The study by Corwin et al. adds to the emerging but limited data on the impact of defendant remorse on sentencing decisions. The authors studied verbal and nonverbal expressions of defendant remorse and whether they were perceived as remorseful by mock jurors. They found that incongruent verbal and nonverbal behavior, as well as mock jurors' willingness to approach emotional situations, resulted in more lenient sentences for defendants. An overarching and as yet unanswered validity concern regarding this line of research in general is whether the use of undergraduate mock jurors reliably models real jurors in actual courtroom settings.
The jury is an entity steeped in tradition. The concept of juries was brought to England by the Norman conquerors, and, “From its inception, the special province of the jury has been the determination of matters of fact” (Ref. 1, p 345). The Sixth Amendment to the U.S. Constitution provides that, in all criminal prosecutions, “the accused shall enjoy…the right to a speedy and public trial, by an impartial jury of the State and district wherein the crime shall have been committed.” The jury has often been described by legal scholars as a foundation of our civil rights and a bastion against tyranny and oppression by government. The 19th century attorney Lysander Spooner added that juries not only must judge the case facts and determine the law, but they must also fathom the moral intent of the accused.2
Researchers have investigated the functioning of juries for decades using a variety of approaches and trying to capture the trial process with experimental paradigms has proved challenging.3 Of note, Weiten and Diamond4 identified six threats to the external validity of jury simulation research, including lack of jury deliberation, credibility problems with role playing, and differences in consequences for real and mock jurors. We now turn to the college student mock juror study by Corwin et al.,5 in which they uncontroversially stated at the outset that defendant remorse is generally accepted as a mitigating factor in capital murder sentencing. In this study, the researchers addressed verbal and nonverbal expressions of defendant remorse that were perceived as remorseful by mock jurors. Also considered was whether need for affect (NFA)—the motivation to approach or avoid emotion-inducing situations—influences sentencing decisions by the study participants.6 The dependent variables were sentencing recommendations in a capital murder trial, either life in prison without parole or the death penalty, and the perceived remorse of the defendant. Consistent with some earlier research studies, Corwin el al.5 found that nonverbal behaviors prevail over verbal behaviors as indicators of remorse.7 In addition, incongruent verbal and nonverbal behavior and a willingness to approach emotional situations led to more lenient defendant sentencing.5 These authors have undertaken an important area of research, and such efforts to develop a deeper understanding of how juries function is laudable. Undoubtedly, jury decisions can have a profound impact on individuals and society.
However, whether undergraduate mock jurors reliably model real jurors in actual courtroom settings remains undetermined.3 While the work by Corwin and colleagues is thought-provoking and interesting, we nonetheless caution that its methodology does not resemble the operations of real juries (ecological validity), and this methodological disconnect prevents the generalization of these results to real-world trials (external validity). Although a lack of ecological validity does not necessarily preclude the external validity of a research project, in this particular study, we believe that neither ecological nor external validity has been achieved, given the inherent limitations of its design. The following sections will outline some of the reasons that we have taken this position.
College Students as Mock Jurors
The participants in the present study were 206 students in Introductory Psychology, mostly white and female, with a mean age of 18.7 years. These students, many of whom were freshmen at the time of their participation in the study, had just embarked on the stage of life when older adolescents commence the process of separating from the family system and moving toward independence. Developmentally, the tasks of establishing an identity and negotiating intimacy were at the forefront for them.8 This population, chosen by the authors, does not represent the more diverse, mature, community-based populations seen in actual juries. Bermant et al.9 pointed out that if the research study characteristics do not reasonably match the actual problem being studied, then “structural verisimilitude” is lacking and the practicality of the findings are limited. Furthermore, research has indicated that age affects jurors' attitudes toward crime. For instance, Higgins et al.10 investigated the effects of mock jurors' age (younger versus older) on decision-making and found that older jurors were more certain of their verdicts and saw the defendant as more responsible for his or her acts than did younger jurors.
