A mock insanity defense trial was presented to 140 college undergraduates by means of an audiotape and synchronized slides. Participants answered a series of questions regarding the case and their predeliberation verdict selections. Consistent with prior research, jurors who selected a guilty verdict neither believed that the defendant could be rehabilitated nor that he suffered from some mental disease or defect. Jurors who opted for a guilty verdict held favorable attitudes toward the death penalty, were crime-control oriented, and held unfavorable attitudes toward the insanity defense. Jurors who rendered a Guilty But Mentally Ill verdict differed significantly in their evaluations of the defendant's mental status. Mean ratings for jurors reaching this verdict were intermediate between the ratings of jurors reaching Not Guilty by Reason of Insanity and guilty verdicts. Although jurors' attitudes and their evaluations of the evidence were both important correlates of verdict selection, evaluation of the evidence was weighted more heavily in the function discriminating among verdicts.
Much of the recent insanity defense research was precipitated by legislative changes made in insanity defense statutes (Low, Jeffries, & Bonnie, 1986; Rogers, 1987). Many of the legislative changes were designed in response to public outcry following highly publicized Not Guilty by Reason of Insanity (NGRI) acquittals (Roberts, Golding, & Fincham, 1987). As reported by Low et al., (1986), 7 states shifted the burden of proof to the defendant, 9 states narrowed their insanity plea standards or tests, and 13 states adopted the Guilty But Mentally Ill (GBMI) verdict option.
It was clear that the role of the GBMI verdict option was to cause a reduction in what might otherwise have been Not Guilty by Reason of Insanity (NGRI) verdicts (Finkel & Duff, 1989; Harris, 1983; Kaufman, 1982). The rationale underlying Guilty But Mentally Ill (GBMI) legislation is that the verdict would serve as a "compromise" option, i.e., an option which would be used by jurors when they were unable to decide between NGRI or guilty (Harris, 1983; Kaufman, 1982; Low et al., 1986; McGraw, Farthing-Capowich & Keilitz, 1985; Rogers, 1987; Slobogin, 1985).
In recent years, a number of simulated studies have featured the GBMI verdict option as a choice among the more traditional verdict options of guilty, NGRI, or not guilty. These studies (Finkel & Duff, 1989; Poulson, 1990; Roberts et al., 1987; Roberts & Golding, 1991; Roberts, Sargent, & Chan, 1993; Savitsky & Lindblom, 1986) have shown that the availability of a GBMI verdict option does cause a large displacement of both NGRI verdicts as well as guilty verdicts. Roberts et al. (1987) simulated juror decision-making by asking students to choose a verdict in response to 16 vignettes. In one condition participants were only allowed to choose from among the traditional verdict choices of NGRI, Guilty, and Not Guilty. In the second condition, participants were allowed to choose from among the traditional verdict choices plus the GBMI verdict. When made available the GBMI verdict was chosen 2.5 times more often than either NGRI or guilty verdicts.
Similarly, Poulson (1990) utilized the simulated jury approach in an attempt to assess how the Guilty But Mentally Ill (GBMI) verdict, along with manipulation of race of defendant and race of victim, affected mock jurors' attributions of criminal responsibility. Based upon predeliberation verdicts, it was shown that NGRI and guilty verdicts were reduced by approximately two-thirds when the GBMI verdict was made available.
Roberts et al. (1993) concluded that "this pronounced `verdict-shifting' phenomenon has been demonstrated for the within- and between-participants research designs, at the individual juror and jury levels of analysis, across diagnostic classes of defendants, and with college and community samples" (p. 262). Hence, the availability of the alternative GBMI verdict option appears to have a strong effect on mock jurors' insanity defense verdict selections.
Evidential/legal factors. Of the legal/evidential factors which affect jurors' insanity defense verdicts, mental status of the defendant has received the most attention and empirical validation. For instance, research conducted by Finkel and Handel (1989), Poulson (1988,1990), Roberts et al. (1987), and Roberts et al. (1993) showed that jurors' evaluations of the defendant's mental disorder significantly affected their verdict selections.
