The Impact of Angry Versus Sad Victim Impact Statements on Mock Jurors’ Sentencing Decisions in a Capital Trial

Abstract

The present study tested the effects of angry and sad victim impact statements (VIS) on jury eligible participants’ decisions. Death qualified participants (N = 581) watched the penalty phase of a capital trial that varied the presence and emotional content of the VIS (angry, sad, or no VIS) along with the strength of mitigating evidence (weak or strong). Results revealed that Angry VIS led to an increase in death sentences, whereas Sad VIS did not. Furthermore, participants who reported becoming angry during the trial were more likely to render a death sentence, but participants who became sad during the trial were not. No interaction was found between VIS and strength of mitigating evidence, but participants exposed to the angry VIS did rate the mitigating evidence as less important to their decisions. The results indicate that VIS are not inherently biasing, nor are all emotions equally impactful on sentencing decisions.

Keywords

emotions and decision making juror decision making victim impact statements

The Impact of Angry versus Sad Victim Impact Statements (VIS) on Sentencing Decisions in a Capital Trial

VIS are statements, either written or oral, that detail the impact of the defendant’s crime on the victim or the victim’s surviving family. Typically, these statements indicate financial losses, physical injuries, and describe changes to the victim’s personal welfare (Booth v. Maryland, 1987). VIS have become increasingly common since the Supreme Court held that such statements were constitutional (Payne v. Tennessee, 1991), and recent legislative changes such as the 2004 Crime Victims’ Rights Act guaranteed crime victims the right to participate in virtually all public criminal proceedings in federal courts (Kyle, Twist, & Higgins, 2005). Twenty-six of the 31 states that enforce the death penalty allow for VIS during the penalty phase of the trial (Death Penalty Information Center, 2015).

Critics of VIS suggest that testimony surrounding the character of the deceased could lead to sentencing decisions based on qualities of the victim rather than the defendant (Greene, 1999; Myers & Greene, 2004), and could introduce testimony that would be nearly impossible for the defense to rebut (Logan, 1999). Although the U.S. Supreme Court has addressed these issues, most of the legal debate surrounding VIS has centered on whether VIS are relevant in establishing the blameworthiness of the defendant and whether the inflammatory nature of VIS creates an unreasonable danger that sentencing decisions will be arbitrary and capricious rather than the product of reasoned judgment (Myers & Greene, 2004).

One of the primary concerns with the legal relevance of VIS is the potential for such testimony to stir negative emotions that will interfere with the reasoning abilities of jurors. A long-held belief in the legal system is that an emotional juror is an irrational juror (Feigenson, 2000; Solomon, 1990). By this logic, decisions arising from emotional states lack reasoned analysis and are therefore prejudicial. As Blumenthal (2001) has noted, the U.S. Supreme Court called specific attention to the potential emotional bias arising from VIS in capital sentencing. In Booth v. Maryland (1987), the Court noted that “the formal presentation of the information by the State can serve no other purpose than to inflame the jury and divert it from deciding the case on the relevant evidence concerning the crime and the defendant” (p. 508). Moreover, numerous instances where VIS have elicited strong emotional displays in jurors and even judges are frequently depicted in the news (see, for example, Logan, 1999; Myers & Greene, 2004).

However, scholars from many fields, including philosophy (e.g., DeSousa, 1987) and neuroscience (e.g., Damasio, 1994), have increasingly argued that emotions are not necessarily antithetical to rational judgment. Damasio (1994) famously proposed that emotion is actually necessary for rationality, as it provides “gut-level” intuitions that guide us toward the most advantageous choice. Furthermore, social cognition research indicates that some negative emotions, like sadness, may actually sustain or even enhance systematic information processing (e.g., Bless, Bohner, Schwarz, & Strack, 1990). In comparison, it has been suggested that anger can lead to more blame and punitive judgments and may also undermine systematic information processing (Lerner & Tiedens, 2006; Tetlock et al., 2007).

One context in which to examine the effects of emotions on legal decision making is in capital trials where jurors have been exposed to VIS. Here, jurors are confronted with often highly emotional testimony just prior to rendering a judgment when the stakes are highest—determining whether a defendant will live or die (Logan, 1999). As we argue next, VIS are perhaps best suited to examine the differential effects of anger and sadness because these two emotions are routinely aroused as a result of this testimony.

Emotions Aroused at Trial

There is anecdotal evidence from real trials that VIS frequently elicit a great amount of sadness in jurors. For example, the trial of Timothy McVeigh included testimony by the mother of a 4-year-old who died in the explosion. She described to the jury a phone conversation she had with a medical examiner who called to say they had uncovered her daughter’s hand in the debris, and wanted to know if she wished to keep it. She responded, “Of course I want it. It’s part of her.” Reportedly, not only did jurors cry upon hearing this story, but so did courtroom reporters and the judge (Logan, 1999).

In addition to anecdotal evidence of emotional responses at trial, Nuñez, Egan-Wright, Kehn, and Myers (2011) conducted a field study that assessed the emotionality of VIS. They obtained 195 VIS from 131 capital trials and analyzed the emotionality of the statements using the Linguistic Inquiry and Word Count (LIWC; Pennebaker, Francis, & Booth, 2001) program. This program produces indexes of anger and sadness expressed by a speaker (or writer) using empirically validated word count algorithms. They found that the predominant negative emotion in the VIS was sadness, lending further credence that sadness is a common emotion experienced when listening to VIS. For example, in one VIS taken from Commonwealth v. Kennedy (2003), the wife of the deceased tells the court what it is like living without her spouse:

How do I begin to tell you how this murder has changed my life forever. How do I express the wave of emotions and feelings I go through each day as I miss being with and growing old with my one and only true love, my soul mate, Michael. It feels as if I have been cut in half. A big part of my life is now gone, missing. (p. 22)

In another report from the same VIS field study, Myers, Nuñez, Mitchell, Kehn, and Wilkowski (2017) found that there was wide variation in VIS testimony in different court jurisdictions. Some courts limit unbridled victim impact testimony by having an attorney question the VIS witness. However, in some jurisdictions, victims are given wide latitude in expressing their opinions, and some of that testimony can be quite angry. For example, in the case of Texas v. Barney Fuller (2004), the victim’s sister said the following during her VIS testimony:

Mr. Fuller, you killed the wrong bitch. The worst bitch in the family is still alive right here and you’re looking at her today. I will be in your life for what time you have left on this earth. I will be here to haunt you until you die like the two that you murdered. When the needle is inserted in your body to kill you, I will be the one last thing that you will think of as that juice is entering in your blood system. Then I will attend your funeral to help your family celebrate.

Given that legal decision makers might be exposed to angry family members, it is important to understand how angry VIS affect decisions as well.

To summarize, although VIS may elicit a variety of emotions from jurors, two distinct emotions such as anger and sadness occur with regularity, and there is reason to believe these two distinct emotions may prompt different judgments on the part of jurors. The purpose of the present article was to examine how the emotions that result from hearing VIS testimony might affect juror information processing and sentencing decisions in capital cases. We were specifically interested in the possibility that the influence of VIS on sentencing decisions depended on whether or not the testimony made jurors feel sad or angry.

Anger, Sadness, and Decision Making

While much has been written of the potential for VIS to be inflammatory (see Blumenthal, 2001; Logan, 1999; Myers & Greene, 2004), little distinction has been made regarding the impact different emotions may have on juror decision making. However, a growing body of research in the area of social cognition suggests that specific negative emotions can exert different influences on information processing and decision making (Lerner & Tiedens, 2006). Most relevant to the potential impact of VIS is the finding that anger leads to greater punitiveness, but sadness does not.

Several different social cognitive theories converge on this same conclusion. According to appraisal-tendencies theory (Lerner & Keltner, 2000; Lerner & Tiedens, 2006), each discrete emotion is linked with a distinctive appraisal pattern in a recursive relationship. The emotion of anger, for example, is often linked to the appraisal of other-blame (i.e., the judgment that others caused harm to one’s self or people who are close to the self; for example, Smith, Haynes, Lazarus, & Pope, 1993; Weiner, 1986).

In a related theoretical approach, Tetlock and colleagues (Lerner, Goldberg, & Tetlock, 1998; Rucker, Polifroni, Tetlock, & Scott, 2004; Tetlock et al., 2007) have argued that angering incidents activate a “prosecutorial mindset,” whereby a person becomes motivated to blame others for wrongdoing and ensure that they are appropriately punished. This punitive mind-set can carry over to entirely unrelated incidents if the initial offender was not found or punished (Goldberg, Lerner, & Tetlock, 1999).

