Legal authorities should be as objective as possible when sentencing a defendant who has been found guilty of committing a crime. However, current research has demonstrated several factors which influence jury decision making in the court room. For example, the gender and ethnicity of jurors (e.g., Golding, Bradshaw, Dunlap, & Hodell, 2007; Perez, Hosch, Ponder, & Trejo, 1993) and defendants (e.g., Demuth & Steffensmeier, 2004; Mazzella & Feingold, 1994) have been shown to affect jury decision making. These factors seriously violate the goal of having a fair and objective trial, irrespective of the race and gender of jurors and defendants. Furthermore, some researchers have argued that when jurors empathize with a defendant they will be less likely to condemn him/her to death and more likely to sentence the defendant to life imprisonment (Garvey, 2000), suggesting that empathy is a significant factor influencing jury decision making. Therefore, this study will focus on the concept of empathy in relation to jury decision making. More specifically, how individual differences in empathy influence mock juror’s sentencing decision when a defendant is found guilty.

Earlier research has demonstrated that jurors who are presented with statements that are aimed at inducing empathy among the jurors report changed perceptions of the defendant (Plumm & Terrance, 2009). Also, mock jurors in an induced-empathy condition found the defendant to be less guilty and less responsible compared to mock jurors in a control condition (Haegerich & Bottoms, 2000). However, there is a lack of research investigating whether jurors’ individual characteristics (e.g., empathy) affect their decisions toward an identical description of the defendant. Therefore, the current study aimed to explore the relationship between empathy and stringency of punishment in mock jurors. The main hypothesis for this study is that mock jurors high in empathy will tend to punish the defendant more leniently compared to mock jurors low in empathy as measured by the Interpersonal Reactivity Index (Cliffordson, 2002; Davis, 1983). In other words, there is a suggested negative relationship between empathy and stringency of punishment in mock jurors.

Jury Decision Making

In the United States alone, more than 150,000 jury trials take place each year (Landsman, 1999; Mize, Hannaford-Agor, & Waters, 2007). Hundreds of thousands of U.S. citizens serve on juries every year, and a significant proportion of the U.S. population will serve on juries at some time in their lives (Devine, Clayton, Dunford, Seying, & Pryce, 2001). Additionally, several countries use a similar system as the U.S. in which the jury is made up entirely of laypersons (e.g., Australia, Canada, UK, New Zealand, Russia, and Spain). Other countries (e.g., France, Germany, Italy, Japan, Poland, and Sweden) use a combined system where the jury is composed of both laypersons and judges educated in the law (Devine, 2012). According to Abramson (1994) the ideological difference between these two systems can be traced back to the way one perceives the role of the jury - either as a representative body where jurors are thought to vote according to their narrow group loyalties or as a deliberative body where the focus is on argument and persuasion to reach a unanimous verdict. In other words, juries are not unique to the U.S. and the amount of jury trials taking place outside the United States is certainly not insignificant. As jury composition in the U.S. differs from several other countries, it is important to increase the investigation of how particular juror characteristics may be influencing decision making in the courtroom in these countries. For instance, in Sweden, both jurors (called lay judges) and judges are equally eligible to decide on a final verdict and sentence length for a defendant (The Swedish courts, 2014, December 20). Increased knowledge about what factors are influencing juror decision making could lead not only to a greater understanding of the attributes of specific sentences but also to interventions that can facilitate the right to a more objective and fair trial for all people. One of the factors that have been shown to influence jury decision making is empathy.

Empathy and Sympathy

Empathy is the ability of one person to put oneself into the position of another person (Davis, 1983). Despite the distinctiveness of the concept of empathy, it is very often confused with other affective experiences, such as sympathy. However, empathy (e.g., “I understand what the defendant is going through”) and sympathy (e.g., “I feel sorry for the defendant”) are distinct (Decety & Michalska, 2010). Sympathy is the ability of a person to respond to the emotions of another person and to have feelings of sorrow and concern for others (Eisenberg & Miller, 1987). Similarly, Clark (2010) defines sympathy as more of a feeling of concern for others than a feeling of genuine psychological understanding for the other person. Thus, sympathy is conceptualized more as an evaluation and (sometimes) succeeding response to a person’s needs than the experience of the exact same emotional state. In contrast, empathy is viewed as a more active process that involves the deliberate action of taking all appropriate steps to go outside of one’s personal self and go into the experience of another individual (Davis, 1996). Additionally, empathy has been hypothesized to play a role in the decision making process among jurors and to influence their verdicts (Garvey, 2000).