Poulson (1988,1990) has shown that jurors who evaluated expert testimony given by the defense as being more effective than the prosecution's expert testimony were significantly more likely to render either a NGRI or GBMI verdict than a guilty verdict. This same tendency was found for jurors who were of the opinion that the defendant was capable of being rehabilitated (Poulson, 1988).
These general findings suggest that jurors' evaluations of legal/ evidential factors (defendant's mental status, expert psychological testimony, and likelihood of rehabilitation) may play an important role in determining how jurors reach their final insanity defense verdicts.
Attitudinal/extralegal. At this point we should emphasize that, although the evidence presented in simulated studies (i.e., Finkel & Handel, 1989; Poulson, 1988, 1990; Roberts et al., 1987) was held constant, jurors still reached disparate verdicts. This suggests that evidence presented during the trial might have been evaluated by jurors in different ways, which, of course, may account for the disparity in their verdict selections. The critical question becomes: what attitudinal factors affect jurors' selection of verdict by influencing how they go about evaluating legal/evidential factors, such as the defendant's mental status?
In the present study we examined jurors' attitudes regarding the insanity defense, attitudes toward attorneys, attitudes about lenient treatment of accused persons, and attitudes regarding the death penalty. The rationale for isolating these particular attitudinal factors stems from previous research (Arafat & McCahery, 1973; Ellsworth, Bukaty, Cowan, & Thompson, 1984; Finkel & Handel, 1989; Roberts et al., 1987) indicating their probable importance in the jury decision-making process.
Research findings (e.g., Finkel & Duff, 1989; Poulson, 1990; Roberts et al., 1987; Roberts & Golding, 1991; Roberts et al., 1993; Savitsky & Lindblom, 1986) have been quite consistent regarding the GBMI verdict's displacement of what might otherwise have been either NGRI or guilty verdicts. However, fewer studies (e.g., Finkel & Duff, 1989; Poulson, 1988; Roberts et al., 1987; Roberts & Golding, 1991; Roberts et al., 1993) have specifically examined how jurors who opt for a GBMI verdict differ in their evaluation of the evidence and their attitudes towards insanity defense issues as compared to jurors who vote guilty or NGRI.
Roberts et al's. (1993) comparison of jurors from the three verdict groups (NGRI, GBMI, and guilty) revealed that significant differences existed among "construal" ratings of the defendant's: level of mental disorder; capacity for controlling psychotic beliefs; capacity for displaying rational behavior; and levels of deserved blame and punishment. Put simply, mean "construal" ratings for people reaching a GBMI verdict tell in the middle of ratings given by jurors who opted for the more extreme NGRI or guilty verdicts.
Using predeliberation verdicts as a classification variable, Poulson (1988) found that jurors rendering GBMI verdicts differed significantly in their evaluations of the defendant's psychological make-up from jurors returning guilty and NGRI verdicts. Neatly paralleling the American Legal Institute (ALI) standard/test for insanity, these differences included: 1) whether the defendant could appreciate the criminality of crime; 2) whether he could have conformed his conduct to the requirements of the law; and, 3) whether the defendant was believed to have suffered from an antisocial personality disorder or from schizophrenia. Consistent with Roberts et al's. (1993) findings, mean ratings given by jurors who opted for the GBMI verdict fell neatly between the mean ratings given by jurors selecting either NGRI or guilty verdicts.
Although most of these studies have produced similar results, we recognized that important differences existed in how these mock trials were conducted. For example, a majority of the previous GBMI studies (e.g., Roberts & Golding, 1991; Roberts et al.,1993) were conducted with the "burden of proof" placed upon the defendant to show by "a preponderance of the evidence" that he or she was "insane" at the time the crime was committed. Indeed, this standard and burden of proof is consistent with a majority of jurisdictions including all Federal jurisdictions (Callahan, Mayer, & Steadman, 1987).
There are other jurisdictions which still place the burden of proof upon the "state" to show "beyond a reasonable doubt" that the defendant was "sane" at the time of the instant offense. Therefore, a most important question is raised as to whether similar results would be found under these conditions. Research conducted by Ogloff (1991) suggests that neither "burden of proof" nor "standard of proof" has a significant effect on how jurors go about deciding their final insanity defense verdicts. The present study is designed to provide a general re-examination of this critical finding.