Other negative emotions are seen as having quite different consequences for judgment. Unlike anger, sadness is typically associated with the appraisal of loss or hopelessness (e.g., Smith et al., 1993; Weiner, 1986). Because of this, sad individuals often feel powerless and uncertain in the face of an overpowering and unpredictable situation. As a result, appraisal-tendencies theory suggests that sadness leads people to focus more on the situational rather than dispositional factors in determining causality (i.e., less blame oriented; Lerner & Keltner, 2000).

Basic social-psychological research has found a number of results consistent with these accounts. For example, numerous studies indicate that participants are more likely to attribute causal responsibility to, blame, and punish an offender when the emotion of anger has been experimentally induced (e.g., Lerner et al., 1998; Quigley & Tedeschi, 1996; Tiedens & Linton, 2001). Goldberg et al. (1999) found that the induction of anger led participants to blame and punish a negligent defendant more in a tort case. By contrast, sadness leads people to explain events in terms of situational causes (e.g., Keltner, Ellsworth, & Edwards, 1993). This can even lead people to behave in a more benevolent fashion in many circumstances (e.g., providing more assistance to people in need; Small & Lerner, 2008).

VIS, Juror Judgments, and the Effect of Specific Emotions

A number of empirical studies have suggested that hearing VIS can have an effect on juror decision making. For example, Luginbuhl and Burkhead (1995) found that participants were significantly more likely to vote for the death penalty when a VIS was present (51%) than when it was absent (20%). Similarly, Myers and Arbuthnot (1999) found that mock jurors who initially voted to convict were significantly more likely to render death penalty judgments when they viewed a VIS (67%) than when they did not (30%).

Other studies have found more modest effects of VIS on sentencing (e.g., Myers, Godwin, Latter, & Winstanley, 2004; Myers, Roop, Kalnen, & Kehn, 2013). In instances in which sentencing judgments were not measured, researchers have found VIS significantly affects judgments that are relevant to sentencing. For example, Greene (1999) and Greene, Koehring, and Quiat (1998) both found that VIS affected liking of and compassion for the victims. But, other studies have failed to uncover evidence for the impact of VIS. Gordon and Brodsky (2007) found that only 33% of the participants found VIS to be “moderately” influential on their judgments. Butler (2008) found that in a sample of 200 venire-persons given a summary of a capital case including the penalty phase, the presence of VIS did not lead to significantly more death penalty judgments. More recent studies by Boppre and Miller (2015) and Wevodau, Cramer, Kehn, and Clark (2014) did not find that VIS affected sentencing. The inconsistencies in findings suggest that other, moderating influences are at work. Drawing on basic social cognition literature, we suggest that the evocation of specific emotions via VIS is critical to obtaining an influence on sentencing decisions.

Specifically, there are indications that the inconsistent effects of VIS can be explained by the presence of angering testimony. For example, two studies (Myers, Lynn, & Arbuthnot, 2002; Platania & Berman, 2006) found that a crying witness providing a VIS failed to produce significant differences in sentencing recommendations, compared with either a stoic VIS or no VIS control condition. While neither study contained a manipulation check of sadness specifically, the null effects of a crying witness suggest sadness has little effect on sentencing decisions. In the study by Boppre and Miller (2015), they similarly found that VIS testimony regarding the “personal and traumatic effects of her mother’s death” (p. 422) had no significant effect on sentencing decisions. Here again, sadness specifically was not assessed, but rather, overall negative affectivity. While this represents conjecture on our part, one possibility is that learning of the effects of the mother’s death mainly evoked sadness rather than anger. Similarly, the study by Wevodau et al. (2014), which evoked general negative affect in response to VIS, failed to affect juror decisions, and so a similar result may have occurred in this study as well.

However, when we see evidence of strong VIS effects on sentencing, evidence of increased anger is typically present. ForsterLee, Fox, ForsterLee, and Ho (2004) provided relatively indirect evidence along these lines. Under conditions that elicited more anger (i.e., VIS for a female defendant), death penalty recommendations were more common. Paternoster and Deise (2011) found that jurors who viewed a VIS rated themselves as more angry, hostile, and vengeful toward the defendant and as having more sympathy for the victims. Furthermore, participants who viewed the VIS were 4 times more likely to render death sentences, and both anger and sympathy mediated such effects. This is an unusually strong effect size in the VIS literature, and so the link between the anger the case evoked and the strong effect on sentencing is worth highlighting. Interestingly, the authors reported that the VIS they used did not evoke more sadness compared with the no VIS group.

Other studies provide more direct, but still correlational, evidence that testimony in the penalty phase that evokes anger is related to mock juror decisions. In a recent study that did not specifically examine the effects of VIS but nevertheless examined emotions and capital sentencing, Georges, Wiener, and Keller (2013) found that anger, disgust, and frustration increased early in the trial and remained above baseline throughout the trial. However, only changes in anger predicted an increased likelihood of death sentence recommendations. In a related study, Nuñez, Schweitzer, Chai, and Myers (2015) had mock jurors rate their emotions before and after witnessing the sentencing phase of a capital trial using a portion of the PANAS-X, a measure of negative and positive affect (Watson & Clark, 1994). While this study did not specifically vary VIS, they found that mock jurors who witnessed VIS testimony reported increases in anger and sadness after watching the trial, but it was only increases in anger (not sadness) that predicted death sentences. Importantly, these studies relied on correlational evidence alone, so the causal effect of anger on sentencing decisions remains unclear.

VIS, Anger, and Reasoned Judgment

As we indicated earlier, one of the critical concerns surrounding the introduction of VIS in capital sentencing is that the emotions they evoke may serve to interfere with reasoned decision making (see, for example, Blumenthal, 2001; Myers & Greene, 2004; Paternoster & Deise, 2011). This concern that VIS interfere with rational decisions was most explicitly expressed in Booth v. Maryland (1987), where the majority opinion referenced Gardner v. Florida (1977) in noting that decisions to impose death “must be, and appear to be, based on reason rather than caprice or emotion” (Booth v. Maryland, 1987, p. 508). The Booth v. Maryland (1987) judgment articulated the Court’s contention that emotions impede rational decisions and cannot play a role in life or death decisions: “The admission of these emotionally charged opinions as to what conclusions the jury should draw from the evidence clearly is inconsistent with the reasoned decision making we require in capital cases” (Booth v. Maryland, 1987, p. 508). In overruling Booth v. Maryland (1987) and South Carolina v. Gathers (1989), the U.S. Supreme Court in Payne v. Tennessee (1991) did not challenge the notion that sentencing decisions must reflect reasoned judgment. Instead, they noted that VIS are not per se inadmissible, and that information about the harm experienced by surviving relatives may be considered on a case-by-case basis. As Justice O’Connor noted,

Most states have enacted legislation enabling judges and juries to consider victim impact evidence . . . the possibility that this evidence may in some cases be unduly inflammatory does not justify a prophylactic, constitutionally based rule that this evidence may never be admitted. (Payne v. Tennessee, 1991, p. 2612)

Although the Court in Booth v. Maryland (1987) and Payne v. Tennessee (1991) failed to explicitly identify what information jurors should instead consider, previous decisions (e.g., Gregg v. Georgia, 1976) make it clear that a careful weighing of the aggravating and mitigating factors is a central responsibility of jurors during the penalty phase of a capital trial. In Lockett v. Ohio (1978), the primacy of mitigating factors in the penalty phase of capital trials was clearly articulated. Here, the Court noted that jurors must consider “any aspect of a defendant’s character or record and any of the circumstances of the offense that the defendant proffers as a basis for a sentence less than death” (p. 604). Evidence that generates strong emotions and serves to interfere with a juror’s capacity to consider mitigating evidence reflects a decision that is “arbitrary and capricious” and a potential violation of the defendant’s 8th Amendment rights under the cruel and unusual punishment clause according to Furman v. Georgia (1972).

