KEY WORDS: emotion, empathy, affect control, mock jurors, legal decisions, criminal identity

INTRODUCTION
Research associated with affect control theory has demonstrated the influence of a social actor's emotion display on perceptions of the social actor's identity (Heise 1979; Smith-Lovin and Heise 1988; Smith-Lovin 1990). Researchers have demonstrated the applicability of the theory in a legal context by showing the influence of victim and criminal emotion displays on perceptions of the victim and criminal. These perceptions, in turn, affect sentence decision making (Robinson, Smith-Lovin and Tsoudis 1994; Tsoudis and Smith-Lovin 1998).

Similarly to this past research, this study explores affect control theory in the legal context; however, this study adds the significance of an empathy variable. Empathy is predicted to influence legal decisions as a mediating factor between emotion displays and perceptions of the criminal and victim. The potential utility of an expanded affect control theory is explored in addition to the practical significance of the study. Past studies have not focused on the observer's (juror's) feelings, in other words, empathy for the victim and the criminal. The emotion displays of the actors may influence the observer's feelings, in turn, affecting the observer's judgment of the criminal.

AFFECT CONTROL THEORY
Affect control theory explains the relations among emotion display, identity, and behavior in social interactions (Heise 1979; Smith-Lovin and Heise 1988). In affect control theory, the social event is the unit of analysis, with each event composed of an actor, the behavior of the actor, and an object of the behavior. In this study the theory is studied in conjunction with a criminal case, in which the actor is the criminal, the behavior is the alleged criminal act, and the object of the behavior is the victim.

According to the theory, the actor and object both have identities, which they maintain throughout their social interactions. A social actor displays an emotion (e.g, sadness, unconcern, happiness) after engaging in a behavior (e.g. hitting, giving, killing). The object of behavior (the victim in this study) also displays an emotion. From this emotion display, an observer infers the actor's (and object's) identity. Identities are linked with expectations of behaviors associated with the identity (Robinson et al. 1994). Based on perceptions of the actor's and object's identities, observers of the behavior make judgments.

Emotion displays, identities, and behaviors each have social meanings, which are generated from the culture (Smith-Lovin and Heise 1988; Heise 1969). They are measured on a semantic differential scale on three dimensions: Evaluation(E), Potency(P) and Activity(A) (EPA scale) (MacKinnon 1994). This study focuses on the evaluation dimension (the continuum of goodness) as relevant to the context of the criminal trial. The evaluation dimension has "good-nice" and "bad-awful" as the end-points of the continuum, which ranges from -4.0 to +4.0 (MacKinnon 1994). These three dimensions are measured on a scale of -4.0 to +4.0 (Osgood 1957, 1975).

According to affect control theory, identities, emotions, and behaviors are parallel to each other on the evaluation dimension (Heise 1979). For example, an individual with a positive identity generally engages in positive behaviors. A mother, high on positive identity, engages in positive behaviors such as caring and encouraging; in other words, the positive identity of mother is expected to engage in positive behaviors. An individual with a negative identity engages in negative behaviors. For example, a criminal, negative in identity, engages in negative behaviors such as stealing and lying. In affect control theory, behaviors confirm or disconfirm the social actor's identity. For example, a positive behavior (such as caring) confirms a positive identity of the mother, while a negative behavior (such as hurting) disconfirms this positive identity.

Emotion displays serve as signals of behavioral disconfirmation or confirmation of identity for either positive or negative behaviors. Positive emotions (e.g., happiness) from the actor, after a positive behavior (e.g., helping) indicate confirmation of a positive identity (Smith-Lovin 1990). For example, an individual after donating money to a charity (positive behavior) may display pleasure (a positive emotion). An emotion display of resentment (negative emotion) after donating money disconfirms a positive identity.

Similarly, the emotion display after a negative behavior also signals identity confirmation or disconfirmation for the actor. A positive emotion after a negative behavior indicates a negative identity for the actor; a negative emotion indicates a more positive identity. For example, an individual, after hitting a child, may display sadness (negative emotion), disconfirming a negative identity; while an emotion display of happiness confirms a negative identity.

Application of Affect Control Theory in a Criminal Trial
In a criminal trial, the emotion displayed by the criminal can signal to observers whether the purported crime confirms or disconfirms the criminal's fundamental identity as a "bad" person (Robinson et al. 1994). If the criminal looks sad, this emotion display indicates that the described criminal behavior confirms a positive identity, suggesting the criminal considers the negative behavior atypical. On the other hand, an emotional display of unconcerned indicates the criminal behavior is more acceptable to the criminal, confirming that the criminal has a negative identity. The observer (in this study, the mock juror) perceives that the criminal has engaged in this behavior in the past and will continue to engage in the behavior in the future. The more the observer perceives the criminal as negative, the harsher the punishment.