Equally important, a number of previous studies (Roberts et al., 1987; Roberts & Colding, 1991; Roberts et al., 1993) did not include detailed psychological testimony as part of their legal vignettes. Instead, Roberts and colleagues elected to eliminate all the diagnostic terminology or psychopathological jargon from their simulated studies. Furthermore, the experts in their studies were not allowed to provide conclusory expert opinions. This approach is consistent with a number of recommendations presented and accepted through the Insanity Defense Reform Act of 1984. However, we observed research (e.g., James, 1960; Ogloff, 1991; Pasewark, 1986; Steadman, Keitner, Braff, & Arvanities 1983) which suggests that diagnoses of psychotic thought and behavior are associated with actual insanity defense plea adjudications. For example, in 1960, James conducted a study of juror evaluation of expert psychiatric testimony using mock jurors. The study revealed that 74% of mock jurors believed psychiatric testimony was helpful in choosing a verdict. Therefore, we elected to maintain the diagnostic terminology and psychopathological jargon which was featured in the actual trial from which our simulated case was drawn. The critical question is whether similar results will be observed in the present study where expert psychological testimony was given a much greater role than in previous research studies.
In sum, the present study was conducted with a number of trial variants different from those in previous studies. As stated earlier, we placed the burden of proof on the "state" to show "beyond a reasonable doubt" that the defendant was sane at the time the crime was committed. We also elected to include detailed psychological testimony; and, also to have expert witnesses provide conclusory testimony. Further, our prototypical insanity defense case (State of Illinois v. John Carlton) was complete with opening and closing arguments, cross-examinations, rebuttals, side-bars, extensive psychological testimony and extensive psychological history.
These critical trial variants will allow us to assess convergent validity. If the results of this study are analogous to previous studies which were conducted with markedly different trial characteristics, we can then have much greater confidence in the robustness and validity of these effects across different research methodologies.
Participants were 140 undergraduate students enrolled in introductory psychology classes at a major Southeastern university. The average age of participants was 19.7 years of age. Ninety-six percent of participants were White and 4% were Black. Sixty-two percent of participants were women and 38% were men. Participants received course credit for their participation. College students were used due to the unavailability of jurors from the local jury pool.
The simulated trial, State of Illinois v. John Carlton, was based on an insanity defense trial which took place in Chicago's Cook County Court. The actual trial was held over a one-week period. In order to reduce the simulated case to 90 minutes, certain testimony had to be deleted. Specifically, even in view of his confession, in the original case several witnesses were called to establish the fact that a crime had occurred and that the correct person had been arrested. These witnesses (i.e., 3 lab experts, 2 police officers, etc.) were removed from the mock-case. However, the medical examiner's testimony, condensed in form, was retained. This testimony, along with the testimony of the arresting officer, and the defendant's confession, sufficed to show that a murder had occurred and the correct person was arrested. Thus, the mock-trial maintained the original trial's bifurcated procedure; first establishing that a crime had occurred, followed by evidence about what caused or precipitated the crime.
In the simulated trial the defendant, John Carlton, was charged with the first degree stabbing death of the victim, Melissa Craft. John Carlton's defense pleaded that he was not guilty by reason of insanity. The prosecution, of course, argued that John Carlton was sane at the time of the crime.