To address this concern more directly, we examined the effect of angry and sad VIS on mock jurors’ abilities to take mitigating evidence into account. This could be achieved in more than one way. First, are jurors less attentive to variations in the strength of mitigating evidence depending on their particular emotional state? Second, are jurors more willing to see mitigating evidence as important to their sentencing judgment based on their particular emotional state? The basic social cognitive literature paints a mixed picture regarding both anger and sadness’s effects on rational judgment. While early studies found that anger increases reliance on heuristic cues (suggesting a lack of rational thought; Bodenhausen, Sheppard, & Kramer, 1994), later studies found that it can sometimes increase systematic thought (as evidenced by increased differentiation of a logically strong vs. weak argument; Moons & Mackie, 2007), especially when rational thought allows one to “move against” another person (Young, Tiedens, Jung, & Tsai, 2011). The picture is similarly mixed with sadness. While some studies indicate that it improves systematic thought (once again evidenced by increased differentiation of a strong vs. weak argument; Bless et al., 1990), other studies fail to find these effects (Nuñez et al., 2015). Evidence that jurors fail to process mitigating evidence in the presence of emotional VIS, or weigh the value of the mitigators differentially depending on the emotions elicited by the VIS, would offer support for the concern that (at least some) emotions elicited by VIS may interfere with reasoned decision making.

Present Study and Hypotheses

In sum, basic research and existing psycholegal research suggest that anger and sadness each have differential effects on judgments. If true, VIS will only be inflammatory if it elicits anger. Sadness should have little effect. The present study was designed to experimentally test this prediction. Participants watched the penalty phase of a death penalty case that either included anger-inducing VIS testimony, sadness-inducing VIS testimony, or no VIS testimony at all. To additionally test how VIS affect mock jurors’ abilities to take mitigating information into account, participants were also exposed to either strong or weak mitigating evidence. At the conclusion of the study, they were asked to make a sentencing decision (life or death) and rate the importance of aggravating and mitigating evidence. We also asked participants to rate their emotions before and after the trial to examine whether reported emotions of anger and sadness increased when exposed to VIS. We tested the following hypotheses:

Hypothesis 1: If a VIS itself leads to more punitive verdicts, then participants who see either the Angry or Sad VIS would give more death penalty verdicts compared with those in the No VIS condition.

In contrast to this standard account, we predicted that the Angry VIS but not the Sad VIS would lead to more death penalty verdicts, compared with participants in the No VIS condition.

Hypothesis 2: We sought to examine how the VIS affected the use of mitigating evidence in sentencing decisions. Given the mixed findings in the social cognitive literature, we tested two competing hypotheses:

Appraisal-tendencies theory (Lerner & Keltner, 2000) suggests that anger leads to a heuristic information processing style, while sadness leads to a systematic information processing style. This suggests that the sad VIS should lead participants to more strongly differentiate between strong and weak mitigating evidence, while angry VIS should lead participants make less of a differentiation.

Other research has found that anger motivates more systematic information processing when it allows one to effectively “move against” another person (Moons & Mackie, 2007; Young et al., 2011). This research would suggest that both the angry VIS and sad VIS would increase the differentiation between strong and weak mitigating evidence compared with the control condition of no VIS.

Hypothesis 3: Finally, we sought to test whether participants rated the same evidence differentially, depending on their emotional state. Specifically, angry participants are predicted to evaluate evidence differently than sad participants and in a manner consistent with their emotions and a desire to punish (Tetlock et al., 2007). If so, we would expect participants in the Angry VIS condition to sentence the defendant to death more often and to rate the prosecution’s evidence (i.e., aggravating factors) as being important to their decisions, relative to those in the Sad or No VIS conditions.

Method

Participants and Design

Five hundred eighty-one death qualified participants (219 male, 352 female) ranging in age from 18 to 66 years (M = 24.67, SD = 9.73) participated. Participants were recruited through the psychology research pools at two different universities (one in the central U.S. region and another in the southeast U.S. region) as well as Amazon’s Mechanical Turk (mTurk). University participants received partial course credit for their participation, and mTurk participants were paid US$8. MTurk participants watched the trial video and had to correctly answer three questions about the video to continue with the study. All of the mTurk sample reported in the present study successfully answered these questions. There were no differences in sentencing, χ²(2) = 2.22, p = .33, across the samples; however, the mTurk sample was older than either of the college samples (M_mTurk = 32.57, M_U2 = 20.23, M_U1 = 19.21) and the college samples had a greater percentage of female participants than the mTurk sample (M_mTurk = 54.4%, M_U2 = 64.5%, M_U1 = 67.1%). Given the lack of sentencing differences for the three groups, we collapsed across samples for the analyses. As discussed in greater detail below, all participants met legal criteria for inclusion on a death penalty jury (i.e., they were “death qualified” and were at least 18 years of age) before they were allowed to participate.

Participants were told that they would serve as mock jurors in the penalty phase of a capital murder trial. They were told the basic facts of the case and that the defendant was found guilty. One question asked of participants (after viewing the trial) was whether they agreed with the initial verdict (from 1 = completely disagree to 9 = completely agree). Ten participants who indicated that they did not agree with the verdict or were unsure about the verdict (i.e., they chose 1 through 5 for their answer) were removed from the study, leaving us with 571 participants (n_U1 = 138, n_U2 = 246, n_mTurk = 187). Participants were randomly assigned to one of two mitigating evidence conditions: strong or weak mitigating evidence; and one of three VIS conditions: No VIS (n = 188; 87 weak and 101 strong), Angry VIS (n = 185; 85 weak and 100 strong), or Sad VIS (n = 198; 83 weak and 115 strong).

Materials and Procedure

All participants completed the death qualification questionnaire, signed a consent form, viewed a reenactment of the sentencing phase of a capital murder trial, and completed a posttrial questionnaire. The entire procedure took approximately 1 hr.

Death Qualification Questionnaire

Potential participants completed a three-item death qualification questionnaire to qualify them to participate. These questions were consistent with the Wainwright v. Witt (1985) criteria. Participants who responded that their attitude toward the death penalty would seriously affect their ability to perform their juror duties, that they would not find the defendant guilty to avoid the possibility of giving the death sentence, or that they would either always or never give the death penalty were excluded from participating and considered not death qualified.

All three samples completed the death qualification questions prior to participating in the present study and only death qualified participants were included in the analyses. In two of the three samples (one university and mTurk), participants completed the questions as part of a separate prior study and only participants who were considered death qualified were allowed to participate in the present study. In the third sample, participants completed the death qualification questions immediately prior to completing the study and non–death qualified participants were screened out from the analyses reported here.

The Positive and Negative Affect Schedule–Revised (PANAS-X)

Participants completed the chosen subscales of the PANAS-X (Watson & Clark, 1994) both before and after watching the sentencing phase of a reenacted capital murder trial. Participants were given a list of 29 emotion words and were asked to rate their current level of each emotion on a 5-point Likert-type scale (1 = very slightly or not at all, 5 = extremely). Six words comprised the Anger subscale (angry, hostile, irritable, scornful, disgusted, and loathing), and five words comprised the Sadness subscale (sad, blue, downhearted, alone, and lonely). Scores on each subscale were summed, allowing Anger scores to range from 0 to 30 and Sadness scores to range from 0 to 25. As reported by Watson and Clark (1994), internal consistency for the subscales was .86 for Sadness, and .82 for Anger. Three other PANAS-X subscales were also administered (Joviality, Attentiveness, and Fear) to disguise our focus on anger and sadness. The Joviality subscale was comprised of eight words (happy, joyful, delighted, cheerful, excited, enthusiastic, lively, and energetic), the Attentiveness subscale was comprised of four words (alert, attentive, concentrating, and determined), and the Fear subscale contained six words (afraid, scared, frightened, nervous, jittery, and shaky).

Development of Trial Materials

All participants viewed a reenactment of the sentencing phase of a capital murder trial based off of actual trial transcripts. The trial and VIS chosen for the study were obtained from a field study of capital trials (Nuñez et al., 2011). As previously mentioned, in developing our trial materials for the present study, the sample of VIS transcripts obtained from the field study was analyzed for the emotional content contained in the VIS by a measure of linguistic content (i.e., LIWC; Pennebaker et al., 2001). In addition, a large sample of respondents also rated the emotional content of a subset of the VIS. The results from this field study by Nuñez et al. (2011) indicated that the predominant negative emotion in the VIS transcripts (as rated by both LIWC and people’s ratings) was sadness. Ratings of anger varied widely across transcripts, with some statements being rated very low in anger and some rated high. Interestingly, none of the transcripts contained high levels of anger without containing similar high levels of sadness (Nuñez et al., 2011).

