When the average American conjures up thoughts regarding the use of mental insanity as a legal defense, s/he might recall a famous instance such as the trial of John Hinckley Jr. who attempted to assassinate then-President Ronald Reagan (United States v. Hinckley, 1981). Although this case was fundamental in changing the laws regarding insanity, cases like United States v. Hinckley are not necessarily representative of trials involving mental illness or legal insanity (Fuller, 2000). While only approximately 1% of legal trials utilize mental insanity defenses, jurors in many more trials likely have questions about defendants’ mental health (Daftary-Kapur, Groscup, O’Connor, Coffaro, & Galietta, 2011). This article will analyze how individual juror characteristics relate to attitudes toward mental illness and the use of mental insanity defenses in legal settings. Additionally, this article will provide suggestions on how this information can be beneficial for jury selection and in crafting case theory.

Mental Illness and the Legal System

Mental insanity is used as a defense in approximately 1% of cases (Daftary-Kapur et al., 2011), but mental illness, in general, affects 1 in 5 Americans (National Alliance of Mental Illness, 2016). The rate of mental illness is even higher among the incarcerated population—an estimated 56% of State prisoners, 45% of Federal prisoners, and 64% of local jail inmates have some diagnosable mental health problem (James & Glaze, 2006) and 16.9% of sampled inmates from local jails have a serious mental illness (United States Department of Justice Archives, 2009). The drastic difference between the rates of mental illness in Americans in general compared to the rates of mental illness amongst the incarcerated populations suggests that mental illness might influence jurors’ attitudes and opinions toward defendants.

Individual Differences and Attitudes

A jury composed of (typically) twelve people provides the potential for a wide variety of attitudes to be present during the decision-making process. Although the United States Constitution requires that all jurors be fair and impartial (U.S. Const. amend. VI), decisions made by jurors are susceptible to prejudice and bias. The expression of prejudice and bias occurs in a variety of situations, ranging from law enforcement (Hall, Hall, & Perry, 2016) to healthcare (Thompson, 2011). Jurors can exhibit biases toward a defendant based on extralegal factors such as the defendant’s gender (McCoy & Gray, 2007), race (Minero & Espinoza, 2015; Pearson, Dovidio, & Pratto, 2007; Sommers & Ellsworth, 2001), religion (Johnson, 1985; Miller, Maskaly, Green, & Peoples, 2011), and political affiliation (Kassin & Wrightsman, 1983). Although there have been ample studies investigating bias based on characteristics of the defendant, those concerned with jury selection are likely more interested in bias related to characteristics of jurors.

Devine (2012) summarizes research regarding relationships between juror characteristics and attitudes or decision-making, but only a small number of studies have specifically examined those relationships in relation to mental insanity defense cases (see Bloechl, Vitacco, Neumann, & Erickson, 2007; Yelderman, West, & Miller, 2017). The limited research has found that males and church attenders have more negative attitudes than their counterparts (Taylor & Dear, 1981). Additional research has also associated politically conservative people and ethnically White people with more negative views of mental illness in general (Alexander & Link, 2003; Corrigan, Edwards, Green, Diwan, & Penn, 2001).

Based on the limited prior research examining individual differences in a legal setting, we conducted two studies to examine whether four variables—gender, race, religion, and political affiliation—are associated with attitudes toward the mentally ill and the use of mental insanity as a legal defense.

Study 1

The first study examined whether individual differences relate to attitudes toward people with mental illnesses. We hypothesized that Whites, religiously affiliated people, political conservatives (i.e., Republicans), and males—compared to their counterparts—would have more negative attitudes toward the mentally ill. These hypotheses were derived from past research (Alexander & Link, 2003; Corrigan et al., 2001; Taylor & Dear, 1981).

Materials

To test this hypothesis, the Community Attitudes Toward the Mentally Ill (CAMI) scale was used (Taylor & Dear, 1981). The CAMI is composed of four subscales—authoritarianism, benevolence, social restrictiveness, and community mental health ideology (CMHI). The authoritarian subscale includes themes related to causes of mental illness, differences between normal and mentally ill people, and how to treat those who have been diagnosed with a mental illness. The benevolence subscale examines attitudes regarding societal responsibility and involvement with those suffering from mental illness. The social restrictiveness subscale measures perceptions of how dangerous the mentally ill are, whether the mentally ill should have any responsibilities, and the normality of mental illness. Lastly, the CMHI subscale examines attitudes about a community’s impact on the mentally ill as well as attitudes about the impact of the mentally ill on the community (Taylor & Dear, 1981). Higher scores on the authoritarian and social restrictiveness subscales indicate more negative attitudes toward people who suffer from mental illness whereas higher scores on the benevolence and CMHI indicate more positive attitudes toward the mentally ill.