John Carlton had been hospitalized for mental problems on two previous occasions. Specifically, the testimony established that John Carlton was first referred for psychological examination at the Illinois State Hospital on April 18, 1971. This referral followed from his arrest for assault and battery. While at the Illinois State Hospital John Carlton was diagnosed as "schizophrenic undifferentiated." On February 19, 1979, John Carlton was again admitted to the Illinois State Hospital following a second altercation with the law. Carlton had been arrested and referred for psychological diagnosis after he assaulted his mother and aunt. During his second stay, John Carlton was again diagnosed as "schizophrenic undifferentiated." He was later released from the hospital when his schizophrenia was said to have gone into remission. Perhaps most important, John Carlton was said to have suffered from delusions which were focused on women and the need for their religious persecution. Following the presentation of testimony and the attorneys' closing arguments, the judge instructed the jurors, focusing on the American Legal Institute (ALI) standard for insanity and the Illinois Pattern of Instructions (IPI) definition of "guilty but mentally ill." Specifically, the mock jurors were instructed as follows:
Importantly, ladies and gentlemen, a person is insane and not criminally responsible for his conduct if, at the time of the crime, as the result of mental disease or mental defect, he lacks substantial capacity to either appreciate the criminality of his conduct, or to conform his conduct to the requirements of the law. (Illinois Revised Statutes, 1985, Chapter 8, Section 6-2. (a))
A person is guilty but mentally ill if, at the time of the commission of the offense, he was not insane, but was suffering from a mental illness. Mental illness means a substantial disorder of thought, mood, or behavior which affected a person at the time of the commission of the offense and which impaired that person's judgment, but not to the extent that he is unable to appreciate the wrongfulness of his behavior or is unable to conform his conduct to the requirements of the law. (Illinois Revised Statutes, 1981, Chapter 8, 6-2. (c))
Design and Procedure
At the beginning of the study, the participants were informed that they would view a simulated insanity defense trial with accompanying slides and that the re-enactment was based on an actual Cook County insanity defense trial. The participants were asked to assume the role of an actual juror and to decide the case as if it were an actual case in which they had been called as jurors. The experimenter asked the participants neither to discuss the case nor the study with each other or friends until the end of the academic semester.
The participants were tested in groups ranging in size from 14 to 20. After viewing the simulated trial, the participants were administered a battery of dependent measures. The primary dependent variable was the mock juror's prediction of verdict selection (guilty, not guilty, NGRI, and GBMI).
In addition, the participants were asked a series of questions regarding their attitudes on relevant extralegal issues and their evaluation of the evidence. From their responses to these questions, scores on five attitudinal and three evaluative clusters of variables were obtained. The attitudinal variables measured attitude about lenient treatment of accused persons, attitude towards the insanity defense, attitude towards the death penalty, attitude toward the defense attorneys, and attitude towards the prosecuting attorneys. The evaluative variables measured degree of belief in the expert testimony of the defense experts versus the prosecution experts, assessment of the defendant's mental status, and belief regarding whether the defendant could be rehabilitated.
In order to determine which of the evaluative and attitudinal factors were important in producing the differences in verdict choice, we conducted a principal components discriminant function analysis. A discriminant function is a weighted linear combination of the predictor variables, with the weights chosen such that the criterion groups differ as much as possible on the resulting discriminant function. In our analysis, verdict choice served as the criterion variable. The predictor variables were the five extralegal/attitudinal and three legal/evidential clusters of variables described in the Methods section and in Appendix 1. To avoid problems associated with multicollinearity among the original variables, these variables were subjected to a principal components analysis, with the resulting orthogonal components used as the predictors in the discriminant analysis. The results of this analysis were then transformed back into a form interpretable in terms of the original variables by correlating the participants' raw scores on the original eight variables with the participants' scores on the two significant discriminant functions (DF). These correlations are given in the structure matrix displayed in Table 1.
Table 2 contains the classification means for the groups on each discriminant function as well as the group means on each of the eight original variables. The classification means indicate that the first function distinguishes between participants choosing a guilty verdict and participants returning a Not Guilty By Reason of Insanity (NGRI) verdict, F(16,252) = 10.71, p < .001). Believing that the defendant was mentally ill, believing the defense's expert testimony more than the prosecution's, being receptive to the insanity defense, opposing the death penalty, and favoring lenient treatment were associated with rendering a NGRI verdict. Conversely, the opposite orientation on these factors was associated with rendering a guilty verdict. The second function separated those who rendered a Guilty But Mentally Ill (GBMI) verdict from those choosing guilty or Not Guilty By Reason of Insanity (NGRI), F(7,127) = 3.40, p < .003). Distrusting the attorneys (especially the prosecution attorney), thinking rehabilitation likely, opposing lenient treatment, not being receptive to the insanity defense, and favoring the death penalty were associated with rendering a GBMI verdict rather than a guilty or NGRI verdict.
