Developing an Understanding of Victims and Violent Offenders

Abstract

This study explores the consequences of fostering empathy—for both victims and perpetrators—after large-scale violent events. Participants (N = 834) read a description of a school shooting and were randomly assigned to one of six conditions revealing varying amounts of background information about the victim and the perpetrator of violence. The impact of empathy on reactions toward the victim and perpetrator were then assessed. Empathy for the perpetrator could be fostered with increased information about his background, resulting in recommendations of increased leniency. Fostering empathy for the victim promoted positive community responses, including increased intentions to engage in helping behavior and make charitable donations. The degree to which participants could make sense of the violent event was also associated with decreases in blame and anger toward the perpetrator. Potential implications of the findings for news media and community coping strategies are explored.

Keywords

empathy trauma coping school shootings

On December 14, 2012, Adam Lanza walked into Sandy Hook Elementary School and murdered 20 children and 6 adults. Only 5 months earlier, James Holmes walked into a packed movie theater in Aurora, Colorado, and opened fire, killing 12 people and injuring another 58. In 2011, Jared Loughner shot and killed 6 people in Tucson, Arizona, wounding 13 others. On April 16, 2007, Seung-Hui Cho shot and killed 32 people at Virginia Tech University before shooting and killing himself. And less than 1 year later, Steven Kazmierczak shot and killed 6 people before shooting himself at Northern Illinois University. As these violent tragedies make national headlines with increasing frequency, communities are left shocked and shattered. Limited systematic research has examined how communities react to victims and offenders after such violence occurs (see, for example, Wicke & Silver, 2009), and there is a need for further investigation into variables that can help individuals make sense of these tragedies. The present study explores one critical variable that may assist coping in the aftermath of a violent tragedy: empathy. Empathy may increase helping behavior toward victims, facilitate meaning making, and foster forgiveness after violence. Specifically, we examine the influence of empathy on reactions to victims, reactions to perpetrators, and the ability to cope after community violence.

Empathy and the Meaning-Making Process

In general, people tend to believe that the world is a benevolent and meaningful place that provides individuals with feelings of security. However, experiencing a violent event can challenge these beliefs, resulting in fear and anger (Janoff-Bulman, 1992). It has been theorized that encountering traumatic events can lead to a search for meaning among those who witness or experience them (Silver & Updegraff, 2013). For example, people possess global meaning in their lives, which involves their beliefs about the world and their larger goals (Park & Folkman, 1997). Situational meaning involves the interpretation of global meanings within the context of specific events (i.e., my college campus is a safe place to go each day to get an education). After a trauma occurs, the meaning-making process involves reconciling global meanings with situational ones (Park & Folkman, 1997). Thus, after a school shooting on a college campus, people have to reconcile how such a violent event could occur in a place they had previously considered safe and secure. Reconciling traumatic events with one’s worldview likely involves searching for “why” an event happened (Silver & Updegraff, 2013). Coming up with an answer to “why?” can facilitate adjustment (Updegraff, Silver, & Holman, 2008).

After violent traumatic events, developing empathy for offenders by understanding their stories may be an integral part of the meaning-making process. In one of the only studies conducted immediately after the Columbine High School shootings, researchers interviewed a small sample of students and parents affected by the tragedy. Results suggested that people who were able to make sense of the shooting did so by empathizing with the two perpetrators and attempting to understand their motivations (Hawkins, McIntosh, Silver, & Holman, 2004). The authors posit that future studies should examine the role of empathizing with perpetrators in communities affected by trauma. Understanding the background of perpetrators may foster empathy and meaning making, facilitating the coping process after community tragedies.

Fostering Empathy

Empathy is defined as either an affective trait (the ability to experience the emotions of another person) or a cognitive ability (the ability to comprehend the emotions of another person; Jolliffee & Farrington, 2006). It involves taking the perspective of another person, while feeling emotionally engaged with that perspective (de Waal, 2008). Empathy is often described as the process of being able to perceive what another person is experiencing and responding to his or her situation (Davis, 1996). Situational empathy—or state empathy—refers to the ability to place oneself in the situation of another actor and to imagine others’ thoughts, feelings, and reactions when taking their role (Plumm & Terrance, 2009). People vary in their trait tendency to empathize with others, which is a stable individual difference factor known as “dispositional empathy” (Archer, Diaz-Loving, Gollwitzer, Davis, & Foushee, 1981). Beyond one’s dispositional level of empathy, empathy can also be fostered cognitively through perceived similarity and perspective taking.

Fostering Empathy Through Similarity

Similarity increases the likelihood that empathic cognitive processes such as role-taking occur, fostering empathy. For example, in a study in which undergraduate students met with a confederate who they were told had cancer, perceived similarity between the participant and the confederate predicted warm and supportive responses during the interaction (Westmaas & Silver, 2006). Having an increased understanding of a victim’s story may highlight similarities between the victim and the observer, leading to empathic processes. For example, a study of women who had never been mothers, were pregnant, or had just given birth found that participants had greater empathetic concern for other women when they had experienced similar life events (Hodges, Kiel, Kramer, Veach, & Villanueva, 2010). Another study by Brown, Wohl, and Exline (2008) found that identifying with the United States after the 9/11 terrorist attacks was associated with more blame and less forgiveness toward the terrorists, or a more “reactive empathy.” This phenomenon has also been demonstrated in mock trials of rape cases, where women who see themselves as similar to the victim are more likely to believe her story and convict the perpetrator (Grubb & Harrower, 2009; Weir & Wrightsman, 1990).

In theory, the more similar a person is to a perpetrator of a crime and the more empathy felt, the more lenient he or she is likely to be in assigning blame and responsibility for the criminal act (see Davis, Bray, & Holt, 1977). That is, a juror who perceives him or herself as similar to a defendant may be more likely to have had similar life experiences, which should lead to more forgiving interpretations of the defendant’s actions. For example, a recent vignette study concerning an abused woman who killed her abuser found that participants who were more distressed by their own experiences with partner violence saw themselves as more similar to the defendant. As a result, participants felt more empathy toward her and found her less culpable of the crime (Stein & Miller, 2012).