To develop the VIS for the current study, we used portions of VIS testimony from transcripts that contained high levels of both anger and sadness (based on the LIWC content analysis). We then created several VIS transcripts and piloted them by having mTurk participants read the VIS testimony and rate the extent to which the VIS made them angry (1 = not at all angry to 9 = extremely angry) and the extent to which it made them sad (1 = not at all sad to 9 = extremely sad). The VIS chosen for the current study was rated by participants as being high in both anger (M_anger = 7.73) and sadness (M_sadness = 6.82). Consequently, in both linguistic content and individual ratings, the VIS was highly emotional.

To create the VIS conditions, the same VIS testimony was filmed so that the witness expressed two distinct demeanors (angry vs. sad). A theater major played the part of the victim’s wife. In the angry VIS condition, she presented the VIS using cues indicative of anger. She spoke in a short, clipped manner, and her face was pinched and tight when she spoke. In the sad VIS condition, she presented the exact same testimony, but this time she used cues of sadness. She choked up on the stand and was on the verge of crying throughout her testimony. See the appendix for a full transcript of the VIS testimony.

Trial Reenactment Procedures

The trial lasted approximately 45 min and was filmed in a mock courtroom used by the law school. Participants were first presented with summary information surrounding the guilt phase of the trial. Specifically, participants were told that the defendant, Anthony Prentice, shot and killed Jeff Mason while attempting to rob the local convenience store where the victim worked as a clerk. They were told that the defendant demanded the victim open the store safe. When Mr. Mason was unable to open the safe, the defendant repeatedly beat him on the head with the pistol and then shot him in the head. There was a witness in the store who saw the crime and identified the defendant. The defendant was found eight blocks from the crime scene, and his T-shirt was splattered with blood, which was later matched to the victim’s. The police found the gun used in the bushes near where the defendant was found, and fingerprints on the gun matched the defendant. The blood on the gun also matched the victim. Death qualified participants were informed that the jury found the defendant guilty of first-degree murder and that they would serve as mock jurors for the penalty phase of the trial.

The videotaped trial began with opening comments from the judge and the attorneys’ opening arguments. Participants then saw a parole officer present aggravating evidence stating that the defendant was on parole at the time he committed the crime and that he committed the crime while committing another felony, meeting the prosecution’s standard of proving beyond a reasonable doubt the presence of at least one aggravating factor. Participants were randomly assigned to receive the angry VIS, the sad VIS, or no VIS whatsoever. The participants assigned to view a VIS then saw the relevant testimony from the victim’s wife (described above).

For the defense, a clinical psychologist presented mitigating evidence, and provided explanations of the tests used in the assessment, along with explanations of error rates for the tests and interpretation of what the scores indicated when compared with normal intellectual functioning. In the strong mitigating evidence condition, the defendant’s IQ was said to be 69, the defendant had experienced extensive childhood sexual abuse, and he was previously diagnosed with schizophrenia. In the weak mitigating evidence condition, the defendant’s IQ was said to be 80, the defendant was physically abused as a child, and he was previously diagnosed with depression. These mitigators were chosen because previous research by Barnett, Brodsky, and Davis (2004) indicated that participants regarded intellectual deficits, childhood abuse, and mental illness as important mitigating factors.

The strength of mitigating evidence was also pilot tested with 20 mTurk participants. They were told that at capital trials the defense often provides expert testimony to convince a jury to show mercy by giving the defendant life rather than death. They were assigned to either the strong or weak mitigating evidence condition, were presented with each part of the mitigating evidence (i.e., regarding the defendant’s IQ, family background, and mental illness), and were asked to rate on a scale of 1 (not at all convinced) to 7 (totally convinced) whether that piece of evidence “convinced them that life in prison was the best sentence.” A repeated measures ANOVA with mitigating strength (Weak or Strong) as the independent variable (between subjects) and the ratings as the dependent variable (as the repeated variable) revealed a significant effect of mitigating evidence, F(1, 18) = 6.86, p = .02. The Weak mitigating condition was rated lower (M = 3.77, SD = 1.52) than the Strong mitigating condition (M = 5.53, SD = 1.52) indicating that participants were more convinced that life in prison was the more appropriate penalty in the Strong rather than the Weak mitigating condition.

After seeing the expert witness testify, participants saw the attorneys’ closing arguments and then received jury instructions from the judge.

Posttrial Questionnaire

After viewing the trial recording, participants were asked to sentence the defendant to either life in prison without the possibility of parole or death, as well as rate their confidence in their decision. Half of the participants completed the chosen subscales from the PANAS-X before the verdict and half completed them after the verdict. Participants then completed a series of questions regarding how the aggravating and mitigating evidence affected their sentence.

Rating of aggravating evidence importance

Two questions assessed the importance of the prosecution’s aggravating evidence. Participants were asked to rate how important each of the following was in reaching their sentences: (a) The defendant committed the crime while he was on parole for another felony. (b) The defendant committed the crime while in the commission of another felony. They rated the questions on a 1 (totally unimportant) to 9 (totally important) Likert-type scale. Answers on the two questions were highly correlated (r = .89), and an Aggravating Evidence Importance variable was constructed as the mean of the two questions.

Rating of mitigating evidence importance

Using the same scale as in the previous measures, three questions assessed the importance of defense evidence in decisions. Participants rated how important each of the following was in their sentencing decisions: (a) The defendant has a low IQ. (b) The defendant has mental health problems. (c) The defendant had a difficult childhood. Responses on the three items were highly related (α = .86), and a Mitigating Evidence Importance variable was created as the mean of the three items.

Results

Manipulation Checks

To examine whether participants’ sadness and anger increased after watching the trial, a series of paired sample t tests were conducted on the PANAS-X Sadness and Anger scores. Approximately, 10 participants did not complete some of the PANAS-X questions and could not be included in the analyses. Thus, degrees of freedom vary for these analyses.

Two repeated measures ANOVAS with Time (pretrial, posttrial) as the independent variable and Sadness or Anger as the dependent variable revealed that there were significant increases in sadness, F(1, 563) = 132.38, p < .001, d_z = .48, and anger, F(1, 561) = 268.41, p < .001, d_z = .69, after viewing the trial. See Table 1 for means and confidence intervals (CIs).

Table 1:

Changes in Anger and Sadness Before and After Trial Exposure: Descriptive Statistics

Emotion	M	SE	95% CI lower	95% CI upper
Sadness
Pretrial	6.67	.12	6.44	6.90
Posttrial	8.13	.14	7.85	8.40
Anger
Pretrial	6.95	.09	6.78	7.12
Posttrial	9.49	.17	9.16	9.82

Note. CI = confidence interval.

Next we examined whether VIS condition (No VIS, Sad, or Angry) predicted PANAS-X changes in anger and sadness scores from Time 1 to Time 2. We computed change scores for anger and sadness by subtracting the Time 1 PANAS-X score (pretrial) from the Time 2 PANAS-X score (posttrial; ΔAnger = PANAS-X_2Anger − PANAS-X_1Anger; ΔSad = PANAS-X_2Sad − PANAS-X_1Sad) and conducted two ANOVAS with VIS condition predicting change scores. With regard to sadness, the analysis did not yield a significant effect. As can be seen in Table 2, ΔSad scores did not vary according to VIS condition, F(2, 561) = .47, p = .62, η² = .002. All participants, regardless of condition, reported equal increases in sadness after watching the trial. ΔAnger scores also did not reach conventional levels of significance, F(2, 569) = 2.19, p = .06, η² = .008, one-tailed test. As expected, however, ΔAnger scores increased for all groups, but were greatest for the Angry VIS condition, compared with the Sad and No VIS conditions (see Table 2).

Table 2:

Change in Sad and Anger Scores by VIS Condition: Descriptive Statistics

Emotion	VIS condition	n	M	SE	95% CI lower	95% CI upper
ΔSad	No VIS	182	1.34	.24	0.87	1.81
	Sad VIS	197	1.42	.22	1.00	1.85
	Angry VIS	185	1.63	.23	1.18	2.08
ΔAnger	No VIS	181	2.08	.24	1.61	2.55
	Sad VIS	197	2.65	.27	2.12	3.18
	Angry VIS	184	2.86	.28	2.30	3.41

Note. CI = confidence interval; VIS = victim impact statements.

The second manipulation check was conducted to test whether mitigating evidence strength affected sentences, and we examined the No VIS condition to perform this comparison. A chi-square analysis revealed a significant effect of evidence strength on sentencing, χ²(1) = 3.53, p = .03, ϕ = .08. When mitigating evidence was weak, jury eligible participants were more likely to sentence the defendant to death, and when mitigating evidence was strong, they were more likely to choose a life sentence.