Those who prefer univariate presentation of results should focus on the last eight rows of Table 2. We should underscore that on every variable, except the trust of the attorneys variables, the mean ratings for the GBMI group fall neatly between the mean ratings given by the jurors who opted for a guilty verdict and jurors who opted for a NGRI verdict.
Fisher's procedure was used to make pairwise comparisons among the groups on each of the variables, including the discriminant functions. When employed to make pairwise comparisons among three and only three groups. Fisher's procedure has been found to hold family wise error at or below the nominal rate and to have more power than commonly employed alternative procedures (Levin. Serlin. & Seaman, 1994).
Data from the present study clearly support and extend previous research findings (e.g., Finkel & Duff, 1988; Poulson, 1988; Roberts et al., 1987; Roberts & Golding, 1991; Roberts et al., 1993). For example, compared to jurors who reached either a NGRI or a guilty verdict, data from the present study show that jurors who rendered a GBMI verdict differed significantly in their evaluation of the defendant's mental status. Mean evaluation ratings for jurors reaching a GBMI verdict were intermediate between the ratings of persons reaching NGRI and guilty verdicts.
When measuring receptivity to the insanity defense and opposition to the death penalty, our data show that jurors who vote GBMI are more like jurors who vote guilty. That is, jurors who voted GBMI were generally opposed to the insanity defense and they were supportive of the death penalty. Again, this finding is quite consistent with previous research (Poulson, 1988; Roberts et al., 1987; Roberts & Golding, 1991; Roberts et al., 1993).
However, we placed the burden of proof on the "state" to show "beyond a reasonable doubt" that the defendant was sane at the time the crime was committed. We also elected to include detailed psychological testimony and to have expert witnesses provide conclusory testimony. Yet, despite the methodological trial variants featured in the present study, our results are amazingly similar to previous findings.
As such, our date tentatively support Ogloff's (1991)findings which show that burden of proof and standard of proof does not have a significant effect on how jurors go about determining their insanity defense verdicts. Instead, it appears that jurors may rely on their own intuitive understanding of what proof should be established for determining whether a defendant should be excused of criminal wrongdoing by virtue of some mental disease or mental defect.
Overall, these results clearly suggest that previous findings regarding characteristics of jurors who opt for a GBMI verdict are not an artifact of how the trial is presented. Instead, this appears to be a robust finding which holds across different experimental methodologies, settings, samples, and times. It appears that jurors who vote GBMI may best described as middle-of-the-road jurors, particularly as they evaluate the defendant's mental status in relation to the charged offense.
Although the results revealed in previous studies (i.e. Finkel et al., 1989; Poulson, 1988; Roberts et al., 1987) appear both stable and robust, there may be a methodological artifact related to when the final verdict was requested. Jurors may align the evidence in a way which is most consistent with their verdict choice (i.e. verdict-driven). In other words, jurors' responses to the follow-up dependent measure questions may be constructed in such a way as to justify their verdict preference.
Research conducted by Poulson and Brondino (1990) and Roberts et al. (1993) suggest that mock jurors' attribution of criminal responsibility is not significantly affected by whether jurors' verdicts are polled at the beginning or at the end of the dependent measures questionnaire. Therefore, differences observed among the three verdict choice groups do not appear to be a simple artifact of verdict placement.
Factors Which Discriminate Among Jurors' Verdict Preferences
Evidential/legal. Of all the variables included in the model, mental status of defendant assumed the strongest relationship with verdict. As expected, NGRI verdict jurors tended to evaluate evidence in a manner which is opposite of jurors who returned guilty verdicts. Jurors who returned an NGRI verdict believed the defendant to be unable to conform his conduct to the requirements of the law and unable to appreciate the criminality of his actions. They believed the defendant suffered from hallucinations brought about by his acute schizophrenia.