All individual characteristics were self-reported. Participants indicated their gender (male or female), political affiliation (no affiliation, Democrat, Republican, Independent, or other), race (White-American, Native-American, African-American, Asian-American, Hispanic-American, or Other), and religion (Catholic, Eastern Orthodox, Protestant, Jewish, Hindu, Buddhist, Muslim, Mormon, Atheist, Agnostic, “I believe in God, but do not have a particular religious affiliation,” or Other).

Participants

Participants (n = 441) were students at a university in the western United States who received partial credit in their social science courses. The study was hosted online as part of a larger survey examining attitudes toward a variety of legal issues.

Analysis

Because there were too few participants in some race categories, we separated participants into two groups—White participants and Non-White participants—and, similarly, participants were separated into two religious affiliation groups—those affiliated with a specific religion (Affiliated) and those who were not (Non-Affiliated). Statistical comparisons examined differences in CAMI subscale scores between males and females, White and Non-White participants, Affiliated and Non-Affiliated participants, and participants of various political affiliations: No Affiliation, Republican, Democrat, Independent, vs. Other.

Results

The study found mixed results that supported some hypothesis but not others and, in one instance, suggesting the exact opposite of our hypothesis.

Gender. The analyses for gender found no statistically significant differences between males and females on any of the CAMI subscales and, therefore, did not support our hypothesis.

Race. The analyses for race found a statistically significant difference on the authoritarianism, social restrictiveness, and CMHI subscales (p<.05) as well as a marginally significant difference on the benevolence subscale (p<.10). Surprisingly, White participants had lower authoritarianism and social restrictiveness subscale scores and higher benevolence and CMHI subscale scores, which is the opposite of what was hypothesized based on our review of prior literature.

Religious Affiliation. The analyses for religion found a statistically significant difference only on the authoritarianism subscale, which supported our hypothesis—but the lack of significant differences on the other three subscales made this support questionable.

Political Affiliation. Lastly, the analyses for political affiliation found significant differences on all four subscales. For the authoritarianism subscale, Republicans had significantly higher scores than the Democrat and Independent groups. For the benevolence subscale, Republicans had significantly lower scores than the Democrat and Independent groups. Republicans also had significantly higher social restrictiveness scores than the Independent group and marginally higher scores than the Democrat group. Finally, keeping with the trend, Republicans had significantly lower CMHI (i.e., attitudes toward the mentally ill in communities) scores than the Democrat and Independent groups. These results supported our hypothesis.

Discussion

Overall, Study 1 suggested that there are important differences among participants. Although there were no differences based on gender, the three other variables produced significant differences. Most notably, participants self- identifying as Republican had more negative opinions toward the mentally ill than Democrats and Independents. Additionally, Whites and Non-Affiliated participants had more positive views of the mentally ill than non-White and religiously affiliated participants based on their lower authoritarianism scores along with lower social restrictiveness and higher CMHI scores for Whites. These results would suggest that, in legal cases that have elements of mental illness, a jury composed of White, non-religiously affiliated, non-Republicans would have the most positive views of defendants with mental illnesses.

Study 2

While Study 1 focused on attitudes toward the mentally ill, it did not include any analyses of attitudes toward specific legal defenses. Therefore, a second study was conducted to examine whether individual differences relate to differing opinions regarding the existence of mental insanity and its use as a legal defense. For Study 2, we hypothesized that the same demographic groups from Study 1 (Non-Whites, religiously affiliated, and Republicans) that had more negative attitudes toward people with mental illnesses would also report a significantly lower self-reported belief in mental insanity and less acceptance of insanity defenses. Additionally, because of the Study 1 results, we hypothesized that gender would have no significant relationship with attitudes toward mental insanity or legal defenses.

Materials

To test the hypotheses, the study included questions that specifically asked participants whether they believe mental insanity should be allowed as a legal defense and whether the concept of mental insanity is even real. Participants also reported all demographic information.

Participants and Procedure

Participants (n = 550) were students at a university in the western United States who received partial credit in their social science courses for participating. The study was hosted online as part of a larger survey regarding legal attitudes.