Furthermore, according to Davis (1983) the concept of empathy is best described as multidimensional, consisting of four distinct constructs. These are perspective taking, fantasy, empathic concern, and personal distress. Starting with perspective taking, this element of empathy involves the tendency to put oneself into the psychological situation of another. People high in perspective taking are also believed to be good at anticipating other peoples’ reactions which make rewarding interpersonal relationships possible (Davis, Conklin, Smith, & Luce, 1996). The fantasy element of empathy instead involves the ability to put oneself into the feelings and behaviors of imaginary characters in movies, books, and plays. People high in fantasy will tend to empathize easily with imaginary characters and this element of empathy has been shown to display a relationship with measures of emotionality (Davis, 1983). The empathic concern element of empathy embraces feelings of concern for others and of the four different aspects of empathy is the construct most closely related to sympathy. This involves feelings of uneasiness about other people’s misfortunes and troubles together with a concern for their lack of wellbeing. Finally, personal distress incorporates self-oriented experiences and feelings of anxiety and nervousness in tense interpersonal situations. People high in personal distress tend to experience great tension and anxiety in situations that involve people in emotional or physical need. The two concepts of perspective taking and empathic concern are the most closely related to this study. Research has shown that jurors who put themselves into the defendant’s position (i.e., perspective taking) often change their perception of the defendant (Haegerich & Bottoms, 2000). Additionally, Davis (1983) found a significant intercorrelation between the two concepts of perspective taking and empathic concern, indicating that they are closely related to each other.

Research about empathy in the courtroom has largely focused on the hypothesis that induced empathy for a defendant would lead to less guilty verdicts by the jurors. For example, the study by Haegerich and Bottoms (2000) investigated whether induced empathy among mock jurors affected their decisions in a child sexual assault case. Here it was found that jurors who were asked to take the defendant’s perspective demonstrated more empathy for the defendant than jurors who were not asked to take the defendant’s perspective. More importantly, jurors in the perspective taking condition not only demonstrated more empathy for the defendant but also found the defendant less guilty and less responsible for the crime. They were also more likely to consider abuse to be a mitigating factor in the homicide. Interestingly, thirty percent of the jurors in the empathy condition also indicated that the atrocities experienced by the defendant justified no punishment for the defendant. Thus, this study clearly suggests that different levels of empathy may play a significant role in the decision making process among jurors, and that perspective taking is very important for the ability to feel empathy in criminal cases.

Furthermore, a study by Plumm and Terrance (2009) had the aim of evaluating the impact of induced-empathy in a case involving a woman who killed her abusive partner and entered a plea of not guilty by reason of self-defense. Here it was demonstrated that empathy played a role in the jury’s perception of the woman. Specifically, participants in a no empathy condition rated the defendant as more mentally unstable compared to participants in an empathy present condition. This suggests that induced empathy could affect jurors’ perceptions of the defendant, especially of their mental health, and perhaps also their tendency to convict the defendant.