However, it is possible that increased similarity to a perpetrator may have the opposite impact, known as “the black sheep effect” (Marques, 1990). People want to maintain positive self-images and positive beliefs about the group(s) to which they belong. If people see themselves as similar to a perpetrator who has done something violent, they may feel more anger toward that person to protect their positive self- or group image. Overall, research over the past 30 years seems overwhelmingly to favor the similarity-leniency effect for defendants, rather than the opposite. Perceptions of similarity between mock jurors and defendants in attitudes (Griffitt & Jackson, 1973), religious background (Kerr, Hymes, Anderson, & Weathers, 1995), native language (Stephan & Stephan, 1986), and race (Johnson et al., 2002) have resulted in recommendations for more lenient sentences.

Fostering Empathy Through Perspective Taking

It is also possible to increase empathic response toward individuals without similarity, by simply asking people to take another’s perspective and facilitating emotional engagement with that perspective (de Waal, 2008). In the case of violence, when an observer is better able to take the perspective of a victim (or his or her family), there will likely be an increase in supportive behaviors offered. This was demonstrated by Anastasio and Costa (2004), who manipulated the amount of information provided about a victim of a violent crime. Referring to the victim by name and describing additional personal information increased the level of empathy felt for the victim. Facilitating empathy toward a victim through perspective taking has reliably predicted helping and altruistic behavior (Davis, 1996), and increases in empathy have been shown to contribute to charitable giving (Sargeant, 2010).

However, taking the perspective of perpetrators can, in turn, foster empathy for them. Increases in empathy through perspective taking, even when the observer is dissimilar to the perpetrator, can contribute to leniency. For example, Hurst and Foley (2005) explored the role of empathy in influencing sentencing decisions in the case of filicide. Undergraduates read scenarios about a woman who killed her two children and pled not guilty by reason of insanity. Perspective-taking ability influenced participants’ feelings about the mother’s culpability and her punishment. That is, participants who scored higher on a perspective-taking scale were more likely to find the mother not guilty by reason of insanity. In addition, Plumm and Terrance (2009) conducted mock trials using a case of a woman who killed her husband in self-defense. The opening and closing statements during the trial were designed either to induce empathy for the woman on trial or not. In the empathy induction condition, lawyers in the case asked the jury to imagine themselves in that woman’s situation that night. Participants in this condition rated the woman on trial as more mentally stable and were less likely to find her guilty (see also Archer, Foushee, Davis, & Aderman, 1979; Haegerich & Bottoms, 2000, for similar results).

Much of the literature on fostering empathy for perpetrators has focused on domestic and family violence, and often involves female offenders—perhaps because it is easier to foster empathy for this genre of offender. The current study focuses on fostering empathy for victims and offenders after a mass shooting on campus by a male offender and examines how empathy impacts potential behaviors toward the victims and perpetrators after this type of violence.

We predicted that dispositional empathy would be associated with the amount of empathy felt for both victims and perpetrators. We hypothesized that the background information provided about the victim and perpetrator meant to foster similarity, as well as the background information provided about the perpetrator meant to foster perspective taking, would increase empathy felt toward each of them. We also hypothesized that empathy for the victim would predict positive community responses, along with more negative responses toward the perpetrator (i.e., increased anger and blame). Finally, we hypothesized that both empathy for the perpetrator and the ability to make sense of the event would predict positive responses toward the perpetrator (i.e., forgiveness).

Method

The Present Study

In the present study, we varied the degree and content of background information provided after a school shooting to examine which scenario fostered the most empathy for victims and perpetrators. In addition, we explored how fostering empathy for victims and perpetrators promoted positive community responses and influenced reactions toward the perpetrator (i.e., feelings of blame and anger toward him). We used a 2 × 3 factorial design with two conditions for the victim (positive information about the victim’s background to foster feelings of similarity and a neutral information control condition) and three conditions for the perpetrator (positive information about the perpetrator’s background to foster feelings of similarity, traumatic background information to facilitate perspective taking, and a neutral background information control condition). We also assessed the relative importance of participants’ degree of dispositional empathy in responses toward the victim and perpetrator.

Participants

Participants were 861 undergraduate students enrolled in psychology courses at a large public university on the west coast of the United States. Undergraduate students were an appropriate sample for this study as college and university settings are potential sites for school shootings. Twenty-seven participants were dropped due to substantial missing data (more than 25%), for a total of 834 participants. Participants were 75% female, ranging in age from 17 to 43 years (M = 20.2, SD = 1.94). The majority of participants were Asian American (53.4%), followed by Caucasian (20.2%), Latino (15.1%), and African American (2.1%), which is consistent with the demographic make-up of the university (55.7% of admitted applicants were Asian American in the fall of 2012). Two percent of participants did not report their ethnicity, and 9.1% reported “Other.”

Measures

Interpersonal reactivity index

The Interpersonal Reactivity Index is a 28-item scale that assesses dispositional empathy (Davis, 1980) comprised of four 7-item scales: perspective taking, fantasy, empathetic concern, and personal distress. Each question is answered on a 5-point Likert-type scale ranging from 1 (“does not describe me well”) to 5 (“describes me very well”). Internal reliability on the four scales ranges from 0.71 to 0.77 and test–retest reliabilities on the four scales range from 0.62 to 0.71 (Davis, 1980). As with almost all empathy measures, females tend to score higher than males on the scale (Davis, 1980). Total scores on the Interpersonal Reactivity Index for the present sample ranged from 25 to 105, with a mean score of 69.2 (SD = 11.5) and a Cronbach’s alpha = .78. There were no significant differences between the experimental groups on dispositional empathy scores.