Analysis

As with Study 1, certain groups were combined for the analysis because of an insufficient number of participants in certain categories; the same White/Non-White and Affiliated/Non-Affiliated categories were used. Therefore, statistical comparisons examined scores between males and females, White and Non-White participants, Religious Affiliated and Non-Affiliated participants, and political affiliation categories of No Affiliation, Republican, Democrat, Independent, vs. Other.

Results

The results of Study 2 were mixed regarding our hypotheses concerning attitudes about mental insanity and the insanity defense.

Gender. The analyses for gender found no statistically significant differences between males and females on either of the questions relating to mental insanity and, therefore, supported our hypothesis and the findings from Study 1.

Race. The analyses for race found a statistically significant difference only for the question asking whether mental insanity exists, with Whites being more likely to believe that insanity exists than Non-Whites. This finding provided support for our hypothesis and was in line with the findings in Study 1.

Religious Affiliation. The analyses for religion found a statistically significant difference for whether defendants should be able to use insanity as a defense and a marginally significant difference on whether insanity exists. Non-Affiliated participants scored higher on both questions—they were more likely to believe that insanity should be allowed as a legal defense and more likely to believe that insanity exists.

Political Affiliation. In stark contrast to Study 1, the political affiliation variable did not predict differences between the No Affiliation, Democrats, Republicans, Independents, or Other groups on either of the two questions.

Discussion

Results of Study 2 were less robust than Study 1. Significant differences between participants were only found based on race and religious affiliation. Further, differences on race were found for both questions of interest whereas differences based on religious affiliation were only found for one of the presented questions. Still, White and Non-Affiliated participants were more likely to believe that mental insanity exists. In addition, Non-Affiliated participants were more supportive of the idea that insanity should be allowed as a legal defense. However, neither gender nor political affiliation predicted significant differences for the two questions, suggesting that these two variables might be of limited utility when selecting jurors in mental insanity cases.

General Discussion

Previous research has found that pre-conceived attitudes can affect a person’s perception of events and potentially play a role in that person’s decision-making (Fazio, 1986). Therefore, prejudicial or biased attitudes toward another person could lead to biased and prejudicial decision-making. The findings from these two studies provide a foundation for attorneys and consultants to apply to the voir dire process. The findings suggest that White potential jurors would have more positive attitudes toward the mentally ill and more acceptance of the existence of mental insanity compared to non-White potential jurors. This finding was surprising and opposite of previous research (Corrigan et al., 2001) as well as what was initially hypothesized. However, this result might have been caused in part by combining all non-White participants into one comparison group. If enough non-White participants completed the survey to allow for multiple group comparisons, the same pattern of results might not be found.

Another significant finding was that non-religiously affiliated people had more positive attitudes toward the mentally ill, were more likely to believe that mental insanity truly exists, and were more likely to believe that mental insanity should be allowed as a legal defense compared to religiously affiliated people. These findings support results of previous studies (Taylor & Dear, 1981) and aligned with the researchers’ hypothesis. However, because all religiously affiliated people were combined into one group, the results do not indicate whether certain religious affiliations have more or less positive attitudes toward mental illness and insanity compared to other religious affiliations.

Interestingly, political affiliation was a significant predictor in Study 1, with Republicans having the most negative attitudes toward the mentally ill compared to Democrats and Independents, but was non-significant in Study 2. The results of Study 1 align with findings of previous research regarding the attitudes of conservatives (i.e., Republicans) toward the mentally ill (Alexander & Link, 2003) but this effect did not carry over to conservatives’ attitudes toward mental insanity defenses. This lack of a significant effect in Study 2 might be attributable to the legal ramifications and implications. Political conservatism is closely associated with authoritarianism and deferring to authority figures (Perlin, 1997) which could suggest that conservatives would defer to those who make the law (judges, lawmakers, et cetera) and align with the courts which allow mental insanity defenses.

Finally, there were no gender differences on attitudes toward the mentally ill or use of mental insanity as a legal defense, which contradicts previous research (Taylor & Dear, 1981). Regarding this lack of significant differences across gender, some previous studies have found gender to be a weak predictor in general (Devine, 2012), so it is not entirely surprising that there were no gender differences.