Likewise, Bornstein11 pointed out ecological validity problems in jury-simulation studies, encompassing the composition of mock juror samples (undergraduates versus representative community adults), the research setting (laboratory versus courtroom), the trial medium (written summaries, transcripts, and audio- or video-taped simulations or live ones), and the presence or absence of deliberation. Bornstein did not find significant differences between students and community adult samples in most of the studies reviewed. However, McCabe and Krauss,12 in a broadly cited psycholegal research meta-analysis, noted that the absence of definite evidence of the differences between undergraduate and more representative samples of mock jurors should not be taken as evidence of the absence of differences. In evaluating the derivation of these differences, they used a two-stage research approach. In the first stage, they used undergraduates, and in the second stage they tested a more representative juror sample. They found that college students were three times less likely to favor civil commitment of sexually violent predators after bias correction intervention (having the defendant's attorney acknowledge mock juror's negative reactions to his client) than if they had not been exposed to the intervention. This difference was not seen in the more representative sample of mock jurors in the second stage of the study, suggesting that college-aged mock jurors were more readily influenced by forces beyond the evidence and facts of the case.
The neurobiological immaturity of adolescents as mock jurors in research studies, like those enlisted in the study by Corwin et al.,5 is another limitation deserving of mention. Research findings now show that the brain maturation process extends well into the third decade of life. In a methodologically elegant magnetic resonance imaging (MRI)-based prospective longitudinal study by Gogtay et al.,13 the dynamic progression of human cortical brain maturation in healthy children and adolescents was studied. After assessing pre- and postpubertal samples over a 10-year period, the authors found that the prefrontal cortex, which is responsible for reasoning and decision-making, is one of the last portions of the brain to mature. This brings into question how much credence can be given to a research paradigm that substitutes the cognitive functioning of primarily teenaged mock jurors for that of the average community juror, who is 43 years of age.14
The Simulated Trial Setting and the Portrayal of Mock Remorse by Actors
Corwin et al.5 employed four five-minute-long courtroom simulation videos to depict the courtroom setting for analysis by mock jurors. The readers of this study and similar ones are faced with the naturally arising question of whether having mock juror research participants observe a brief video clip of an actor portraying different verbal and nonverbal expressions of remorse truly simulates what jurors perceive and assimilate during their observations of a real defendant during a capital murder trial that may last weeks or longer. In the latter, jurors have the advantage of observing the defendant day in and day out for hours at a time as he responds to the spectrum of courtroom personnel and proceedings (e.g., testimony of prosecution and defense witnesses, attorney motions, sidebar discussions, and audience members).
Further, can an actor playing a person convicted of homicide and awaiting sentencing genuinely capture the complexity of emotions and behaviors related to remorse that an actual defendant in this setting would display? And are college students psychologically equipped, given their limited life experience, to determine what indicators of remorse are reliable and trustworthy? In defense of Corwin et al.5 the existing evidence shows that neither law enforcement officials nor college students are particularly adept at detecting nonverbal deception and that prisoners may be better at it than both of these groups.15 Admittedly, the detection of deception (and for that matter acting as though one has remorse) is a difficult task.
Ten Brinke et al.16 conducted the first study that investigated the discrepancies between true and false remorse. They described detailed facial, verbal, and body language cues to emotional deception and the characteristics of so-called crocodile tears. According to the authors' review, liars were more likely to speak slowly, to be more hesitant, and to use fewer first-person pronouns. In addition, when compared with those persons expressing genuine remorse, falsified remorse was associated with a greater range of emotion and abrupt transitions between positive and negative facial expressions.
In this vein, Klaver et al.,17 in their study on psychopathic offenders and nonverbal indicators of deception, analyzed videotapes from 45 male offenders and suggested that deception in offenders was probably different from that seen in the general population. According to these authors, psychopathic offenders had higher rates of head movement but appeared to be less anxious or nervous while lying when compared with the general population. For the sake of argument, we will presume that actors more closely resemble the general population than do criminal offenders or psychopaths and that their rendition of remorse in research studies like the current work by Corwin et al.5 therefore would not necessarily approximate that of murder defendants.