Conversely, jurors opting for a guilty verdict did not believe the defendant suffered from schizophrenia. They were of the opinion that the defendant could conform his conduct to the requirements of the law and could appreciate the criminality of his actions. They believed the expert testimony of the prosecution and did not believe the defendant could be rehabilitated. Analogous to the findings of Roberts et al. (1987) and Ellsworth et al. (1984), the data showed that some people do not accept alleged mental illness as a basis for criminal exculpation. Rather, these jurors conclude that allegedly mentally ill persons should be held responsible for their acts and should be punished in a manner equal to a person who claims no mental illness.
Unlike jurors who selected the guilty verdict, jurors who selected the GBMI verdict thought that rehabilitation was possible. However, it is important to note that the GBMI verdict allows for the defendant to be sentenced in the same way as if he or she had been found guilty, but the GBMI verdict also promises mental health evaluation or treatment during the term of the sentence (Kaufman, 1982).
With regards to mock juror evaluation of expert testimony, data featured in the present study show that significant differences exist among jurors who opt either for an NGRI verdict, a GBMI verdict, or a guilty verdict. Again, mean ratings given by jurors who opted for a GBMI verdict fell intermediate to mean ratings given by jurors who voted NGRI and jurors who voted guilty.
Attitudinal/extralegal. Jurors selecting the NGRI verdict were found to hold favorable opinions of the insanity defense. Consistent with Ellsworth et al. (1984), jurors who returned a NGRI verdict were more likely to be oriented toward due process of law (lenient treatment of defendants) than crime control, and they were more likely to be opposed to the death penalty. Jurors who returned a guilty verdict fell in the opposite direction on these extralegal dimensions. These findings are supportive of the research conducted by Ellsworth et al. (1984), Finkel and Handel (1989), Poulson (1988), and Roberts et al. (1987). For example, Ellsworth et al. (1984) reported that between 80% and 90% of the participants who were death-qualified (and crime-control oriented) rejected the insanity defense for cases which involved schizophrenia.
Jurors who returned a GBMI verdict were similar to those who returned a guilty verdict in that they held unfavorable attitudes towards the insanity defense, were in support of the death penalty, and were oriented toward crime control as opposed to due process of law. Unlike those who rendered guilty verdicts, they were suspicious of both prosecution and defense attorneys.
Overall, results from the present study show a significant relationship between jurors' attitudes and their verdict choices. This finding, however, requires some qualification. With respect to discriminating between selection of the NGRI versus the guilty verdict, evidential factors (mental status and expert testimony) were more heavily weighted than were attitudinal factors. The results of this study indicate that jurors' assessment of the defendant's mental status is the most important factor in the determination of a final insanity defense verdict. In many ways this seems commonsensical because the defense invokes the defendant's mental status as a causal basis for his criminal wrongdoing.
Limitations of the Study
Critics have claimed that unless an actual case is occurring, the mock jurors may not care about the outcome of the simulated case. Because the participants used in the present study were college students, although eligible for actual jury selection, one might wonder whether similar effects might be found with a different population; namely, older adults. Although our response to this important issue awaits further empirical testing, a number of researchers (e.g., Hastie, Penrod, & Pennington, 1983; Kalven & Zeisel, 1966; Simon, 1967) have reported that mock jurors show a nearly equal level of involvement in simulated cases as do actual jurors in real cases.
An additional potential limitation of the present study is that it focused on predeliberation verdicts. One may argue that this results in the restriction of the external validity of the study. The basic question is whether predeliberation verdicts map onto postdeliberation verdicts. Again, past research has shown that a high correlation exists between pre- and postdeliberation verdicts (Hastie et al., 1983; Kalven & Zeisel, 1966). However, future studies may be best served by examining pre- and postdeliberation verdicts.
One would expect that when the Guilty But Mentally Ill (GBMI) option was unavailable, jurors who would have returned a GBMI verdict would otherwise have to opt for a either a straight guilty verdict or a Not Guilty By Reason of Insanity (NGRI) verdict. To more clearly explain these potential differences, however, a repeated measures design is necessary. Therefore, future research should be conducted with the use of a repeated measures design in which each juror participates in two conditions: GBMI available and GBMI option not available. In doing so, we may capture a more refined analysis of how jurors from the various verdict categories differ in regards to their evaluation of legal and extralegal factors already mentioned. Further, particular emphasis should be placed on identifying characteristics of jurors who select the GBMI category when it is available.