In order to improve the understanding of how jurors make decisions in capital cases, the Capital Jury Project (CJP) was initiated as a nationwide interview effort involving many jurors in the U.S. (Bowers, 1995). Using the data from this project, Garvey (2000) asked 187 jurors if they felt sympathy for a defendant in different cases that they had been involved in. It was established that half of the jurors responded that they indeed did so. Further, it was also found that jurors who thought that the defendant was severely abused as a child or had obtained a misfortune in life were more likely to have felt sympathy for the defendant. This suggests that there are individual differences in how jurors sympathize with defendants and that this difference could, in turn, emanate from their belief that the defendant has lived a rough life. Also, this tendency could arguably be strongly related to the concept of empathic concern from Davis’s (1983) multidimensional approach to empathy, which involves feelings of concern and sympathy for others. Moreover, most people would probably expect that jurors would empathize more with the victim than the defendant when they are deliberating a punishment for that guilty defendant. After all, the victim has presumably not done anything wrong according to the law. However, in another study by Garvey (2000) as part of the CJP, jurors who imagined the victim as a member of his or her own family were also more likely to have imagined being in the defendant’s situation. Garvey (2000) therefore proposes that empathy is not a scarce resource but rather a capacity or quality of character. Thus, a juror who imagined being in the victim’s situation may also tend to imagine being in the defendant’s situation and empathize with both of them. This suggests that perhaps some jurors are more empathetic and will tend to direct that empathy toward both the victim and the defendant. In contrast, Breithaupt (2012) has argued that people tend to choose one side of a dispute and only feel empathy for that side while blocking their empathy for the other side. This implies that jurors would choose to either feel empathy for only the victim or the defendant. An additional study using the data from the CJP reported moderately strong correlations between the perceptions of the jury that the victim had a troubled life (e.g., had alcohol and/or drug problems) and the tendency to choose a life sentence rather than a death sentence for the defendant (Sundby, 2003). This was true even though jurors themselves reported not being affected by the victim’s attributes. Also, juries tended to value victims who played no role in the crime more than victims who engaged in some type of risky behavior related to the crime (since it was easier for the jurors to empathize with an “innocent” victim). More research is needed in order to investigate how these different empathic responses affect verdicts and sentencing decisions in courts. However, because of the limitations of the current study only empathic feelings toward the defendant will be considered.

Aims and Hypotheses

Current research has focused a lot of attention on the strength of external forces and induced empathy on the sentencing decisions among jurors. For example, both Haegerich and Bottoms (2000) and Plumm and Terrance (2009) each manipulated the court case in order to create two different conditions, one low in induced empathy and one high in induced empathy. However, no current study has attempted to study empathy and juror decision making from an individual difference perspective. This is problematic since individuals who are part of a jury in a court are presented with the exact same information in a case. Hence, individual differences could potentially explain different opinions among jurors. The aim of this study is therefore to investigate if individual differences in empathy among the mock jurors are affecting their decisions to the exact same court case. In light of previous research on empathy and juror decision making, three hypotheses are proposed. H1: There is a negative correlation between overall empathy and stringency of punishment among mock jurors. H2: There is a negative correlation between the empathic concept of perspective taking (PT) and stringency of punishment among mock jurors. H3: There is a negative relationship between the empathic concept of empathic concern (EC) and stringency of punishment among mock jurors. Additionally, the differences between law-students and non-law students will be investigated since research has shown that judges (law educated) and jurors (non-law educated) frequently disagree about their verdicts (Eisenberg et al., 2005).



The current study used a correlational design with the above mentioned variables of overall empathy (including all four empathic subscales), perspective taking, and empathic concern, taken from the Interpersonal Reactivity Index (Davis, 1983). Each of these variables’ relationship with stringency of punishment was assessed by the Pearson product-moment correlation coefficient.


In total, 291 participants (123 male, 166 female, and 2 other) were recruited to be part of the study. All participants were university students. Eighty (27.5%) of the participants were students of the Law department and 211 (72.5%) participants were students of other departments. Among the law students, 39 (48.8%) were female and 39 (48.8%) male. The corresponding number among the non-law students was 127 (60.2%) females and 84 (39.8%) males. The age of the participants ranged between 18 to 39 years (M = 23.29, SD = 3.24). In the current study, 36 (12.4%) participants reported having previous experiences of the crime described in the scenario, while the rest, 255 (87.6%) participants did not report having any such experiences. The reason for including both law students and non-law students in the study was made because, in Sweden, the juries in the lower court consist of both one law-educated judge and three non-law-educated lay judges (i.e., jurors). That is, a 25/75% distribution of law educated/non-law educated judges and jurors. The current study received ethical approval through the student supervisor.


Stringency of punishment. In order to measure stringency of punishment, a court case was constructed for the present study where the participants had to decide a length of sentence for a guilty defendant. The sentence length was answered on a scale from 1 (one year imprisonment) to 10 (10 years imprisonment). The length of the sentence scale was decided in accordance to the Swedish law which states that the minimum sentence length for aggravated assault is 1 year imprisonment and the maximum sentence length is 10 years imprisonment.