Level of empathy

A series of questions assessed empathy for both the victim (Daniel, 5 items) and the perpetrator (Tyler, 7 items), which were developed based on the Interpersonal Reactivity Index (Davis, 1980). Each item was answered on a 5-point scale (with endpoints from 0 = strongly disagree to 4 = strongly agree). Questions concerning empathy for the victim included items such as, “I feel sorry for Daniel’s family” and “I am touched by Daniel’s story.” Total scores on the victim empathy questions ranged from 0 to 20, with a mean score of 15.9 (SD = 3.5) and a Cronbach’s alpha = .83. Questions concerning empathy for the perpetrator included items such as “I feel sorry for Tyler” and “I am touched by Tyler’s story.” Total scores on the perpetrator empathy questions ranged from 0 to 24, with a mean score of 15.9 (SD = 4.4) and a Cronbach’s alpha = .80.

Positive community responses

Questions assessing positive community responses were designed for the purpose of this study and were assessed with 4 items, each answered on 5-point scales (with endpoints 0 = strongly disagree to 4 = strongly agree). Questions included items such as “I would give to a charitable fund in honor of Daniel” and “I would reach out to Daniel’s friends at my school to offer support.” Total scores on the positive community response scale ranged from 0 to 16, with a mean score of 10.2 (SD = 3.3) and a Cronbach’s alpha = .81. Average community responses tended to be positive (M = 2.54, SD = 0.83), above the midpoint of the scale

Negative response toward the perpetrator

Questions assessing negative responses toward the perpetrator (Tyler) were designed for the purpose of this study and were also assessed with 4 items each, answered on 5-point scales (with endpoints 0 = strongly disagree to 4 = strongly agree). Questions included items such as “If he were alive for a trial, I would want Tyler to receive the worst punishment available” and “Only Tyler is to blame for shooting Daniel.” Total scores on the negative response toward the perpetrator scale ranged from 1 to 20, with a mean score of 9.8 (SD = 3.3) and a Cronbach’s alpha = .65. Overall, mean scores tended to be negative (M = 1.96, SD = 0.66), below the midpoint of the scale.

Making sense of the event

The degree to which participants could make sense of the violent act was assessed using a question based on prior research (Silver & Updegraff, 2013) and scored on a 5-point scale that ranged from 0 to 4 (0 = not at all to 4 = a great deal). Scores on the scale were normally distributed, with a mean of 1.87 (SD = 1.07).

Demographics

Participant’s gender, ethnicity, age, birth place, and parents’ income were also collected.

Procedures

Participants volunteered to participate for course credit on a social science experiment website. Participants were first shown an information page online that informed them that the study examined judgments following a violent event and would take approximately 30 min to complete. After reading the description, individuals consented to participate by checking a button at the bottom of the page. Participants first completed the Interpersonal Reactivity Index (Davis, 1980) to assess baseline dispositional empathy. All participants then read the same passage describing a school shooting that took place at a university, which was based on the actual events that took place at Northern Illinois University in 2008:

On March 16, 2008 at approximately 4:15 p.m., Tyler Mason walked into a large auditorium in Raven Hall. An Introduction to Psychology course was in progress with over 200 students in attendance. Using the door attached directly to the stage, Tyler walked into the auditorium wearing a long black coat. From underneath the coat, he took out a shotgun in one hand, and a hand gun in the other. Tyler started firing bullets into the auditorium of students from the stage. Students ran frantically out of the auditorium and tried to hide behind their chairs. After a few minutes of firing, Tyler shot himself in the head, killing himself before the police arrived. Six students were killed during the shooting, and 18 were wounded.

Following the scenario, participants were randomly assigned to one of six combinations of information about the victim and perpetrator:

Description of the victim

Participants were randomly assigned to one of two conditions (50% of participants to each condition): basic or positive background regarding the victim, Daniel. Basic information indicated that he was 19 years old, a college sophomore, from California, and studying psychology and biology to go to medical school. Additional positive information about the victim was provided to half the participants including what he wanted to study in medical school, his charity work, quotes from his family and friends, his interests and goals, and his close relationship with his younger brother. This information, which was based on public information about an actual victim of the Northern Illinois University shooting, was meant to foster feeling of similarity with the victim.

Description of the perpetrator

Participants were randomly assigned to one of three conditions: basic, positive, or traumatic background regarding the perpetrator, Tyler. Basic information indicated that he was 22 years old, from California, raised by his mother, and attending graduate school to obtain his master’s degree in social work. In addition to the basic background, additional positive information was provided to one third of the participants. They learned that he was a straight A student, read quotes from his classmates and professors about his kindness, the fact that he bought a diamond ring to get engaged to his girlfriend, and lack of history of violent behavior (to foster similarity because the perpetrator is a “good person”). In addition to the basic background, one third of the participants learned about Tyler’s traumatic background (to foster perspective taking), including that he grew up in poverty, that his mother suffered from delusions, that he was neglected, stuttered, was teased in school, had a history of suicide attempts, was diagnosed with paranoid schizophrenia, and that he stopped taking his medication. All of the positive and traumatic background information was based on public information about the shooter at Northern Illinois University.

After reading the scenarios and descriptions of the people involved, all participants answered the same set of questions that assessed (a) level of empathy (for the victim and for the perpetrator), (b) positive community responses (i.e., likelihood of providing support to the victims’ family and friends, donating to charity), (c) negative responses toward the perpetrator (i.e., blame and anger directed at the perpetrator), (d) ability to make sense of the event, and (e) demographic background information.

Ethics approval for the study was obtained from the Institutional Review Board of the University of California, Irvine; all participants provided informed consent prior to their participation.

Data Analysis

Composite variables were created to represent dispositional empathy, empathy for the victim, empathy for the perpetrator, positive community responses, and negative responses toward the perpetrator. A stepwise ordinary least squares regression analysis was first used to examine which variables predicted empathy toward the victim and empathy toward the perpetrator (participant demographics, dispositional empathy, and background provided). Next, stepwise ordinary least squares regression analysis was used to examine which variables predicted positive community reactions and negative reactions toward the perpetrator. All of the dependent variables were approximately normally distributed.