Hypothesis 1: The Effect of VIS on Sentencing Decisions

As previously stated, if a VIS itself leads to more punitive verdicts, then participants who see either the Angry or Sad VIS would give more death penalty verdicts compared with those in the No VIS condition. To determine the overall effect of VIS on sentencing decisions, we compared the No VIS with the Angry and Sad VIS conditions. In other words, we collapsed the two emotion VIS conditions into one condition called VIS. The logistic regression analyses revealed a significant effect of VIS on sentencing, χ²(1) = 4.29, p = .04, odds ratio (OR) = 1.45. As can be seen in Figure 1, participants were more likely to give the defendant a death sentence when VIS was present rather than absent. The OR indicates that the odds of receiving the death penalty if VIS are presented is 1.45 times greater than the No VIS condition (OR 95% CI = [1.02, 2.07]). If we were simply measuring the role of VIS on sentencing, we would conclude that VIS lead to more punitive sentencing.

Figure 1:

Proportion of the Sample That Selected Life and Death Sentences

Recall that we also predicted that the Angry VIS, but not the Sad VIS, would lead to more death penalty verdicts compared with participants in the No VIS condition. Thus, we examined the effect of No VIS, Sad VIS, and Angry VIS on sentencing (i.e., not collapsing across emotion type). A logistic regression analysis revealed a significant effect of VIS on sentencing, χ²(2) = 14.01, p < .001. Analysis of binomial proportions revealed that the Angry VIS differed significantly from the Sad VIS (Z = 3.15, p = .002) and the No VIS condition (Z = 3.41, p < .001), but there was not a significant difference in sentencing between the No VIS and Sad VIS conditions (Z = .30, p = .76; see Figure 1). The odds of receiving the death penalty in the Angry VIS condition were 2.02 times greater than the No VIS condition (OR 95% CI = [1.34, 3.05]). These results suggest that it is not VIS per se that affects verdicts, but rather VIS that arouse anger, which affect verdicts.

Hypothesis 2: The Role of Mitigating Evidence Strength on Sentencing

The second hypothesis examined how exposure to VIS affected participants’ ability to take mitigating evidence into account. A logistic regression was run with VIS (Angry, Sad, No VIS) and Mitigation Strength (Weak or Strong) as the independent variables and Sentence (Life or Death) as the dependent variable. The analysis did not reveal a significant VIS × Mitigation Strength interaction, χ²(2) = 2.10, p = .15, ϕ = .06. However, the pattern of results displayed in Figure 2 show that the participants in the Angry VIS condition were more sensitive to the Mitigation Strength than the participants in the Sad VIS condition. Removing the nonsignificant interaction, a significant effect for both emotionality of the VIS, χ²(2) = 9.17, p = .001, ϕ = .06, and Mitigation Strength, χ²(2) = 4.27, p = .002, ϕ = .09 (see Table 3), was found.

Figure 2:

Proportion of the Sample That Selected Life and Death Sentences Partitioned by the Strength of the Mitigating Evidence

Table 3:

Frequencies of Life and Death Sentences Varying VIS and Mitigation Strength

Condition	Life	Death
No VIS
Weak	46 (53%)	41 (47%)
Strong	67 (66%)	18 (34%)
Sad VIS
Weak	47 (57%)	36 (43%)
Strong	69 (60%)	46 (40%)
Angry VIS
Weak	28 (33%)	57 (67%)
Strong	51 (51%)	49 (49%)

Note. VIS = victim impact statements.

Hypothesis 3: VIS and Importance of Aggravating and Mitigating Evidence

To test whether participants rated the same evidence differentially, depending on their emotional state, an ANOVA was performed with VIS (No, Sad, or Angry) and Mitigation Strength as the independent variables and Aggravating Evidence Importance as the dependent variable. No significant effects were found for either the full model (with interaction terms) or the model that included only main effects. As can be seen in Table 4, the mean Prosecution Importance scores suggest that there might be a ceiling effect for Aggravating Evidence Importance scores. An examination of the variable revealed that 60% of our sample had a score of 8 or higher on this measure (where scores could range from 1 to 9) and 86% of the sample rated the importance of the prosecution evidence above a 5. Thus, the majority of our participants indicated that the prosecution evidence was important to their decisions.

Table 4:

Mean Ratings of Strength of Prosecution and Defense Evidence

Condition	Prosecution	Defense
Mitigation Strength
Weak	7.45 (.11)	4.94 (.12)
Strong	7.45 (.10)	5.75 (.11)
VIS
No VIS	7.39 (.13)	5.67 (.15)
Sad VIS	7.43 (.13)	5.27 (.14)
Angry VIS	7.50 (.13)	5.07 (.15)
Mitigation Strength by VIS
No VIS
Weak	7.48 (.19)	5.36 (.21)
Strong	7.30 (.18)	6.00 (.20)
Sad VIS
Weak	7.27 (.20)	4.85 (.22)
Strong	7.60 (.17)	5.72 (.19)
Angry VIS
Weak	7.56 (.19)	4.62(.22)
Strong	7.44 (.19)	5.53 (.20)

Note. Standard errors appear in the parentheses. VIS = victim impact statements.

To test whether ratings of Mitigating Evidence Importance varied by condition, an ANOVA was performed with VIS and Mitigation Strength as the independent variables and Mitigating Evidence Importance as the dependent variable. The analysis yielded significant main effects for VIS and Mitigation Strength, F(2, 565) = 4.38, p = .01, η² = .02; F(1, 565) = 23.01, p < .001, η² = .04, respectively. The interaction between VIS and Mitigation Strength was not significant. As can be seen in Table 4, participants rated the importance of the defense evidence higher when they were in the strong, rather than the weak, mitigating evidence condition (p = .04, d = .40). With regard to VIS, pairwise comparisons revealed that participants in the Angry VIS condition rated the importance of the defense evidence significantly lower than the No VIS (p = .05, d = .30) condition. The Sad VIS condition did not significantly differ from either the Angry or No VIS conditions (p > .05, d = .11 and .20, respectively; see Table 4).

Mediational Analysis

As previously mentioned, we found that anger and sadness increased after watching the trial, but only ΔAnger scores were related to VIS condition. However, another important consideration is whether either ΔAnger scores or ΔSadness scores affected sentencing decisions. Two logistic regressions were run with the ΔAnger scores and ΔSadness scores as the independent variables and sentence as the dependent variable. The analysis with ΔSadness scores was not significant (β = .02, Z = .57, p = .57). However, ΔAnger scores were associated with sentence (β = .10, Z = 3.02, p = .001, 95% CI = [.04, .16]). Increases in anger were associated with increases in death penalty decisions.

In our final test, we examined the possibility that changes in anger might mediate the relationship between Angry VIS and sentencing decisions. As previously reported, there was a significant association between VIS and Sentence, with participants in the Angry VIS condition more likely to recommend the death sentence (Direct effect: β = .62, 95% CI = [.20, 1.04]). ΔAnger scores were also found to be associated with sentence (β = .10, 95% CI = [.04, .16]) and VIS condition (β = .77, 95% CI = [.05, 1.50]). Thus, the primary conditions for mediation were met. Using the Hayes PROCESS macro for SPSS with 5,000 bias-corrected bootstrapped CIs at 95% confidence, the indirect effect was tested using the Angry versus No VIS conditions (coded as 1 and 0, respectively). Results indicated that the indirect effect of VIS condition on sentencing decision through ΔAnger scores was marginally significant (β = .07, 95% CI = [.01, .20]). Thus, the conditions for mediation were confirmed by the analyses, with ΔAnger scores mediating the significant effect of VIS and Sentence.

Discussion

We examined the differential impact of sadness- and anger-inducing VIS on decisions and found that when the trial included VIS, participants were more likely to give a death sentence than when the trial contained no VIS. However, when we analyzed the Sad and Angry VIS separately, only the Angry VIS led to an increase in death sentences; sentencing in the sad condition failed to differ from the no VIS condition. Moreover, changes in reported anger mediated the relationship between Angry VIS and sentencing. Consequently, the present findings provide confirmation that much of the inconsistencies in the literature surrounding the effects of VIS on sentencing may be attributable to the emotions evoked by the VIS. Anger promotes death penalty sentences, whereas sadness does not.