The hypothetical court case. Michael Carlsson (the defendant) has assaulted the plaintiff by first shoving him to the ground and then, when the plaintiff was lying down on the ground, allocating several kicks to the plaintiff’s head and body. As a result of the assault, the plaintiff has suffered life threatening brain damage and pain and bruises over the left eyebrow. The court found the defendant guilty of aggravated assault. Both Michael Carlsson (the defendant) and the plaintiff were above 18 years of age at the time of the event.

To reduce the risk of subjects empathizing with the victim and not the defendant, as predicted by Breithaupt’s (2012) argument, special care was taken to portray the victim in the crime as anonymously as possible by excluding the victim’s name, his relation to the defendant, and background story. Apart from the description of the crime, some personal information about the defendant was also provided. This personal information had the aim of inducing empathy for the defendant among the mock jurors. It revealed that the defendant had been the victim of bullying in school, that his father left the family very early, and that he had lately been thinking about the meaning of his own life and whether it was really worth for him to continue living.

Validation of test materials. In order to ensure that the personal information that accompanied the court case did in fact induce empathy, a small pilot study was conducted with 24 participants (17 females and seven males) who rated how much they empathized with a defendant who had committed a crime. All participants were students with an age range between approximately 20–35 years. Thirteen participants read a case about Michael (defendant 1) who committed an aggravated assault and was declared guilty by the Swedish court. His personal information explained that he had grown up without his father and also that he had been the victim of bullying in school. In contrast, eleven participants read a case about Marcus (defendant 2) who also committed an aggravated assault and was declared guilty by the Swedish court. His personal information explained that he had been sexually abused by his father from an early age and that he had found it difficult to find any close friends. An independent samples t-test demonstrated that there was no significant difference in the empathy score between defendant 1 (M = 5.00, SD = 2.16) and defendant 2 (M = 4.55, SD = 2.16; t(22) = .51, p > .05, two-tailed). In other words, both cases were quite similar to each other in terms of induced empathy and both cases had a mean score around the midpoint of the empathy scale which was scored from 1 (no empathy) to 10 (much empathy). It was therefore decided to only retain defendant 1 (Michael) for use in the current study. This was done in order to avoid the use of a relative judgment of the two defendants against each other. Having two defendants in a repeated measures design would potentially have encouraged the participants to compare the cases between each other when deciding a punishment, thus potentially influencing their opinions and negatively bias the final results.

Interpersonal Reactivity Index (IRI). The IRI is a 28-item self-report questionnaire consisting of four 7-item subscales, each of which examines a specific aspect of empathy (Davis, 1983). The scale consists of the empathic aspects of perspective taking, fantasy, empathic concern, and personal distress. The perspective taking scale assesses the tendency to put oneself into the situation of another person in one’s everyday life. One sample item from the perspective taking scale is “Before criticizing somebody, I try to imagine how I would feel if I were in their place”. The fantasy scale investigates the tendency to put oneself into the feelings and acts of imaginary characters in books, plays, and movies. One sample item from the fantasy scale is “After seeing a play or a movie, I have felt as though I were one of the characters”. The empathic concern scale measures the tendency to have feelings of warmth, sympathy, and concern for other people around oneself. One sample item from the empathic concern scale is “I often have tender, concerned feelings for people less fortunate than me”. Finally, the personal distress scale assesses one’s own feelings of discomfort and unease in reactions to others’ emotions. One sample item from the personal distress scale is “When I see someone who badly needs help in an emergency, I go to pieces”.

According to Davis (1980), all scales have substantial test-retest reliability as well as satisfactory internal validity. The Swedish version of the IRI was validated by Cliffordson (2002), and demonstrated an acceptable alpha reliability (ranging from .71 to .80), which is similar to the alpha reliability (ranging from .71 to .77) reported by Davis (1983). In the present study, all sub-scales together with the overall empathy had acceptable internal consistency, ranging from α = .70 to α = .84.