According to affect control theory, the victim's emotion display can produce similar inferences about the victim's identity. A negative emotion display (e.g., sadness) after being victimized shows that the victim has a positive identity and that victimization disconfirmed that identity. This disconfirmation suggests to the observer that the victim did not believe that he/she "deserved" this treatment. If the victim has a positive emotion display (e.g. unconcern) after a criminal act, this suggests that the negative act confirmed a negative identity (Smith-Lovin and Heise 1988). Most victims express a negative emotion (e.g., sadness) after a criminal act because most victims have positive identities. If the victim displays a positive emotion, observers make more negative inferences about the victim's identity because of an inference that such emotions reflect routine involvement in events. The criminal behavior is seen as less negative, resulting in a less harsh punishment.

There has been some empirical work using this theory. Robinson et al. (1994) explored affect control theory predictions regarding emotion display by a criminal and ensuing character assessments of the criminal by mock jurors. A display of sadness led to a more positive evaluation of the criminal and a shorter recommended sentence, as mock jurors inferred the criminal was remorseful, the behavior was atypical, and the criminal was less likely to engage in future criminal behavior. A subsequent study examined the contribution of the victim's emotion displays on mock jurors' judgments about a criminal (Tsoudis and Smith-Lovin 1998). Victim emotion displays similarly affected identity inferences about the victim. When the victim was sad, the participants perceived the victim as more positive and subsequently resulted in a harsher punishment.

The model in Figure 1 shows the relationships between criminal and victim emotion displays, their identities, and the harshness of the criminal's punishment based on studies.

None of these studies have taken into account the observer's (juror's) feelings, in other words, empathy for the victim and the criminal. The emotion displays of the actors may influence the observer's feelings, in turn, affecting the observer's judgment of the criminal. There may be additional mediating variables between perceptions of the actor's identity and punishment for the criminal; for example the criminal's credibility. These variables may also be significant to research in further understanding legal decision-making. Nonetheless, this study specifically focuses on empathy as a mediator between emotion displays and identity.

EMPATHY AND ITS RELATIONSHIP TO AFFECT CONTROL THEORY
Empathy can be defined on numerous dimensions, including identifying with the feelings of another. Empathy motivates people to help others (Berkowitz and Macaulay 1970). An observer who empathizes with a victim may want to help reduce the victim's distress because he or she appreciates what the victim is feeling. Compassion is another dimension of empathy (Smith, Keating and Stotland 1989). For example, an emotion display of sadness from an individual can bring out compassion from an observer. Observers typically react compassionately (attempt to alleviate distress) to expression cues of another person's distress (Tannebaum 1975). The emotion display may influence the level of empathy an observer feels for the emotion displayer, whether it is the victim or the criminal.

Researchers have focused on empathy as a response to a stimulus (e.g., in this study, the victim or criminal; Duan and Hill 1996; Barrett-Lennard 1962; Greenson 1960, 1967; Hoffman 1984; Stotland 1969). The observer's and stimuli's feelings will become similar. For example, a sad victim will evoke sad emotions for the observer. As previously stated, affect control theory postulates parallels among emotions, identities, and behaviors. Similarly, the present study includes the observer's emotional response to the emotion display of the displayer (whether actor or object) as parallel/similar to each other. The emotion display of the actor will influence the empathy experienced by the observer. The greater the emotion display, the more empathy the observer will have for the social actor.

Research looking specifically at the relationship between empathy and legal decision making has found that empathy is an important construct (Deitz and Byrnes 1981). Defense attorneys continuously report that empathy for their client is important for the client's case (Cohen 1961), realizing that they should encourage jurors to consider the case from the criminal's position (Black 1956). According to Archer et al. (1979), mock jurors who empathize with a criminal perceive the criminal's behavior as more lawful and place less responsibility on the criminal for the criminal act. Mock jurors with strong, positive empathy for a criminal were less likely to find the criminal guilty where the criminal was depicted as having stabbed a victim.

Similarly, Deitz and Byrnes (1980, 1981) found a significant association between rape victim empathy scores and recommended sentences for criminals. Respondents who empathized with rape victims gave longer prison terms to the criminal accused of the crime, expressed greater certainty of guilt, and placed less responsibility on the victim than respondents who did not empathize with the victim (Deitz 1980). Chaikin and Darley (1973) found that the higher the distress of the victim, the more responsibility placed on the criminal.

However, these studies focus on empathy as the only independent variable. This study attempts to provide a more complete picture of legal decision making, exploring the interaction among a criminal's emotion display, a victim's emotion display, empathy for both the criminal and victim, and identity perceptions of the criminal and victim.

Furthermore, in affect control theory, emotion displays relate to an observer's perceptions of the social actor's emotions, identity, behaviors, and judgments. Thus, it would seem important for the theory to include other observer variables, for example, variables directly associated with feelings the observer may have for the social actor. As indicated by empathy theory, the observer's feelings about the social actor can have a significant impact on judgments made about the actor. Empathy manifests itself as an emotion experienced by the observer. This empathy, in turn, will influence the observer's perceptions of the social actor.

