journal of Emergency Medicine

    (2019)
  • S.C.S. Caravita et al.

    Unique and interactive effects of empathy and social status on involvement in bullying

    Social Development

    (2009)
  • E. Chaux

    Role of third parties in conflicts among Colombian children and early adolescents

    Aggressive Behavior

    (2005)
  • Y.-Y. Cheng et al.

    Report of the survey and implementation plan for school bullying

    (2009)
  • P. Desrumaux et al.

    Workplace bullying: How do bystanders’ emotions and the type of bullying influence their willingness to help?

    Nordic Psychology

    (2018)
  • P. Fischer et al.

    The unresponsive bystander: Are bystanders more responsive in dangerous emergencies?

    European Journal of Social Psychology

    (2006)
  • C. Forsberg et al.

    Bystanders to bullying: Fourth- to seventh-grade students' perspectives on their reactions

    Research Papers in Education

    (2014)
  • C. Forsberg et al.

    Students' views of factors affecting their bystander behaviors in response to school bullying: A cross-collaborative conceptual qualitative analysis

    Research Papers in Education

    (2018)
  • J.F. Hair et al.

    Multivariate data analysis

    (1998)
  • D.L. Hawkins et al.

    Naturalistic observations of peer interventions in bullying

    Social Development

    (2001)
  • G.J. Hwang et al.

    The effectiveness of the virtual patient-based social learning approach in undergraduate nursing education: A quasi-experimental study

    Nurse Education Today

    (2022)
  • L.N. Jenkins et al.

    Bullying participant roles and gender as predictors of bystander intervention

    Aggressive Behavior

    (2017)
  • Cited by (5)

    1

    Present address: National Yang Ming Chiao Tung University, 1001 University Road, Hsinchu, 300, Taiwan, ROC.

    View full text