We also tested how VIS affected participants’ abilities to take mitigating evidence into account. Previous U.S. Supreme Court decisions (e.g., Gregg v. Georgia, 1976) make it clear that a careful weighing of the aggravating and mitigating factors is a central responsibility of jurors during the penalty phase of a capital trial, and if emotions affect how jurors weigh such evidence, this could be evidence of biased decision making. As mentioned earlier, if we found that jurors were less attentive to variations in the strength of mitigating evidence depending on their particular emotional state, this would be some evidence of arbitrary or capricious decision making. Overall, we found that participants were more likely to render a death sentence when mitigating evidence was weak, but the VIS condition did not moderate this effect. Though we did not discover a significant interaction between emotion and evidence strength, the pattern of results suggested that participants in the angry VIS condition were more sensitive than those in other conditions to the strength of mitigating evidence. Furthermore, though the interaction between VIS emotion and evidence strength did not reach conventional levels of significance, the possibility that angry participants weigh evidence more carefully needs to be further explored in future research. Further evidence of biased reasoning (i.e., capricious judgments) is demonstrated if jurors evaluate the importance of the mitigating evidence according to their particular emotional state. We found that angry VIS led participants to rate the mitigating evidence of lower importance to their sentencing decisions. This is consistent with intuitive prosecutor theory (Goldberg et al., 1999), but it is at odds with jurors’ central responsibility outlined by the courts to properly weigh mitigators in a reasoned fashion when sentencing in capital trials (e.g., Gregg v. Georgia, 1976).

Emotionality and Sentencing Decisions

We assessed participants’ reported anger and sadness before and after watching the trial video and found that both sadness and anger increased after watching the trial. However, only changes in anger were related to the VIS manipulations and sentencing decisions. Sadness was not. Our findings were quite similar to those reported by Georges et al. (2013) and Nuñez et al. (2015). In both studies, mock jurors viewed the sentencing phase of a capital trial and reported increases in sadness and anger during trial, but only anger affected sentences. Given that three studies have now found no effect of sadness on juror decision making, it might be tempting to regard sadness as a nonprejudicial aspect of a trial.

However, the results of our study suggest one other possibility. We found that sadness increased during the trial, but it increased equally in both VIS conditions. Because of this, we were not able to experimentally test its effects on decision making as there was no true nonsad control group. This may have been a problem in other legal studies that have examined sadness. Jury trials, especially capital trials, may elicit great sadness in decision makers. Thus, the effect sadness has on decision making might be difficult to test if every aspect of the trial (e.g., the details of the murder, the victim, the family) elicits sadness. It is also possible that sadness is inherent in all legal trials in which the defendant may lose their freedom and innocent victims may have suffered. Future research should examine whether there are some trial contexts that do not inherently elicit sadness, and then the emotion could be manipulated and tested.

VIS

The results of our study suggest that VIS, in and of themselves, may not lead to capricious decision making. Studies cited earlier that have found significant effects of VIS on decision making may have used VIS that elicited primarily anger in their participants. As we noted, much of the research on VIS and emotions have failed to separate out the negative emotions of anger and sadness or produce clear evidence of angry responses in jurors (e.g., Boppre & Miller, 2015; Myers et al., 2002; Platania & Berman, 2006; Wevodau et al., 2014). One exception is the study conducted by Paternoster and Deise (2011), who found that anger mediated the effects of a VIS on death sentences. Thus, while the effects of VIS on sentencing have been inconsistent, VIS that elicit anger consistently increase the likelihood that defendants will receive the death penalty.

Recall that the two models of emotion and judgment addressed here were the appraisal tendency model and the intuitive prosecutor model. The appraisal model suggests that emotions, such as anger, that prime certainty result in reduced (i.e., heuristic) information processing, whereas emotions, like sadness, that prime uncertainty should lead to more extensive processing. We should therefore expect to see evidence of greater processing in the sad participants relative to the angry ones (e.g., evidence that sad participants perceive greater differences between strong and weak mitigators). Conversely, the intuitive prosecutor model suggests that individuals perceive the evidence (such as mitigating and aggravating) as consonant with their particular emotion. So, when anger provokes a need to punish, the aggravators are judged as more important and the mitigators as less important. The current data suggest that anger motivates how evidence is judged, not how deeply the evidence is processed, consistent with the intuitive prosecutor model (Tetlock et al., 2007). This is supported by two findings. First, participants exposed to the angry VIS were not less likely to differentiate between the strong and weak mitigating evidence. Second, these participants rated the mitigating evidence as less important to their decisions. Thus, regardless of the strength of this evidence, they were motivated to discount it, as the intuitive prosecutor model would predict. Nonetheless, the evidence for this more specific conclusion remains tentative. Future research should include more definitive tests of a reasoning account (e.g., thought-listing protocols) to more clearly support this hypothesis.

Mitigating Evidence

In the present study, we specifically examined whether certain kinds of emotions evoked by a VIS (anger vs. sadness) might have a differential impact on how jurors consider mitigating evidence, and by extension, sentence the defendant. Consistent with predictions from intuitive prosecutor theory (Tetlock et al., 2007), the angry VIS led participants to judge the mitigating evidence as less important compared with participants who saw no VIS. This effect did not emerge for the participants who saw the sad VIS. Consequently, the angry VIS had a unique effect on ratings of the importance of the defense testimony. This finding is certainly troubling, but the extent to which it clearly points to an unwillingness to “consider” mitigating evidence is less apparent. As noted, differences in sentencing between weak and strong mitigators were still evident when participants were exposed to the angry VIS. This finding suggests that participants may have been processing the mitigation evidence to some degree.

Sentencing decisions are affected by a great many factors, and individual difference measures (such as death penalty attitudes) can potentially override any subtle effects variables such as VIS may have on mitigation evidence considerations. Indeed, research has shown that support for the death penalty affects not only the weight jurors give mitigating evidence, but also, in some instances, it has led jurors to judge mitigating factors as aggravating factors (e.g., Barnett, Brodsky, & Price, 2007; Garvey, 1998; Stevenson, Bottoms, & Diamond, 2010). By death qualifying the sample (a necessary approach when studying sentencing decisions), the present study contained those most likely to fail to use mitigation evidence properly in their sentencing decisions. Even if jurors “consider” mitigation evidence when sentencing, an angered and death qualified juror may assess the mitigating evidence in a different way than a sad or neutral juror.

Limitations

The findings here should be interpreted with caution as the present study was not without limitations. Although we included a number of factors designed to enhance the ecological validity of the study (e.g., diverse samples, videotaped trial, transcripts modified from actual capital trials, representative VIS, death qualification), there were aspects of the study that deviated in some important ways from what typically transpires in capital trials. For example, although we sought to extend the diversity of our sample by recruiting community members using mTurk, our sample still contained a large proportion of college students, which vary in a number of respects from actual juries such as age and heterogeneity of education and socioeconomic status (SES). While sample characteristics may matter less to decisions than many suppose (see, for example, Bornstein, 1999), seeking the sample most representative of the population of interest should always be the goal.

There were other methodological factors that readers should take note of when considering the generalizability of the findings. For example, jurors took part in only the penalty phase of the trial and were not exposed to the guilt phase of the trial but instead read a brief summary of the case facts. After the penalty phase, participants were not given the opportunity to deliberate as a group. To keep the length of the task to a manageable level for a single session, we chose to place our emphasis on the detail provided in the penalty phase (and including a detailed VIS). Although previous studies on VIS have both included the guilt phase along with the penalty phase (e.g., Mitchell, Myers, & Broszkiewicz, 2016) and have included group deliberation (e.g., Myers et al., 2002), the majority of the studies, thus far, have failed to include these features. In future studies, researcher might seek to address these factors and carefully examine the role of the guilt phase and the deliberation phase in sentencing by varying these factors.

The VIS used in the present study were exactly the same in the angry versus sad condition. It was only the demeanor of the actress that was varied. After extensively testing various VIS, we found that all of the VIS obtained in the Nuñez et al. (2011) field study were high in sadness. That may be the nature of VIS. We chose a VIS for this study that had high LIWC scores in both anger and sadness and had the actress portray one emotion over another. This is not a perfect way to examine emotional VIS experimentally, but it may be a more ecologically valid approach. All VIS (especially at murder trials) might be inherently sad. And it might be that sadness alone does not affect decisions. However, when anger is added to the mix, it may motivate jurors to punish the defendant more harshly.