Suggestions for Jury Selection and Case Theory

These findings suggest that, for defense attorneys seeking sympathetic jurors for cases dealing with mental illness or insanity, it would be best to keep as many White, non-religiously affiliated potential jurors as possible. If the case involves aspects of a defendant’s mental illness but does not use a mental insanity defense, defense attorneys might also try to limit the number of Republicans selected. Conversely, for prosecuting attorneys who seek jurors that are unsympathetic to mental illness and the insanity defense, these results suggest keeping as many non-White and religiously affiliated potential jurors as possible. Maximizing the number of Republicans would also benefit the prosecution if the case involves aspects of mental illness but not a mental insanity defense.

Although keeping or dismissing potential jurors based on these personal characteristics might sound good in theory, it is, in some ways, illegal in practice. Previous Supreme Court rulings such as J.E.B. v. Alabama (1994) and Batson v. Kentucky (1986) prevent attorneys from dismissing potential jurors based strictly on that juror belonging to a clearly identifiable group (i.e., a specific race, ethnicity, or gender). Therefore, another and possibly more applicable use of this information, especially regarding race, would be in creating case theory. Once the jury has been selected, attorneys could use these results to shape their case theory. For instance, defense attorneys might want to limit the mentioning of the defendant’s mental health issues if the jury is composed of primarily non-White, religiously affiliated, Republican jurors. On the other hand, if the jury were composed of White, non-religiously affiliated jurors, these results suggest the jury would be more sympathetic to and accepting of a case theory focusing on the defendant’s mental illness.

Lastly, once attorneys know the characteristic make-up of the jury, they could potentially use certain psychological principles to overcome potential disadvantages. Jurors are more likely to show leniency toward a defendant when the defendant is seen as similar to themselves (Abwender & Hough, 2001), but this effect would likely be limited to when the evidence is weak or ambiguous (Kerr, Hymes, Anderson, & Weathers, 1995). Therefore, defense attorneys might focus on accentuating similarities between the defendant and the jurors, such as race or religion, when the evidence is ambiguous or weak to create the perception amongst jurors that the defendant is part of their “in-group.” In cases of weak or ambiguous evidence, prosecuting attorneys might focus on the differences between the defendant and jurors to create a perception that the defendant is not similar to jurors and is, rather, part of their “out-group.” This similarity-leniency effect does have limitations, however. When the evidence against the defendant is strong, similarities between the defendant and jurors could cause a “black sheep” effect in which jurors want to distance themselves from the defendant and would be more likely to convict (Kerr et al., 1995).

Limitations

As with all studies, there are limitations to the presented findings. The primary limitation is participants self-reported their attitudes and opinions in isolation, meaning these findings might not be generalizable to the group context of jury deliberations. Additionally, our sample was composed entirely of students, which might limit the generalizability of the findings. Although a recent meta-analysis by Bornstein et al. (2017) suggests there are few substantial differences between student and non-student mock jurors, it is still a factor that must be considered. A final concern in the study centers around the lack of consequences associated with participant responses. As with some mock jury studies, participants were under no belief that their responses would have any real-life consequences which may have affected responses (Bornstein & McCabe, 2005).

Conclusion

It is crucial for judges and attorneys to identify and remove potential jurors who might use bias or prejudice in their verdict. With such a large percentage of the prison population suffering from some form of mental illness, it is crucial to understand community sentiment toward people with a mental illness and the insanity defense. Further research is necessary to confirm these findings, but the studies provide a foundation for helping make jury selection and case framing decisions when attorneys faced with a case involving mental illness or mental insanity.

---

Charles P. Edwards, MA [[email protected]] is a doctoral student in the Interdisciplinary Social Psychology Ph.D. Program at the University of Nevada, Reno. He received his MA in psychology in 2014 from Boston University. His research interests are in psychology and law; specifically, jury decision-making.

Monica K. Miller, J.D., Ph.D. [[email protected]] is a Professor of Criminal Justice and Interdisciplinary Social Psychology at the University of Nevada, Reno. She received her J.D. in 2002 from the University of Nebraska College of Law and her Ph.D. in social psychology in 2004 from the University of Nebraska-Lincoln. Her interests involve the application of psychological theories and justice principles to laws and policies. Specifically, she is interested in the role of religion in the legal system (e.g., jury decisions); how the law regulates sexual behavior, pregnancy and family issues; jury decisions in death penalty, medical malpractice and insanity cases; community sentiment and public policy; and vigilante justice in the legal system and the media. Her webpage is: https://www.unr.edu/criminal-justice/people/monica-miller

---