Authors' Notes: Research made possible in part by a grant from the National Science Foundation (#R11-9012742) awarded to Ronald L. Poulson. We would also like to thank Frank Dane, Kelley Davenport, Reid Hastie, and two anonymous reviewers 7for their insightful comments regarding the legal aspects of the study, as well as Robin O'Hearn for help with the data collection. We would also like to dearly thank Marion Eppler for her many editorial suggestions regarding the present manuscript.
Variable DF1 DF2
Mental status of defendant .87 -.19
Evaluation of expert testimony .75 .12
Receptivity to insanity defense .65 .28
Opposition to death penalty .54 .25
Favoring lenient treatment -.34 .35
Believing rehabilitation unlikely -.41 .43
Trusting the prosecuting attorneys -.10 .50
Trusting the defense attorneys .13 .29
Variable Guilty GBMI NGRI
Discriminant Function 1 -1.29(a) -0.01(b) 2.32(c)
Discriminant Function 2 0.48(a) -0.39(b) 0.46(a)
Mental status of defendant 2.86(a) 4.61(b) 6.98(c)
Evaluation of expert testimony -2.47(a) -0.71(b) 3.43(c)
Receptivity to insanity defense 2.01(a) 2.22(a) 3.02(b)
Opposition to death penalty 1.56(a) 1.81(a) 2.82(b)
Favoring lenient treatment 1.88(a) 1.89(a) 2.50(b)
Believing rehabilitation unlikely 6.82(a) 5.47(b) 4.91(b)
Trusting the prosecuting attorneys 2.21(a) 1.81(b) 2.00(ab)
Trusting the defense attorneys 2.03(a) 1.75(a) 1.77(a)
Note: Within each row, means having the same letter in their
superscripts are not significantly different from each other at
the .05 level.
Arafat, I., & McCahery, K. (1973). The insanity defense and the juror. Drake Law Review, 22, 538-549.
Callahan, L., Mayer, C., & Steadman, H. (1987). Insanity defense reform in the United States-Post Hinckley. Mental and Physical Disabilities Laws Reporter, 11, 54-59.
Ellsworth, P.C., Bukaty, R.M., Cowan, C.L., & Thompson, W.C. (1984). The death-qualified jury and the defense of insanity. Law and Human Behavior, 8, 81-93.
Finkel, N.J., & Duff, K.B. (1989). The insanity defense: Giving jurors a third option. Forensic Reports, 2, 235-263.
Finkel, N.J., & Handel, S.F. (1989). How jurors construe "insanity." Law and Human Behavior, 13, 41-59.
Harris, J.D. (1983). Note: Guilty but mentally ill: A critical analysis. Rutgers Law Journal, 14, 453-477.
Hastie, R., Penrod, S.D., & Pennington, N. (1983). Inside the jury. Cambridge, MA: Harvard University Press:.
Insanity Defense Reform Act of 1984. (1984). Public Law 98-473, Sections 401-406.
Illinois Pattern Jury instructions--Criminal (2nd ed.). (1981). St. Paul, MN: West.
James, R.M. (1960,Winter). Jurors' evaluation of psychiatric testimony. Ohio State Law Journal, 21, 75-95
Kaufman, I.R. (1982). The insanity plea of trial. New York Times Magazine, 1, 16-20.
McGraw, B., Farthing-Capowich, D., & Keilitz, I. (1985). The "guilty but mentally ill" plea and verdict: current state of knowledge. Villanova Law Review, 30, 117-191.
Kalven, H. Jr., & Zeisel, H. (1966). The American jury. Boston: Little, Brown.
Levin, J.R., Serlin, R.C., & Seaman, M.A. (1994). A controlled, powerful multiple-comparison strategy for several situations. Psychological Bulletin, 115, 153-159.
Low, P.W., Jeffries, J.C., & Bonnie, R.J. (1986). The trial of John W. Hinckley Jr.: A case study in the insanity defense. Mineola, NY: Foundation Press.