The participants were recruited in different locations at various departments at the University. They were informed that their participation was entirely voluntary and that they could be confident that their responses would be kept anonymous and only be used for research purposes. Once the participants had agreed to participate, they were given the questionnaire. As mentioned above, the subjects first read a court case about a defendant named Michael who had been convicted of aggravated assault for whom they had to decide an appropriate sentence length. In the second part of the questionnaire, participants answered the IRI (Davis, 1983). In order to control for the possibility that participants’ personal experiences of the crime included in the crime description to confound their responses, a control question was included at the end of the questionnaire asking subjects “Have you, or anyone in your close family, any experiences of a similar crime that was described in the court case?” Subjects answering yes to this question were excluded from the main analyses due to the risk of their earlier crime experiences influencing their sentencing decisions. After completing the questionnaire, the subjects were thanked for their participation together with a short debriefing about the study’s real purpose.


As displayed in Table 1, all subscales except the personal distress scale reached their absolute maximum score whereas no sub-scale reached its absolute minimum score. The sentence length scale did reach both its absolute maximum and minimum score, with the mean value located approximately in the middle of the scale.

Table 1

Means, standard deviations, maximum scores, minimum scores, intercorrelations, and coefficient alphas for overall empathy, subscales, and sentence length.

Overall empathy (OE) Perspective taking (PT) Empathic concern (EC) Fantasy (FS) Personal distress (PD) Sentence length (SL)

Maximum 125 35 35 35 32 10
Minimum 61 13 16 10 8 1
OE .84 .70** .96** .95** .71** -.15*
PT .54** .70 .54** .32** -.11 -.22**
EC .77** .40** .77 .59** .34** -.22**
FS .78** .24** .46** .80 .36** -.06
PD .57** -.08 .26** .28** .76 .01
SL -.14* -.18** -.19** -.05 .01 1.00

Note. Intercorrelations are presented below the diagonal, and attenuation corrected intercorrelations are presented above the diagonal. Coefficient alphas are presented in boldface along the diagonal.

aPossible range, 28–140, bPossible range, 7–35, cPossible range, 1–10. *p < .05, **p < .01.

According to Davis (1983), among the empathic subscales (i.e., perspective taking, empathic concern, fantasy, & personal distress), the perspective taking scale and the empathic concern scale had a positive relationship between each other as well as the fantasy scale and the empathic concern scale. As can be seen in Table 1, this was also true for the present study with perspective taking and empathic concern, and fantasy and empathic concern demonstrating the highest correlations between each other.

Overall Empathy and Subscales

The association between overall empathy and subscales (i.e., perspective taking & empathic concern) with stringency of punishment was examined using the Pearson correlation coefficient. There was a small negative correlation between overall empathy and stringency of punishment, r = -.14, n = 255, p = .022, with high levels of overall empathy associated with a shorter sentence length for the defendant. There was also a small negative correlation between the empathic concept of perspective taking (PT) and stringency of punishment, r = -.18, n = 255, p = .003, with high levels of perspective taking correlated with a shorter sentence length for the defendant. Finally, there was a small negative correlation between the empathic concept of empathic concern (EC) and stringency of punishment, r = -.19, n = 255, p = .002, with high levels of empathic concern associated with a shorter sentence length for the defendant. Thus, all three hypotheses of a negative correlation between overall empathy (i.e., hypothesis 1) and subscales (i.e., hypotheses 2 & 3) with stringency of punishment were supported.

Explained Variance of Stringency of Punishment

In order to investigate the explained variance of stringency of punishment by the four empathic subscales, a standard multiple regression was conducted to examine their combined contribution to the explained variance in sentence length. Preliminary analyses were performed to ensure no violation of the assumption of multicollinearity. The total variance explained by the model as a whole was 5.3%, F(4, 250) = 3.47, p = .009. As displayed in Table 2, only empathic concern (β = -.17, p = .027) made a significant unique contribution to explaining sentence length, while perspective taking (β = -.13, p > .05), fantasy (β = .051, p > .05), and personal distress (β = .024, p > .05) all did not. Thus, in this study, empathic concern was the unique and only significant predictor of sentence length.