Results

Predicting Empathy Toward the Victim

Table 1 reports the results of the analysis predicting empathy toward the victim. In the first model, the covariates of gender and race were entered as predictors of empathy toward the victim. As expected, females showed more empathy toward the victim than did males. In the second model, elevated dispositional empathy scores significantly predicted empathy toward the victim, but gender was no longer a significant predictor when dispositional empathy scores were included, indicating that the gender differences were fully explained by differences in dispositional empathy. In the final model, background information about the victim (basic or positive) and the perpetrator (basic, positive, or traumatic) were entered into the regression model. Neither background information about the victim nor the perpetrator significantly predicted empathy felt toward the victim of the shooting. Dispositional empathy level remained the only significant predictor in the model, explaining 7.0% of the variance in empathy toward the victim (Table 1).

Table 1.

Predictors of Greater Empathy Toward the Victim: OLS Regression (N = 834).

	Model 1			Model 2			Model 3
Predictor Variable	b	β	t-Ratio	b	β	t-Ratio	b	β	t-Ratio
Gender^a	−0.19	−0.13	(−3.64)**	−0.09	−0.06	(−1.65)	−0.09	−0.06	(−1.64)
Race^b
Asian American	−0.07	−0.05	(−1.21)	−0.09	−0.07	(−1.56)	−0.09	−0.07	(−1.52)
Latino	−0.10	−0.05	(−1.22)	−0.10	−0.05	(−1.26)	−0.09	−0.05	(−1.21)
Other	−0.02	−0.01	(−0.26)	−0.03	−0.02	(−0.39)	−0.03	−0.01	(−0.36)
Dispositional empathy score				0.40	0.24	( 6.94) **	0.40	0.24	( 6.90)**
Victim^c
Positive background							0.02	0.02	(0.47)
Perpetrator^c
Positive background							0.00	0.00	(0.04)
Traumatic background							−0.07	−0.05	(−1.26)
F(df )	3.92	(4, 830)**		12.93	(5, 829)**		8.38	(8, 826)**
Adjusted R²	0.01			0.07			0.07

Note. OLS = ordinary least squares.

Male = 1, Female = 2.

Reference group for race is Caucasian.

Reference group is basic background information.

p < .05. **p < .01.

Predicting Empathy Toward the Perpetrator

Empathy toward the victim and empathy toward the perpetrator were significantly correlated (r = .38, p < .001). Table 2 reports the results of the analysis predicting empathy toward the perpetrator. In the first model, the covariates of gender and race were entered as predictors of empathy toward the perpetrator. Females showed more empathy toward the perpetrator than males. In the second model, elevated dispositional empathy scores significantly predicted empathy toward the perpetrator, but gender was no longer a significant predictor, again indicating that the gender difference was fully explained by differences in dispositional empathy. In the final model, background information on the victim and perpetrator were entered. The victim’s background information was not a significant predictor of empathy toward the perpetrator. However, background information about the perpetrator significantly predicted empathy toward him. Participants who read positive background about the perpetrator had more empathy for him than participants who only read basic background. In addition, participants who read about the perpetrator’s traumatic background reported more empathy for him than participants who only read basic background information. Dispositional empathy level remained a significant predictor in the final model, which explained 8.0% of the variance in empathy toward the perpetrator.

Table 2.

Predictors of Greater Empathy Toward the Perpetrator: OLS Regression (N = 834).

	Model 1			Model 2			Model 3
Predictor Variable	b	β	t-Ratio	b	β	t-Ratio	b	β	t-Ratio
Gender^a	−0.21	−0.13	(−3.77)**	−0.09	−0.06	(−1.64)	−0.10	−0.06	(−1.79)
Race^b
Asian American	0.02	0.02	(0.35)	0.00	0.00	(0.01)	0.00	0.00	(−0.06)
Latino	−0.04	−0.02	(−0.44)	−0.04	−0.02	(−0.45)	−0.04	−0.02	(−0.54)
Other	0.02	0.01	(0.23)	0.01	0.00	(0.12)	0.00	0.00	(0.03)
Dispositional empathy score				0.45	0.26	(7.50)**	0.45	0.26	(7.50)**
Victim^c
Positive background							−0.06	−0.04	(−1.19)
Perpetrator^c
Positive background							0.16	0.11	(2.75)**
Traumatic background							0.14	0.09	(2.38)*
F(df )	3.71	(4, 830)**		14.40	(5, 829)**		10.33	(8, 826)**
Adjusted R²	0.01			0.07			0.08

Note. OLS = ordinary least squares.

Male = 1, Female = 2.

Reference group for race is Caucasian.

Reference group is basic background information.

p < .05. **p < .01.

Predicting Positive Community Responses

Table 3 reports the results of the analysis predicting positive community responses. In the first model, gender was a significant predictor of positive responses toward the victim, with females more likely to indicate that they would take action in the community than males. Asian American students were significantly less likely than White students to report positive community responses. In the second model, increased dispositional empathy levels significantly predicted likelihood of positive community responses; gender was no longer a significant predictor. Asian American students remained significantly less likely than White students to report the likelihood of positive community responses. In the third model, empathy toward the victim and empathy toward the perpetrator were added as predictors of positive community responses. Empathy toward the victim was the strongest predictor of positive community responses. In addition, increased empathy toward the perpetrator predicted positive community responses. Dispositional empathy also remained a significant predictor of positive community responses, as did Asian American ethnicity. The final model, which included the background of the victim and perpetrator, explained 23.9% of the variance in positive community responses.

Table 3.

Predictors of Positive Community Responses: OLS Regression (N = 834).