In short, we find that Corwin et al.5 entered problematic methodological waters when they attempted to simulate juror assessment of criminal defendant remorse by substituting for the defendant an actor appearing briefly on a videotape. Perhaps their results will provide some insight into general indicators of deception and remorse perception. However, we believe that designs utilizing exposure to in vivo criminal defendant samples in real trials are far more preferable and that the courts are becoming more friendly to the idea of psycholegal research within their walls.3
The authors teasingly stated that they had collected data on the participants' political affiliation in this study, yet they did not report on how this variable may have affected the study participants' sentencing recommendations. This variable is of import that cannot be ignored when attempting to understand jury decisions on sentencing capital murder trials. In fact, according to a 2004 Gallup Poll of 6,500 adults, three quarters of those who identified themselves as political conservatives were in favor of the death penalty, whereas only about one half of the political liberals were similarly inclined.18
Congruent Verbal and Nonverbal Remorse and Blameworthiness
It is worthy of mention that the congruence between verbal and nonverbal remorse and blameworthiness is not necessarily as straightforward in jury decisions as might be assumed. For instance, Jehle et al.19 studied the effects of a defendant's remorse when accompanied by other accounts such as apologies, excuses, justifications, and denials. They found that denials were more often associated with defendants' being found not guilty than were shows of remorse or remorselessness, even though denial was rated as the least believable among the accounts. This finding calls into question whether defendants necessarily stand to benefit by giving a remorseful apology. It may result in a paradoxical jury response by not helping them to appear deserving of a lighter punishment. The results obtained by Corwin et al.5 may be consistent with this phenomenon, in that they found inconsistent verbal and nonverbal behavior to be associated with lighter sentences. They surmised that defendants exhibiting too much remorse (both verbally and nonverbally) could appear more guilty and thus deserving of the severest punishment. These findings are consistent with evidence that perceptions of greater responsibility are associated with an increased likelihood of capital punishment.20
Need for Affect
As part of their study, Corwin et al. utilized the need for affect (NFA) scale developed by Maio and Esse,6 thus allowing them to assess how individual differences in motivation to approach or avoid emotion-inducing situations influenced sentencing decisions. As it turned out, mock jurors more willing to approach emotional situations recommended more lenient sentencing.5 Maio and Esse6 found that the need for affect was related to a variety of other individual affective processes, such as emotional intensity and having more extreme attitudes toward controversial matters. One might therefore posit, contrary to the current findings by Corwin et al.,5 that jurors high in the need for affect would be more inclined to assign capital punishment, since they are more open to strong emotions and controversial opinions. Curiously, in the early developmental research on the need for affect scale, it was reported that people high in NFA were more likely to become involved in a macabre emotion-inducing event (e.g., the death of Princess Diana).6 Thus, a high NFA in that research indicated an emotional affinity for being connected with the death of a luminary, but a high NFA in the present study did not contribute to being in favor of the death sentence for a capital murder defendant. The need for cognition, described by Maio and Esse in the same 2001 study, was found to correlate positively with NFA. That is, feelers (individuals with a higher affinity for emotion-inducing experiences) and thinkers (individuals with a higher need for cognition in their approach to the world) were related in NFA.6 Perhaps had Corwin et al. also looked at need for cognition, they would have shed light on the question of why those jurors with greater NFA showed a preference for assigning a life sentence rather than the death penalty.
The decision-making process surrounding sentencing for a juror serving in a capital murder trial is complicated. Individual, familial, social, political, religious, moral, and group dynamic influences contribute in various degrees to the ultimate outcome. Nevertheless, this study is a contribution in the important area of jury research, an area of scientific inquiry that ideally will lead to the promotion of justice for criminal defendants.
Disclosures of financial or other potential conflicts of interest: None.
- © 2012 American Academy of Psychiatry and the Law