Table 2

Standard multiple regression of each of the subscales on sentence length.

Subscale B SE B β

Perspective taking (PT) -.082 .045 -.13
Fantasy (FS) .024 .034 .051
Empathic concern (EC) -.11 .047 -.17*
Personal distress (PD) .013 .037 .024

Note. *p < .05.

Law Students vs. Non-law Students

The decision to include both law students and non-law students in the study was made in order to increase the similarity of an authentic Swedish jury composition (i.e., a 25/75% law educated/non-law educated distribution) and thus, enhance the ecological validity of the study. Finally, in order to investigate if there were any differences between law students and non-law students in sentencing decision and overall empathy, two independent samples t-tests were conducted. The first demonstrated that there was no significant difference in sentencing decision between law students (M = 4.62, SD = 2.34) and non-law students (M = 4.53, SD = 2.47; t(253) = .27, p > .05, two-tailed). Further, there was no significant difference in overall empathy between law students (M = 95.83, SD = 12.11) and non-law students (M = 98.32, SD = 11.20; t(253) = -1.55, p > .05, two-tailed). This indicates that the law students and non-law students were rather similar to each other in terms of both their sentencing decisions and overall empathy scores.


The present study had the main objective of investigating the relationship between the level of empathy in mock jurors and their sentencing decision toward a guilty defendant. There was a significant negative correlation between overall empathy and stringency of punishment among the mock jurors. This supported the hypothesis that increasing levels of empathy were associated with decreasing levels of stringency of punishment. The current findings also support the conclusions from the studies by Haegerich and Bottoms (2000) and Plumm and Terrance (2009), which both concluded that empathy plays a role when it comes to the decision making of jurors in court cases. In addition, it was found that increasing levels in each of the empathic subscales of perspective taking and empathic concern were associated with decreasing levels of stringency of punishment. These findings relate to the study by Sundby (2003) which found a relationship between the perceptions of the jury that the victim had a troubled life (e.g., had alcohol and/or drug problems) and the tendency to choose a life sentence rather than a death sentence for the defendant. The awareness of others’ misfortunes and concern for other people’s problems is a significant attribute of the concept of empathic concern. This, together with the fact that the personal information that was included in the court case included information about the defendant explaining that he had lived a tough life, provides findings in line with the previous research.

Moreover, in the present study, 5.3% of the variability in the sentencing decision was explained by the variability in empathy among the participants. While this is a rather low R squared, this, could be due to the short court case and limited personal information about the defendant as compared to a real court case. Interestingly, no differences in overall empathy or sentencing decision were found between law-students and non-law students. One explanation for this might be that the law-students were still studying and lacked real court training. This could have affected their empathy scores and sentencing decision to more resemble non-law students. Had the study been conducted with real jurors and judges the results may potentially have been different. Supporting this argument, the mentioned research by Eisenberg et al. (2005) showed that judges (law educated) and jurors (non-law educated) frequently disagreed about their verdicts. Whether they would also differ in their sentencing decision could possibly be examined in future research.

The rather low correlation between overall empathy and stringency of punishment in the current study could further have several explanations. As was mentioned above, the court case and the personal information that was presented to the participants were very short and limited. This could perhaps have led to the participants not developing particularly strong emotions for the defendant and thus not empathizing with him very much. Had the information about the defendant’s life been more comprehensive, the correlation might have been significantly stronger between empathy and stringency of punishment. Also, no information was provided of what led the defendant to commit the crime in the first place. If this information had been portrayed the defendant as “innocent” in terms of not initiating the fight, then participants may have empathized more with the defendant. This is also in line with the study by Sundby (2003) where juries tended to value victims who played no role in the crime over those who engaged in some type of risky behavior related to the crime.

In addition, when jurors are making their decisions in a real court they are often involved in some kind of deliberation process. This process probably has an influence on the final verdict decision. For example, a study by Patry (2008) demonstrated that jurors were much more likely to fall prey to the attractiveness leniency bias (assigning a more lenient punishment to an attractive defendant) when they did not deliberate as compared to when they did deliberate. The social process of juror decision making is something that was intentionally excluded from the current study. Here the participants made their sentencing decision completely alone and anonymous. The disadvantage is that this does not correspond well with the real world scenario of deliberating jurors. It would therefore be interesting for future researchers to investigate the role of empathy in relation to the jury deliberation process. This will further improve the generalisability of the results and aid the practical utility of the findings.