	Model 1			Model 2			Model 3			Model 4
Predictor Variable	b	β	t-Ratio	b	β	t-Ratio	b	β	t-Ratio	b	β	t-Ratio
Gender^a	−0.25	−0.13	(−3.86)**	−0.09	−0.05	(−1.44)	−0.04	−0.02	(−0.73)	−0.04	−0.02	(−0.74)
Race^b
Asian American	−0.14	−0.09	(−2.00)*	−0.17	−0.11	(−2.48)*	−0.13	−0.08	(−2.11)*	−0.14	−0.08	(−2.12)
Latino	−0.15	−0.07	(−1.61)	−0.15	−0.07	(−1.68)	−0.11	−0.05	(−1.29)	−0.11	−0.05	(−1.31)
Other	−0.01	−0.00	(−0.05)	−0.02	−0.01	(−0.20)	−0.01	0.00	(−0.08)	−0.01	0.00	(−0.07)
Dispositional empathy score				0.59	0.30	(8.69)**	0.37	0.19	(5.61)**	0.37	0.19	(5.63)**
Empathy—Victim							0.42	0.34	(10.53)**	0.42	0.35	(10.51)**
Empathy—Perpetrator							0.13	0.11	(3.45)**	0.13	0.11	(3.38)**
Victim^c
Positive background										−0.03	−0.02	(−0.59)
Perpetrator^c
Positive background										0.01	0.00	(0.10)
Traumatic background										0.01	0.01	(0.17)
F(df )	5.29	(4, 830)**		19.72	(5, 829)**		38.98	(7, 827)**		27.24	(10, 824)**
Adjusted R²	0.020			0.101			0.242			0.239

Note. OLS = ordinary least squares.

Male = 1, Female = 2.

Reference group for race is Caucasian.

Reference group is basic background information.

p < .05. **p < .01.

Predicting Negative Responses Toward the Perpetrator

Table 4 reports the results of the analysis predicting negative responses toward the perpetrator. In the first model, gender was a significant predictor of negative reactions toward the perpetrator, with males more likely to report negative reactions than females. In addition, all ethnicities, including Asian American, Latino, and Other, reported less negative reactions to the perpetrator than did Caucasian participants. In the second model, increased dispositional empathy significantly predicted decreases in negative reactions toward the perpetrator; gender was no longer significant. Ethnicity remained a significant predictor. In the third model, empathy toward the perpetrator was the strongest predictor of negative responses toward the perpetrator, leading to fewer negative reactions toward him. Empathy toward the victim predicted significantly more negative reactions toward the perpetrator. Ethnicity and dispositional empathy scores also remained significant predictors. In the fourth model, background about the victim did not significantly predict negative reactions toward the perpetrator. Positive background information about the perpetrator was also not a significant predictor of reactions to the perpetrator. However, learning about the perpetrator’s traumatic background predicted significantly more positive reactions toward him. In addition, ethnicity, dispositional empathy scores, empathy toward the perpetrator, and empathy toward the victim remained significant predictors of negative reactions toward the perpetrator.

Table 4.

Predictors of Negative Responses Toward the Perpetrator: OLS Regression (N = 834).

	Model 1		Model 2		Model 3		Model 4		Model 5
Predictor Variable	b	β	b	β	b	β	b	β	b	β
Gender^a	0.12	0.08	0.07	0.05	0.05	0.04	0.06	0.04	0.08	0.05
	(2.34)*		(1.26)		(1.05)		(1.13)		(1.71)
Race^b
Asian American	−0.24	−0.18	−0.23	−0.18	−0.22	−0.16	−0.21	−0.16	−0.15	−0.12
	(−4.13)**		(−3.99)**		(−3.92)**		(−3.95)**		(−2.93)**
Latino	−0.18	−0.10	−0.18	−0.10	−0.17	−0.09	−0.16	−0.09	−0.16	−0.09
	(−2.30)*		(−2.32)*		(−2.38)*		(−2.35)*		(−2.42)*
Other	−0.28	−0.13	−0.28	−0.13	−0.27	−0.13	−0.25	−0.12	−0.22	−0.11
	(−3.34)**		(−3.30)**		(−3.38)**		(−3.26)**		(−3.01)**
Dispositional empathy score			−0.21	−0.13	−0.13	−0.08	−0.13	−0.08	−0.20	−0.12
			(−3.59)**		(−2.31)*		(−2.27)*		(−3.71)**
Empathy—victim					0.18	0.18	0.16	0.17	0.10	0.10
					(5.17)**		(4.88)**		(2.99)**
Empathy—perpetrator					−0.32	−0.34	−0.31	−0.33	−0.29	−0.31
					(−9.88)**		(−9.75)**		(−9.41)**
Victim^c
Positive background							−0.06	−0.05	−0.06	−0.05
							(−1.52)		(−1.55)
Perpetrator^c
Positive background							−0.01	−0.01	−0.02	−0.02
							(−0.23)		(−0.48)
Traumatic background							−0.32	−0.23	−0.29	−0.20
							(−6.34)**		(−5.74)**
Anger									0.16	0.26
									(8.21)**
Makes sense									−0.06	−0.10
									(−3.12)**
F(df )	5.99		7.43		20.76		20.97		25.56
	(4, 830)**		(5, 829)**		(7, 827)**		(10, 824)**		(12, 822)**
Adjusted R²	0.023		0.037		0.142		0.193		0.261

Note. OLS = ordinary least squares.

Male = 1, Female = 2.

Reference group for race is Caucasian.

Reference group is basic background information.

p < .05. **p < .01.

In the final model, whether or not the participant was able to make sense of the shooting (i.e., meaning making) was added into the analysis. The degree to which the shooting made sense significantly predicted less negative reactions toward the perpetrator. The perpetrator’s traumatic background, empathy felt toward the victim and perpetrator, dispositional empathy, and ethnicity all remained significant predictors in this final model. The final model explained 26.1% of the variance in reactions toward the perpetrator.

Discussion

In examining the extent to which empathy for victims and perpetrators can be fostered, our hypotheses were partially supported. Our prediction that dispositional empathy would predict the amount of empathy felt for both victims and perpetrators across conditions was supported. In addition to dispositional empathy levels, knowing additional information about a perpetrator (information that fosters feelings of similarity and perspective taking) increased empathy toward him. However, knowing more information about the victim did not lead to increases in empathy (participants reported high levels of empathy for the victim in both conditions). Our hypothesis that empathy for the victim would predict both positive community responses and negative responses to the perpetrator (increased anger and blame) was supported. Finally, our hypothesis that empathy for the perpetrator and the ability to make sense of the event would predict positive responses toward the perpetrator (i.e., forgiveness) was also supported. In addition, knowing traumatic background information about the perpetrator predicted less blame and anger toward him.