Despite the fact that the Supreme Court has ruled that VIS are permissible, there have been lingering doubts among legal scholars about whether VIS could lead to capricious decision making. Our study (as well as previous research) suggests that VIS alone are not the problem. The problem appears when VIS communicate anger, which may arouse a motivation to punish in jurors. The good news, however, is that sadness does not seem to affect sentencing decisions. Given previous research that shows sadness is the most common negative emotion in actual VIS statements (Nuñez et al., 2011), this is good news indeed.

These results offer implications for the admissibility of VIS as well as some potential for future research in this area. As we noted earlier, Payne v. Tennessee (1991) argued that VIS is not per se inadmissible in capital sentencing. Instead, the Court reasoned that the harmful effects of VIS must be considered on a case-by-case basis. However, as Myers and Greene (2004) noted previously, the Court offered little guidance on what judges should consider when evaluating the potential harmful effects of VIS. The findings here suggest that statements that generate anger are those most likely to affect juror sentencing judgments. As the present findings indicate, VIS that promote anger influence jurors’ beliefs about the importance of the mitigators presented during the penalty phase of the trial. Future research might aim to identify more precisely the different elements within VIS testimony that are most likely to generate angry responses in jurors. For example, research on emotional contagion (e.g., Hatfield, Cacioppo, & Rapson, 1993) suggests that witnesses who display anger when testifying would be more likely to elicit a corresponding anger response in jurors. Moreover, while we focused our analysis on the impact of angry and sad VIS on mitigating evidence, future research activities could be aimed at better understanding the impact VIS has on the evaluation of aggravating factors when engaged in capital sentencing. Finally, there may be specific elements within the VIS content that judges may specifically look for when evaluating the admissibility of the VIS testimony, and this too is an area ripe for exploration by researchers.

The Payne v. Tennessee (1991) decision appears to have settled the issue of whether VIS are admissible, in general, in capital sentencing. Consequently, the findings presented in this study do little to inform the debate surrounding the constitutionality of introducing VIS in capital trials. Instead, the present findings shed needed light on the previous inconsistencies in findings regarding sentencing and VIS in the jury simulation literature and point to the importance of distinguishing among the types of emotions VIS provoke. Consistent with the social-psychological research on emotions and judgment, VIS that elicit anger promote a different mind-set in jurors than VIS that elicit sadness. Research that is directed at identifying what aspects of VIS testimony elicit anger in jurors has the potential to inform the courts. Significant advances in this area would better educate judges on the kinds of limits that should be placed on VIS testimony on a case-by-case basis in a manner in keeping with Payne v. Tennessee (1991). In terms of policy, it would be challenging to implement practices whereby sad witnesses may testify but angry witnesses may not. However, justices must regularly make decisions concerning whether the prejudicial aspects of testimony outweigh any probative value, and so VIS testimony that does little more than provide an angry witness the opportunity to unload his or her anger and frustrations might be identified and either postponed until the witness is more composed or disallowed altogether.

Conclusion

There is a growing body of evidence that anger leads to punitive decision making. This has been demonstrated in the social-psychological literature (e.g., Goldberg et al., 1999), the general psycholegal literature (e.g., Georges et al., 2013), and in studies that specifically examine the impact of VIS (e.g., the present study). Sadness, although less studied, does not appear to be associated with a particular decision style (at least in the jury context studied here). In the present study, jury eligible participants exposed to sad VIS made similar sentencing decisions as those who were not exposed to VIS. The present study therefore provides support for the notion that any study on the potential harmful effects of VIS must distinguish the particular emotion elicited in jurors. While the courts may regard emotions in general as antithetical to reasoned decision making, the empirical research on this matter suggests not all emotions are equally impactful on sentencing decisions. Anger, but not sadness, is linked to a tendency to rate mitigating factors as less important in sentencing decisions, and correspondingly, a greater likelihood of sentencing the defendant to death.

Footnotes

Appendix

This research was supported by an NSF grant (SES-0851004) to the 1st and 3rd author and by an NSF grant (SES- 0851183) to the 2nd author.

Narina Nuñez is a professor in the psychology department at the University of Wyoming. She studies jury decision making and how emotions affect decision making.

Bryan Myers is a professor of psychology at the University of North Carolina–Wilmington. He studies jury decision making and capital sentencing.

Benjamin M. Wilkowski is an associate professor of psychology at the University of Wyoming. He studies anger and aggression.

Kimberly Schweitzer is an assistant professor of criminal justice at the University of Wyoming. She studies how jurors weigh evidence in making their decisions.

References

Barnett

M. E.

Brodsky

S. L.

Davis

(2004). When mitigation evidence makes a difference: Effects of psychological mitigating evidence on sentencing decisions in capital trials. Behavioral Sciences & the Law, 22, 751-770.

Barnett

M. E.

Brodsky

S. L.

Price

J. R.

(2007). Differential impact of mitigating evidence in capital case sentencing. Journal of Forensic Psychology Practice, 7, 39-45. doi:10.1300/J158v07n01_04

Bless

Bohner

Schwarz

Strack

(1990). Mood and persuasion: A cognitive response analysis. Personality and Social Psychology Bulletin, 16, 331-345. doi:10.1177/0146167290162013

Blumenthal

J. A.

(2001). The admissibility of victim impact statements at capital sentencing: Traditional and non-traditional perspectives. Drake Law Review, 50, 67-91. Retrieved from http://heinonline.org/HOL/Page?handle=hein.journals/drklr50&;g_sent=1&id=77

Bodenhausen

G. V.

Sheppard

Kramer

G. P.

(1994). Negative affect and social perception: The differential impact of anger and sadness. European Journal of Social Psychology, 24, 45-62. doi:10.1002/ejsp.2420240104

Booth v. Maryland, 482 U.S. 496 (1987).

Boppre

Miller

(2015). How victim and execution impact statements affect mock jurors’ perceptions, emotions, and verdicts. Victims & Offenders, 9, 413-435. doi:10.1080/15564886.2013.845124

Bornstein

B. H.

(1999). The ecological validity of jury simulations: Is the jury still out? Law and Human Behavior, 23, 75-92. doi:10.1023/A:1022326807441

Butler

(2008). The role of death qualification in venirepersons’ susceptibility to victim impact statements. Psychology, Crime & Law, 14, 133-141. doi:10.1080/10683160701483534

10.

Commonwealth v. Kennedy. Case No. 1046, First District Court of Pennsylvania. Decided, March, 2003.

11.

Damasio

(1994). Descartes’ error: Emotion, reason, and the human brain. New York, NY: HarperCollins.

12.

Death Penalty Information Center. (2015). Legal issues: States that allow victim impact statements. Retrieved from http://www.deathpenaltyinfo.org/legal-issues-states-allow-victim-impact-statements

13.

DeSousa

(1987). The rationality of emotions. Cambridge, MA: MIT Press.

14.

Feigenson

(2000). Legal blame: How jurors think and talk about accidents. Washington, DC: American Psychological Association.

15.

Furman v. Georgia, 408 U.S. 238 (1972).

16.

Gardner v. Florida, 430 U.S. 349 (1977).

17.

Garvey

S. P.

(1998). Aggravation and mitigation in capital cases: What do jurors think? Columbia Law Review, 98, 1538-1576. doi:10.1002/acp.2880

18.

Georges

L. C.

Wiener

R. L.

Keller

S. R.

(2013). The angry juror: Sentencing decisions in 1st degree murder. Applied Cognitive Psychology, 27, 156-166. doi:10.1002/acp.2880

19.

Goldberg

J. H.

Lerner

J. S.

Tetlock

P. E.

(1999). Rage and reason: The psychology of the intuitive prosecutor. European Journal of Social Psychology, 29, 781-795. doi:10.1002/(SICI)1099-0992(199908/09)29:5/6<781::AID-EJSP960>3.0.CO;2-3

20.

Gordon

T. M.

Brodsky

S. L.

(2007). The influence of victim impact statements on sentencing in capital cases. Journal of Forensic Psychology Practice, 7, 45-52. doi:10.1300/J158v07n02_03

21.

Greene

(1999). The many guises of victim impact evidence and effects on jurors’ judgments. Psychology, Crime & Law, 5, 331-348. doi:10.1080/10683169908401776

22.

Greene

Koehring

Quiat

(1998). Victim impact evidence in capital cases: Does the victim’s character matter? Journal of Applied Social Psychology, 28, 145-156. doi:10.1111/j.1559-1816.1998.tb01697.x

23.

Gregg v. Georgia, 428 U.S. 153 (1976).

24.