Another possible limitation of the current study relates to the fact that the participants all came from a rather homogenous background. All participants were highly educated and rather young. This is a problem if one wants to generalize the results to the general population. Students are more educated than the general population, which means that they might be more rational when it comes to making moral decisions as well as having a more liberal view about prison as a form of punishment. Also, although most people agreed to participate in the study, some people refused to do so. It could be hypothesized that those who refused to take part in the study may differ in some systematical way compared to those who participated. This could potentially have affected the relationship between empathy and stringency of punishment. One additional major limitation of the current study relates to the fact that the participants were not actual jurors but merely students although some will likely be jurors in the future. This is problematic since real jurors likely have some practical experience of judging court cases and therefore may be evaluating the evidence more objectively and justly (whether this really is the case could be debated). Students on the other hand, might be more influenced by the irrelevant factors of the case and thus less objective and unbiased in their sentencing decisions. By only using mock jurors, the ecological validity and supposedly the external validity of the study is weakened. Consequently, Devine et al. (2001) have reported that the use of mock jurors have been criticized by many judges for lacking realism. At the same time, no other approach is capable of controlling influential extraneous variables that are related to the case as effectively as is possible with the use of mock jurors. Finally, despite having found a statistically significant association between empathy and stringency of punishment, the implications of the findings are marginal due to the small magnitude of explained variance.

Practical Implications

The results of the present study could conceivably have some practical implications in real court situations. For example, by including personality-related questions (e.g., items from the IRI) on screening questionnaires for potential jurors, the judges and attorneys would have information about the individual characteristics of the jurors. This information could be used to construct juries that are more balanced in terms of including jurors with both high and low levels of the target characteristic (i.e., empathy).

Theoretically, the study builds on the already established research that has related individual differences in juror characteristics (e.g., gender and ethnicity) to their sentencing decisions. The finding in the present study that empathy plays a role (although a very small one) in the sentencing decision among jurors and thus should be included on the list of individual differences affecting jury decision making is both intriguing and important for legal authorities to be aware of. This is especially relevant in a Swedish legal context since both judges and jurors in this country make sentencing decisions.

Future Research

It might be fruitful to combine the design of the present study with two different empathy conditions – one low in empathy and one high in empathy. This would make it possible to investigate potential interaction effects between individual differences in empathy and the level of empathy induced in the case. For instance, using the IRI to measure individual differences in empathy would make it feasible to examine if people low or high in empathy show a smaller or larger change in their sentencing decision as a function of the empathic characteristics of the case. Here it could be hypothesized that jurors low in overall empathy would show a smaller difference in their sentencing decision between the low and the high level of induced empathy in the case. However, jurors high in empathy might instead show a larger difference in their sentencing decision between the low and high level of induced empathy in the case, thus supporting an interaction effect.

Also, it would be interesting for future researchers to investigate the role of empathy and its relation to other individual characteristics among the jurors. Combining several different individual characteristics into the same study will lead to more practically useful results which are not constrained inside the walls of our scientific labs. Ideally, engaging real jurors would make the results much more robust and likely to be taken seriously by both the legal authorities and scientific community. The usefulness of our research to improve the legal authority and thus peoples’ lives should always be of paramount priority in psychological research.


The present study demonstrated a significant negative correlation between empathy and stringency of punishment together with a significant amount (although a very small proportion) of explained variance from empathy in the sentencing decisions among mock jurors. In other words, this study builds on the already established research that has highlighted the importance of empathy in jury decision making (e.g., Garvey, 2000; Haegerich & Bottoms, 2000; Plumm & Terrance, 2009; Sundby, 2003). However, it also expanded the current research by showing that individual differences in empathy among mock jurors affected their sentencing decisions. The next decade will hopefully see a clarification and expansion in research of how precisely empathy is related to juror decision making and what can be done to control its impact in the court room.