Implications for Community Coping

Although our findings are preliminary and the effect sizes were small (Cohen, 1988), if replicated they offer a number of important practical implications for communities coping after a violent tragedy. Our results suggest that providing details about the background of perpetrators of violence may increase empathy toward them. In this study, making sense of the violent event was also associated with decreases in blame and anger toward the perpetrator. Although more research is needed, it is possible that promoting empathy—for both victims and offenders—after violence might be an innovative way to reduce anger, while promoting meaning making and coping. Finding ways to reduce trauma after large-scale violence is an important goal for future research.

Implications for News Media

The findings of this study also have important implications for the way that violent tragedies are presented in the media. News stories often give out little factual information about the perpetrator of violence or the victims, instead sensationalizing violence, linking it with factors such as mental illness (“Help them before they Kill,” New York Post, Benjamin, August 13, 2012), and focusing on trivial aspects of the event (“Aurora shooting suspect James Holmes ‘hot,’ some Twitter users say,” New York Daily News, Duerson, July 25, 2012). In a media analysis of articles published after the Columbine shootings, Muschert (2009) found that the news media coverage often focused on how schools no longer seem safe, perpetuating public fears that such an incident is likely to happen again.

A recent study by McGinty, Webster, and Barry (2013) used a large national sample to study the impact of information revealed by news stories about mass shootings on public attitudes toward gun policies and people with mental illness. They found that when participants read a story about a mass shooting committed by someone specifically with a serious mental illness, they were more likely to report negative attitudes toward people with mental illness compared with the control group. The way that stories are reported in the media, particularly the information that is revealed about the perpetrator of mass violence, appears to have an important impact on public attitudes and perceptions.

Rather than perpetuating fear of violent events, our findings indicate that after a violent tragedy, it might be useful to provide detailed accounts of both the victim’s and perpetrator’s stories to encourage perspective taking and meaning making. Not only did empathy for the perpetrator decrease negative feelings such as blame and anger, but it also predicted positive community responses. Media news stories that paint a detailed, accurate picture of a perpetrator to help the public understand why the event occurred may help promote positive adjustment and coping (Updegraff et al., 2008).

Implications for the Criminal Justice System

Our study found that empathy toward a perpetrator is associated with the amount of information known about his or her background—whether it is positive information or traumatic background information that helps the crime make sense. Providing information about a perpetrator’s past (i.e., mitigating life history information) may be important for criminal defense teams to get across to juries when making sentencing decisions because it may result in a more lenient sentence (Schroeder, Guin, Pogue, & Bordelon, 2006). Furthermore, as dispositional empathy levels predicted empathetic responses, dispositional empathy may be an important factor for lawyers to examine when picking potential jurors for violent criminal cases.

Our study also provides initial evidence that fostering empathy for an offender may promote healing for the victim. This is consistent with principles of restorative justice (Braithwaite, 1989) in which the focus is on repairing the harm caused by a crime rather than punishing the offender. By sitting down and meeting an offender, victims may come to have a better understanding of his or her story—fostering empathy—which can lead to healing (Strang et al., 2006). Our findings lend additional support for pursing restorative justice strategies for community healing after violent events.

Study Limitations

Despite its potentially important findings, several limitations of this study should nonetheless be noted. First, this was a scenario study involving students reading about a violent community event and responding online. Participants reported the likelihood of engaging in behaviors regarding the victim and perpetrator, which may or may not translate to actual behaviors after such an event occurs. Nonetheless, participants read about a shooting on a university campus that was based on a recent real case, with as many true details as possible incorporated into the research materials. Moreover, this was a very specific event in which the perpetrator died during a school shooting that took place on a college campus. Although mass shootings generate a tremendous amount of media attention and public concern (Bliss, Emshoff, Buck, & Cook, 2006), they are very rare events (Heilbrun, Dvoskin, & Heilbrun, 2009; Mulvey & Cauffman, 2001). The extent to which these findings apply to other forms of violent crime that take place in other contexts, and events in which the perpetrator does not die, needs further exploration. Finally, our study sample included undergraduate students who were predominately female and Asian American. Although undergraduate students are an appropriate sample for studying the impact of school shootings, the extent to which these findings apply to other campuses and other student populations needs further exploration.

Future Directions for Research

In this study, the race of the victim and perpetrator were not specified in the scenario. Considering the fact that race was a significant predictor of reactions to the perpetrator in this study, and there is evidence that perspective taking and blame may present itself differently cross-culturally (Morris & Peng, 1994; Takaku, Weiner, & Ohbuchi, 2001), this is an important factor to examine in future studies. In addition, varying the gender of the perpetrator may be an interesting future direction, as much of the previous literature on fostering empathy for perpetrators has focused on female offenders (Hurst & Foley, 2005; Plumm & Terrance, 2009). Although the scenarios in this study were meant to foster feelings of similarity and perspective taking, the degree to which participants felt similar to the victim or perpetrator or were able to take their perspective was not assessed. Future studies are needed to specifically examine the validity of the similarity and perspective-taking variables by measuring the degree to which participants identify with the victim and offender in the study and the degree to which they are able to take their perspective. It is possible that adding measures of similarity and perspective taking would explain additional variance in the model. It may also be important to assess the “closeness” of the observer to the event (e.g., if the violence occurred in one’s classroom versus one’s school versus one’s city), which could be assessed in future studies. In addition, making sense of the crime was only assessed with a single item. A more sophisticated measurement of the concept of meaning making would be helpful in future studies that examine the role of empathy in fostering meaning after crime. Another important step for research in this area is to move beyond scenario studies in the lab and conduct research in an applied research setting and to assess reactions of victims, perpetrators, and communities after a large-scale violent event occurs. For example, after a school shooting, does empathy for the perpetrator have an impact on post-traumatic stress symptoms in the community? Does having empathy for one’s perpetrator help victims find meaning after a crime occurs and promote healing? The role of empathy in both emotional responses and behaviors after a violent event occurs is relatively unexplored, but these findings indicate it may be an important factor in fostering forgiveness and helping victims and communities cope with violent events.