Hatfield

Cacioppo

Rapson

R. L.

(1993). Emotional contagion. Current Directions in Psychological Science, 2, 96-99.

25.

Keltner

Ellsworth

P. C.

Edwards

(1993). Beyond simple pessimism: Effects of sadness and anger on social perception. Journal of Personality and Social Psychology, 64, 740-752. doi:10.1037/0022-3514.64.5.740

26.

Kyle

Twist

S. J.

Higgins

(2005). On the wings of their angels: The Scott Cambell, Stephanie Roper, Wendy Preston, Louarna Gillis, Nila Lyn Crime Victims’ Rights Act. Lewis & Clark Law Review, 9, 581-611. Retrieved from http://heinonline.org/HOL/Page?handle=hein.journals/lewclr9&;g_sent=1&id=601

27.

Lerner

J. S.

Goldberg

J. H.

Tetlock

P. E.

(1998). Sober second thought: The effects of accountability, anger, and authoritarianism on attributions of responsibility. Personality and Social Psychology Bulletin, 24, 563-574. Retrieved from https://www.researchgate.net/profile/Philip_Tetlock/publication/228079226_Sober_Second_Thought_The_Effects_of_Accountability_Anger_and_Authoritarianism_on_Attributions_of_Responsibility/links/557758da08aeacff20004b00.pdf

28.

Lerner

J. S.

Keltner

(2000). Beyond valence: Toward a model of emotion-specific influences on judgment and choice. Cognition and Emotion, 14, 473-493. doi:10.1080/026999300402763

29.

Lerner

J. S.

Tiedens

L. Z.

(2006). Portrait of the angry decision maker: How appraisal tendencies shape anger’s influence on cognition. Journal of Behavioral Decision Making, 19, 115-137. doi:10.1002/bdm.515

30.

Lockett v. Ohio, 438 U.S. 586 (1978).

31.

Logan

(1999). Through the past darkly: A survey of the uses and abuses of victim impact statements in capital trials. Arizona Law Review, 41, 143-192. Retrieved from http://heinonline.org/HOL/Page?handle=hein.journals/arz41&;g_sent=1&id=157

32.

Luginbuhl

Burkhead

(1995). Victim impact evidence in a capital trial: Encouraging votes for death. American Journal of Criminal Justice, 20, 1-16. doi:10.1007/BF02886115

33.

Mitchell

Myers

Broszkiewicz

(2016). Good or essential? The effects of victim characteristics and family significance on sentencing judgments and perceptions of harm. Psychiatry, Psychology and Law, 23, 651-669. doi:10.1080/13218719.2015.1084662

34.

Moons

W. G.

Mackie

D. M.

(2007). Thinking straight while seeing red: The influence of anger on information processing. Personality and Social Psychology Bulletin, 33, 706-720. doi:10.1177/0146167206298566

35.

Myers

Arbuthnot

(1999). The effects of victim impact evidence on the verdicts and sentencing judgments of mock jurors. Journal of Offender Rehabilitation, 29, 95-112. doi:10.1300/J076v29n03_05

36.

Myers

Godwin

Latter

Winstanley

(2004). Victim impact statements and mock juror sentencing: The impact of dehumanizing language on a death qualified sample. American Journal of Forensic Psychology, 22, 39-55.

37.

Myers

Greene

(2004). The prejudicial nature of victim impact statements: Implications for capital sentencing policy. Psychology, Public Policy, and Law, 10, 492-515. doi:10.1037/1076-8971.10.4.492

38.

Myers

Lynn

S. J.

Arbuthnot

(2002). Victim impact statements and juror judgments: The effects of harm information and witness demeanor. Journal of Applied Social Psychology, 32, 2393-2412. doi:10.1111/j.1559-1816.2002.tb01869.x

39.

Myers

Nuñez

Mitchell

Kehn

Wilkowski

(2017). The heterogeneity of victim impact statements: A content analysis of capital trial penalty transcripts. Unpublished manuscript.

40.

Myers

Roop

Kalnen

Kehn

(2013). Victim impact statements and crime heinousness: A test of the saturation hypothesis. Psychology, Crime & Law, 19, 129-143. doi:10.1080/1068316X.2011.614244

41.

Nuñez

Egan-Wright

Kehn

Myers

(2011, March). Impact of different methods of victim impact statement delivery at capital trials: Emotionality of statements and its impact on sentencing decisions. Talk presented at the 4th International Congress of Psychology and Law, Miami, FL.

42.

Nuñez

Schweitzer

Chai

Myers

(2015). The impact of anger on juror decisions in a capital trial. Applied Cognitive Psychology, 29, 200-209. doi:10.1002/acp.3094

43.

Paternoster

Deise

(2011). A heavy thumb on the scale: The effect of victim impact evidence on capital decision making. Criminology, 49, 129-161. doi:10.1111/j.1745-9125.2010.00220.x

44.

Payne v. Tennessee, 111, S. Ct., 2597 (1991).

45.

Pennebaker

J. W.

Francis

M. E.

Booth

R. J.

(2001). Linguistic inquiry and word count: LIWC 2001. Mahwah, NJ: Lawrence Erlbaum.

46.

Platania

Berman

(2006). The moderating effect of judge’s instructions on victim impact testimony in capital cases. Applied Psychology in Criminal Justice, 2, 84-101. Retrieved from http://docs.rwu.edu/cgi/viewcontent.cgi?article=1021&;context=fcas_fp

47.

Quigley

B. M.

Tedeschi

J. T.

(1996). Mediating effects of blame attributions on feelings of anger. Personality and Social Psychology Bulletin, 22, 1280-1288. doi:10.1177/01461672962212008

48.

Rucker

D. D.

Polifroni

Tetlock

P. E.

Scott

A. L.

(2004). On the assignment of punishment: The impact of general-societal threat and the moderating role of severity. Personality and Social Psychology Bulletin, 30, 673-684. doi:10.1177/0146167203262849

49.

Small

D. A.

Lerner

J. S.

(2008). Emotional policy: Personal sadness and anger shape judgments about a welfare case. Political Psychology, 29, 149-168. doi:10.1111/j.1467-9221.2008.00621.x

50.

Smith

C. A.

Haynes

K. N.

Lazarus

R. S.

Pope

L. K.

(1993). In search of the “hot” cognitions: Attributions, appraisals, and their relation to emotion. Journal of Personality and Social Psychology, 65, 916-929. doi:10.1037/0022-3514.65.5.916

51.

Solomon

(1990). A passion for justice. Reading, MA: Addison-Wesley.

52.

South Carolina v. Gathers, 490 U.S. 805 (1989).

53.

Stevenson

M. C.

Bottoms

B. L.

Diamond

S. S.

(2010). Jurors’ discussions of a defendant’s history of child abuse and alcohol abuse in capital sentencing deliberations. Psychology, Public Policy, and Law, 16, 1-38. doi:10.1037/a0018404

54.

Tetlock

P. E.

Visser

P. S.

Singh

Polifroni

Scott

Elson

S. B.

. . . Rescober

(2007). People as intuitive prosecutors: The impact of social-control goals on attributions of responsibility. Journal of Experimental Social Psychology, 43, 195-209. doi:10.1016/j.jesp.2006.02.009

55.

Texas v. Ronald Barney Fuller, 3rd Dist. AP-74, 980 (2004).

56.

Tiedens

L. Z.

Linton

(2001). Judgment under emotional certainty and uncertainty: The effects of specific emotions on information processing. Journal of Personality and Social Psychology, 81, 973-988. doi:10.1037/0022-3514.81.6.973

57.

Wainwright v. Witt, 469 U.S. 412 (1985).

58.

Watson

Clark

L. A.

(1994). The PANAS-X: Manual for the Positive and Negative Affect Schedule-Expanded Form. Iowa City: University of Iowa. doi:10.13072/midss.438

59.

Weiner

(1986). An attributional theory of motivation and emotion. New York, NY: Springer-Verlag.

60.

Wevodau

A. L.

Cramer

R. J.

Kehn

Clark

J. W.

(2014). Why the impact? Negative affective change as a mediator of the effects of victim impact statements. Journal of Interpersonal Violence, 29, 2884-2903. doi:10.1177/0886260514527170

61.

Young

M. J.

Tiedens

L. Z.

Jung

Tsai

M. H.

(2011). Mad enough to see the other side: Anger and the search for disconfirming information. Cognition and Emotion, 25, 10-21. doi:10.1080/02699930903534105