The ideal of unbiased, rational legal decision-making contrasts with the reality of human cognitive biases. This study explores the influence of cognitive and emotional factors on moral judgments within the legal field. Traditional legal frameworks emphasize rationality, yet abundant evidence demonstrates the pervasive effects of biases in human decision-making. For example, decisions about punishment are often driven by emotional biases, and people tend to overestimate the damage caused by intentional harm compared to accidental harm, even when the actual harm is identical. While the law sometimes acknowledges these factors, they often remain implicit or overlooked. This research aims to illuminate the interplay of these factors in legal decision-makers by comparing the moral decisions of criminal judges, attorneys, and a control group. The study focuses on moral evaluation, punishment assignment, and harm assessment of third-party aggressions, examining the influence of (a) information about the transgressor's mental state, (b) gruesome language (GL) in harm descriptions, and (c) ongoing physiological states. These three factors are crucial because inferences about mental states are central to moral and legal deliberations, emotionally arousing elements like GL can bias decisions, and ongoing physiological states also shape decision processes. While legal decision-makers are not immune to implicit biases, the extent to which their expertise mitigates these biases remains an open question. Previous research suggests that judges are susceptible to various biases, including those related to anchoring, framing, egocentric bias, and implicit racial biases, yet they possess the ability to suppress these biases when consciously motivated. The current research seeks to expand this understanding by exploring the influence of the aforementioned factors on the moral decision-making of legal experts.

The literature extensively documents the influence of cognitive and emotional biases on human decision-making, particularly in moral and legal contexts. Studies highlight the impact of emotional biases on punishment decisions, the tendency to overestimate damage from intentional harm, and the effects of gruesome language on judgments. Prior research has investigated the cognitive biases present in judicial decision-making, such as anchoring biases in damage awards and sentencing, and the presence of implicit racial biases. Moreover, the literature establishes a link between moral and legal decision-making, with similar neural correlates suggesting overlapping cognitive processes. However, the influence of intentionality, emotionally charged language, and physiological states on the decisions of legal experts remains under-researched. This study directly addresses this gap by comparing the responses of legal professionals to those of a control group.

This study employed a quantitative research design comparing the moral judgments of three groups: 45 criminal judges, 60 criminal attorneys, and 64 community members (control group). The researchers used a modified computerized version of a task that measured moral evaluation, punishment assignment, and harm assessment. Participants read 24 scenarios involving a protagonist inflicting harm on a victim, varying by harm type (property damage, physical harm, death), intentionality (accidental or intentional), and language (gruesome or plain). After each scenario, participants rated the moral adequacy of the action, the deserved punishment, and the harm severity using a 9-point Likert scale. A subsample of 86 participants (30 attorneys, 27 controls, and 29 judges) also underwent electrocardiogram (ECG) recordings to measure heart rate variability (HRV) as an indicator of physiological arousal during the task. Executive functioning (EF) was assessed in the subsample using the INECO frontal screening (IFS) battery. The data were analyzed using mixed ANOVAs to examine the interactions between group, language, and intentionality, with age and years of education as covariates. Regression models examined the association of EFs, HRV, and age with moral judgments.

The findings revealed that information about the transgressor's mental state significantly influenced moral decision-making across all groups, with intentional harms consistently judged as morally worse, receiving harsher punishments, and being deemed more damaging than accidental harms. However, judges and attorneys assigned significantly less punishment and harm severity to accidental harms compared to the control group, showing less bias towards the accidental condition. Gruesome language (GL) did not bias judges' and attorneys' decisions, unlike the control group, indicating that legal expertise conferred immunity to this type of language bias on moral decision-making, particularly impacting morality ratings rather than punishment or harm severity ratings. This effect of GL was reflected in significant differences in morality ratings between the control group and the legal professional groups. Interestingly, judges' physiological signals (HRV) showed no association with decision patterns, unlike the attorneys (marginally significant) and controls (fully significant). Executive functions (EFs) significantly predicted harm severity ratings, suggesting their role in moral judgment. The study also included analyses for different types of crimes (death, property damage, and physical harm) and years of experience in criminal law, but these factors did not fully explain the group differences observed.

The results support the hypothesis that legal expertise can attenuate some pervasive biases in moral decision-making. The finding that judges and attorneys were less influenced by gruesome language and physiological arousal suggests that professional training and experience can reduce the impact of emotional influences on legal judgments. While all groups exhibited the "harm-magnification effect," overestimating damage from intentional harm, legal professionals demonstrated improved discernment regarding the intentionality of accidental harms, resulting in fairer punishment assignments. The lack of association between judges' physiological signals and their decisions suggests a higher level of emotional detachment during judgment, perhaps due to the nature of their role and experience. The study highlights a partial discrepancy between actual decision-making and the legal system's underlying assumptions of pure rationality. The findings have implications for the use of juries in legal systems, suggesting the potential for greater bias in non-expert decision-makers. However, the harm-magnification effect remains a challenge, emphasizing the need to account for it in legal settings.

This study is the first to experimentally compare moral decision-making across judges, attorneys, and controls concerning mental state information, gruesome language, and physiological states. Legal professionals show less bias in punishment and harm severity ratings for accidental harm and are less influenced by language or arousal. While the "harm-magnification effect" persists, legal expertise improves intentionality detection for accidental harms, possibly leading to fairer judgments. Future research should explore how specific aspects of legal training and experience contribute to bias reduction and investigate the neural mechanisms underlying these differences. The implications for the use of juries and the design of fairer legal systems are significant.

The study's limitations include the cross-sectional nature of the design, limiting inferences about causal relationships between legal expertise and bias reduction. The sample, while large, might not fully generalize to all legal professionals or populations. The reliance on self-reported data for years of experience and education may introduce bias. The use of hypothetical scenarios instead of real-world cases could limit ecological validity. Future longitudinal studies with more diverse samples and real-world data could enhance the findings' generalizability.