Ogloff, R.P. (1991). A comparison of insanity defense standards on juror decision making. Law and Human Behavior, 15, 509-531.
Pasewark, R.A. (1986). A review of research on the insanity defense. Annals of the Academy of Political and Social Science, 484, 100-114.
Poulson, R.L. (1988). The effects of legal and extralegal factors on juror attribution of criminal responsibility: An empirical analysis (Doctoral dissertation, Northwestern Univ., 1988). Dissertation Abstracts International, 49, 5065B.
Poulson, R.L. (1990). Mockjuror attribution of criminal responsibility: Effects of race and the guilty but mentally ill (GBMI) verdict option. Journal of Applied Social Psychology, 20, 1596-1611.
Poulson, R.L. & Brondino, M.J. (1990). Jurors and the guilty but mentally ill verdict: A question of evidence vs. verdict-driven attributions of criminal responsibility. Unpublished manuscript.
Roberts, C.F., & Golding, S.L. (1991). The social construction of criminal responsibility and insanity. Law and Human Behavior, 15, 349-376.
Roberts, C.F.. Golding, S.L.. & Fincham, F.D. (1987). Implicit theories of criminal responsibility: Decision making and the insanity defense. Law and Human Behavior, 11, 207-232.
Roberts, C.F., Sargent, E.L., & Chan, A.S. (1993). Verdict selection processes in insanity cases. Juror construals and the effects of guilty but mentally ill instructions. Law and Human Behavior, 17, 261-275.
Rogers, R. (1987). APA's position on the insanity defense: Empiricism versus emotionalism. American Psychologist, 42, 840-848.
Savitsky, J.C., & Lindblom, W.D. (1986). The impact of the guilty but mentally ill verdict on juror decisions: an empirical analysis. Journal of Applied Social Psychology, 16, 686-701.
Simon, R.J. (1967). The jury and the defense of insanity. Boston: Little, Brown.
Slobogin, C. (1985). The guilty but mentally ill verdict: An idea whose time should not have come. The George Washington Law Review, 53, 494-527.
Steadman, H., Keitner, L., Braff, J., & Arvanities, T. (1983). Factors associated with a successful insanity plea. American Journal of Psychiatry, 140, 401-405.
Complete details about the questions are available by request to the first author. Here we present only a terse description of each of the eight predictor variables.
Attitude about lenient treatment: Two Likert-type items summed such that a high score indicates a preference for lenient treatment. The correlation between the two items was r = .32
Attitude toward the insanity defense: Four Likert-type items summed so that a high score indicates receptivity rather than opposition to the insanity defense. Cronbach's alpha was .67.
Attitude toward the death penalty: a single item, "are you strongly in favor, somewhat in favor, somewhat opposed, or strongly opposed to the death penalty," with high scores indicating opposition to the death penalty.
Belief regarding expert testimony: Four none-point items summed such that high scores indicate that the defense's expert testimony was believed more than was the prosecution's. Cronbach's alpha was .87.
Assessment of the defendant's mental status: Six Likert-type items summed so that high scores indicate that the participant believed the defendant to be insane. Cronbach's alpha was. 88.
Belief that the defendant could be rehabilitated: Two items summed such that high scores indicate disbelief in the possibility of rehabilitation. The correlation between the two items was .56.
Attitude toward the defense attorney: A single Likert-type item, "defense attorneys have to be watched carefully, since they will use any means they can to get their client acquired," with high scores indicating trust.
Attitude toward the prosecuting attorneys: A single Likert-type item, "prosecuting attorneys have to be watched carefully since they will use any means they can to get convictions," with high scores indicating trust.
By Ronald L. Poulson, Department of Psychology, East Carolina University, 311 Rawl, Greenville, NC 27858-4353 , Ronald L. Braithwaite, School of Public Health, Emory University, Atlanta, GA 30322 , Michael J. Brondino, Department of Psychology, University of South Carolina, Columbia, SC 29208 and Karl L. Wuensch, Department of Psychology, East Carolina University, 104 Rawl, Greenville, NC 27858-4353