Conclusion

As school shootings and mass violence make headlines with increasing frequency, it is important to consider how best to care for individuals and communities that survive these events—both by decreasing feels of anger and increasing meaning-making and coping skills. The findings of this study indicate that empathy may be an important variable for helping survivors cope with these tragedies. Empathy can be fostered by providing information about both victims and perpetrators after violence. The potential value of fostering empathy in our criminal justice system, in community mental health, and in media coverage of violent events warrants further study.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Author Biographies

Jillian K. Peterson, PhD, is an assistant professor in the Criminology and Criminal Justice Department at Hamline University in St. Paul, Minnesota. She has a doctorate degree in psychology and social behavior from the University of California, Irvine. She specializes in psychopathology and violent crime. She has worked as a mitigation specialist and special investigator on death penalty cases in New York City and Chicago, a research coordinator at the University of Minnesota, and a trial consultant with the National Jury Project. She has published multiple articles related to risk assessment, violence prevention, and the relationship between mental illness and criminal behavior. In addition to teaching and research, she provides training and consultation on forensic psychology topics to practitioners in the criminal justice field.

Roxane Cohen Silver, PhD, is professor in the Department of Psychology and Social Behavior, the Department of Medicine, and the Program in Public Health at the University of California, Irvine. An international expert in the field of stress and coping, she has spent more than three decades studying acute and long-term psychological and physical reactions to stressful life experiences, including personal traumas such as physical disability, loss, and childhood sexual victimization, as well as larger collective events such as war, firestorms, school shootings, terrorist attacks, and other international community disasters. Her research has been funded by the National Science Foundation, the National Institute of Mental Health, the U.S. Public Health Service, and the U.S. Department of Homeland Security. She has served on numerous senior advisory committees and task forces for the U.S. Department of Homeland Security, providing advice to the department and its component agencies on the psychological impact of disasters and terrorism. She is also one of the founding directors of Psychology Beyond Borders, an international nonprofit organization. In 2007, she received the American Psychological Association Award for Distinguished Service to Psychological Science, and in 2010, she received the Public Advocacy Award from the International Society for Traumatic Stress Studies (for “outstanding and fundamental contributions to advancing social understanding of trauma”). In 2011, she received APA’s Award for Distinguished Contributions to Psychology in the Public Interest (Senior Career) and the Award for Outstanding Service to the Field of Trauma Psychology from APA’s Division 56.

References

Anastasio

P. A.

Costa

D. M.

(2004). Twice hurt: How newspaper coverage may reduce empathy and engender blame for female victims of crime. Sex Roles, 51, 535-542. doi:10.1007/s11199-004-5463-7

Archer

R. L.

Diaz-Loving

Gollwitzer

P. M.

Davis

M. H.

Foushee

H. C.

(1981). The role of dispositional empathy and social evaluation in the empathetic mediation of helping. Journal of Personality and Social Psychology, 40, 786-796. doi:10.1037/0022-3514.40.4.786

Archer

R. L.

Foushee

H. C.

Davis

M. H.

Aderman

(1979). Emotional empathy in a courtroom simulation: A person-situation interaction. Journal of Applied Social Psychology, 9, 275-291. doi:10.1111/j.1559-1816.1979.tb02711.x

Benjamin

(2012, August 13). Help them before they kill. New York Post.

Bliss

M. J.

Emshoff

Buck

C. A.

Cook

S. L.

(2006). Patients' perceptions of causes of and solutions for school violence: Implications for policy. The Journal of Primary Prevention, 27, 265-280. doi:10.1007/s10935-006-0032-1

Braithwaite

(1989). Crime, shame and reintegration. Cambridge, UK: Cambridge University Press.

Brown

R. P.

Wohl

M. J. A.

Exline

J. J.

(2008). Taking up offenses: Secondhand forgiveness and group identification. Personality and Social Psychology Bulletin, 34, 1406-1419. doi:10.1177/0146167208321538

Cohen

(1988). Statistical power analysis for the behavioral sciences (2nd ed.). Hillsdale, NJ: Lawrence Erlbaum.

Davis

J. H.

Bray

R. M.

Holt

(1977). The empirical study of decision processes in juries: A critical review. In Tapp

Levine

(Eds.), Law, justice and the individual in society (pp. 326-361). New York, NY: Holt.

10.

Davis

M. H.

(1980). Multidimensional approach to individual differences in empathy. JSAS Catalog of Selected Documents in Psychology, 10, 85.

11.

Davis

M. H.

(1996). Empathy: A social psychological approach. Boulder, CO: Westview Press.

12.

de Waal

F. B. M.

(2008). Putting the altruism back into altruism: The evolution of empathy. Annual Review of Psychology, 59, 279-300. doi:10.1146/annurev.psych.59.103006.093625

13.

Duerson

M. H.

(2012, July 25). Aurora shooting suspect James Holmes ‘hot,’ some Twitter users say. New York Daily News.

14.

Griffitt

Jackson

(1973). Simulated jury decisions: The influence of jury-defendant attitude similarity-dissimilarity. Social Behavior and Personality, 1, 1-7. doi:10.2224/sbp.1973.1.1.1

15.

Grubb

A. R.

Harrower

(2009). Understanding attribution of blame in cases of rape: An analysis of participant gender, type of rape and perceived similarity to the victim. Journal of Sexual Aggression, 15, 63-81. doi:10.1080/13552600802641649

16.

Haegerich

T. M.

Bottoms

B. L.

(2000). Empathy and jurors’ decisions in patricide trials involving child sexual assault allegations. Law and Human Behavior, 24, 421-448. doi:10.1023/A:1005592213294

17.

Hawkins

N. A.

McIntosh

D. N.

Silver

R. C.

Holman

A. E.

(2004). Early responses to school violence: A qualitative analysis of students’ and parents’ immediate reactions to the shootings at Columbine High School. Journal of Emotional Abuse, 4, 197-222. doi:10.1300/J135v04n03_12

18.

Heilbrun

Dvoskin

Heilbrun

(2009). Toward preventing future tragedies: Mass killings on college campuses, public health, and threat/risk assessment. Psychological Injury and Law, 2, 93-99. doi:10.1007/s12207-009-9040-9

19.

Hodges

S. D.

Kiel

K. J.

Kramer

A. D. I.

Veach

Villanueva

B. R.

(2010). Giving birth to empathy: The effects of similar experiences on empathetic accuracy, empathetic concern, and perceived empathy. Personality and Social Psychology Bulletin, 36, 398-409. doi:10.1177/0146167210375618

20.

Hurst

D. R.

Foley

L. A.

(2005). Filicide and insanity defense: Legal authoritarianism and empathy as predictors of guilt. American Journal of Forensic Psychology, 23, 81-91. Available from http://www.forensicpsychology.org

21.

Janoff-Bulman

(1992). Shattered assumptions: Towards a new psychology of trauma. New York, NY: Free Press.

22.

Johnson

J. D.

Simmons

C. H.

Jordan

MacLean

Taddai

Thomas

. . . Reed

(2002). Rodney King and O.J. revisited: The impact of race and defendant empathy induction on judicial decisions. Journal of Applied Social Psychology, 32, 1208-1223. doi:10.1111/j.1559-1816.2002.tb01432.x

23.

Jolliffee

Farrington

D. P.

(2006). Development and validation of the Basic Empathy Scale. Journal of Adolescence, 29, 589-611. doi:10.1016/j.adolescence.2005.08.010

24.

Kerr

N. L.

Hymes

R. W.

Anderson

A. B.

Weathers

J. E.

(1995). Defendant-juror similarity and mock juror judgments. Law and Human Behavior, 19, 545-567. doi:10.1007/BF01499374

25.

Marques

J. M.

(1990). The black sheep effect: Outgroup homogeneity in social comparison settings. In Abrams

Hoggs

(Eds.), Social identity theory: Constructive and critical advances (pp. 131-151). London, England: Harvester Wheatsheaf.

26.

McGinty

E. E.

Webster

D. W.

Barry

C. L.

(2013). Effects of news media messages about mass shootings on attitudes toward persons with serious mental illness and public support of gun control policies. American Journal of Psychiatry, 170, 494-501. doi:10.1176/appi.ajp.2013.12010014

27.

Morris

M. W.

Peng

(1994). Culture and cause: American and Chinese attributions for social and physical events. Journal of Personality and Social Psychology, 67, 949-971. doi:10.1037/0022-3514.67.6.949

28.

Mulvey

E. P.

Cauffman

(2001). The inherent limits of predicting school violence. American Psychologist, 56, 797-802. doi:10.1037//0003-066X.56.10.797

29.

Muschert

G. W.

(2009). Frame-changing in the media coverage of a school shooting: The rise of Columbine as a national concern. The Social Science Journal, 46, 164-170. doi:10.1016/j.soscij.2008.12.014

30.

Park

C. L.

Folkman

(1997). Meaning in the context of stress and coping. Review of General Psychology, 1, 115-144. doi:10.1037/1089-2680.1.2.115

31.

Plumm

K. M.

Terrance

C. A.

(2009). Battered women who kill: The impact of expert testimony and empathy induction in the courtroom. Violence Against Women, 15, 186-205. doi: 10.1177/1077801208329145

32.

Sargeant

(2010). Essentials of donor loyalty. London, England: White Lion Press.

33.

Schroeder

Guin

C. C.

Pogue

Bordelon

(2006). Mitigating circumstances in death penalty decisions: Using evidence-based research to inform social work practice in capital cases. Social Work, 51, 355-364. doi:10.1093/sw/51.4.355

34.

Silver

R. C.

Updegraff

J. A.

(2013). Searching for and finding meaning following personal and collective traumas. In Markman

Proulx

Lindberg

(Eds.), The psychology of meaning (pp. 237-255). Washington, DC: American Psychological Association.

35.

Stein

M. L.

Miller

A. K.

(2012). Distress resulting from perceivers’ own intimate partner violence experiences predicts culpability attributions toward a battered woman on trial for killing her abuser: A path model. Journal of Interpersonal Violence, 27, 2527-2544. doi:10.1177/0886260512436388

36.

Stephan

C. W.

Stephan

W. G.

(1986). Habla Ingles? The effect of language translation on simulated juror decisions. Journal of Applied Social Psychology, 16, 577-589. doi:10.1111/j.1559-1816.1986.tb01160.x

37.

Strang

Sherman

Angel

C. M.

Woods

D. J.

Bennett

Newbury-Birch

Inkpen

(2006). Victim evaluations of face-to-face restorative justice experiences: A quasi-experimental analysis. Journal of Social Issues, 62(2), 281-306. doi: 10.1111/j.1540-4560.2006.00451.x

38.

Takaku

Weiner

Ohbuchi

(2001). A cross-cultural examination of the effects of apology and perspective-taking on forgiveness. Journal of Language and Social Psychology, 20, 144-166. doi:10.1177/0261927X01020001007

39.

Updegraff

J. A.

Silver

R. C.

Holman

E. A.

(2008). Searching for and finding meaning in collective trauma: Results from a national longitudinal study of the 9/11 terrorist attacks. Journal of Personality and Social Psychology, 95, 709-722. doi:10.1037/0022-3514.95.3.709

40.

Weir

J. A.

Wrightsman

L. S.

(1990). The determinants of mock jurors’ verdicts in a rape case. Journal of Applied Social Psychology, 20, 901-919. doi:10.1111/j.1559-1816.1990.tb01467.x

41.

Westmaas

J. L.

Silver

R. C.

(2006). The role of perceived similarity in supportive responses to victims of negative life events. Personality and Social Psychology Bulletin, 32, 1537-1546. doi:10.1177/0146167206291874

42.

Wicke

Silver

R. C.

(2009). A community responds to collective trauma: An ecological analysis of the James Byrd murder in Jasper, Texas. American Journal of Community Psychology, 44, 233-248. doi:10.1007/s10464-009-9262-8