This focus on the observer can also be related to research on the similarity-leniency hypothesis, describing the similarities between the juror and the criminal. Kassin and Wrightsman (1988) describe the similarity hypothesis as the juror favoring their own kind. When a juror has a defendant and a victim, will the juror favor the individual who is most similar to them? Past research shows that jurors will place more blame and support more punitive punishment towards defendants who are less similar to them (Gleason and Harris 1975; McGowen and King 1982). Perhaps the similarity hypothesis can be further elaborated upon through empathy. A juror may have more empathy for an individual with similar characteristics, in turn, resulting in less punishment.

In a legal context, if a juror (the observer) empathizes with the victim, it is predicted that the criminal will be given harsher punishment. If the juror empathizes with a victim, they may believe that punishing a criminal will help the victim by making the victim feel better that justice has been served. Similarly, if the juror empathizes with the criminal, the punishment can be expected to be less harsh. The juror may perceive the criminal's behavior as more lawful, thus, he/she may empathize with the criminal. Such empathy may lead the juror to focus on the criminal's rehabilitation, and perhaps opt for a more lenient punishment.

Whether this empathy exists for the criminal and/or for the victim, the association between empathy and punishment may be mediated by identity. Empathy for the criminal will influence perceptions of the criminal's identity. The more the observer has empathy for the criminal, the more positive the perceptions of the criminal. Similarly, empathy for the victim will influence perceptions of the victim's identity. The more empathy the observer has for the victim, the more positive the perceptions of the victim. In turn, as in past research, it is hypothesized that the identity perceptions of the criminal and the victim will influence the level of punishment issued by a juror.

Figure 2 shows the association described by Figure 1 with empathy for the criminal and victim added to the model.

HYPOTHESES
The independent variables in this study are the victim's and the criminal's emotion displays. Empathy and identity are also independent variables; however, they are also proposed intervening variables. The dependent variable is the degree of harshness of punishment for the criminal. The two emotion displays used in this study are the same as those mentioned in past affect control theory studies: sadness and unconcern (Robinson et al. 1994; Tsoudis and Smith-Lovin 1998.) An actor perceived as sad is similarly perceived as being "less unconcerned", while an actor perceived as unconcerned is similarly perceived as "less sad."

The following hypotheses were generated (with hypotheses 1,4,5, and 8 drawn from past affect control research focusing on the relations among emotions, identities, and punishment):

METHODS
Design
A 2 (criminal case: shooting, assault) x 2 (emotion of the criminal: sad, unconcern) x 2 (emotion of the victim: sad, unconcern) factorial design was used to test the above hypotheses. Thus, there were eight experimental conditions. The evaluation of sadness, a negative emotion, is -1.47 on the evaluation scale of -4 to +4. The evaluation of relaxed/unconcerned, a positive emotion, is 1.58. A relaxed emotion is positive relative to an emotion of sadness.

Participants
One hundred forty-three undergraduate students at a Southwestern university were recruited from sociology courses and paid ten dollars to participate in the study. Half of the participants were male and half were female. These undergraduates have not been exposed to material on affect control theory or legal decision making in their introductory courses. Furthermore, it is important to note that the participants were randomly assigned to a condition. Conditions are randomly assigned so that there is no effect from pre-existing beliefs, perceptions, etc. Some researchers may be concerned that pre-existing conceptions may result in a juror concluding that the criminal is "playing a game" through emotion displays. The random assignment to conditions balances the effect of outside variables. Affect control research similar to this study has continuously demonstrated that the emotion display manipulated in the study does have a significant effect.

Stimuli
Past studies that have explored the relevance of affect control theory on jury decision making (Robinson et al. 1994) have used one criminal case to explore the impact of the criminal's emotion display and perceived identity on sentencing. In this study, two cases of violent acts, were chosen in order to explore any differences between cases. This will help ensure that identity inferences and judgments were based on the general character of a bad (violent) act-- not some idiosyncratic features of the specific vignette.

Furthermore, past research shows that severity of the case can influence perceptions and punishment of a crime (Blumstein et al.1983; Casper and Benedict 1993). Thus, two criminal cases of similar severity were chosen; however, the details of the two criminal cases vary. Both crimes are robberies in which an individual gets hurt; one at a restaurant, the other on the street. Both victims are seriously hurt; one with a gunshot, the other with a pipe. This similarity allows a test of the generality of the observed effects. Dunning (1989) used the same two crimes, which are based on two real New York cases, in his study on construals and social judgments. Affect control researchers highly stress the theory's generalizability across individuals of different subculture and social strata (Heise 1988; Smith-Lovin 1990); thus, it is important to explore.

Both criminal cases were used as stimuli. Each participant read a criminal statement and a victim statement from one of the two trials. The summaries were described to participants as transcripts taken from videotape of the trials. They included a confession by the criminal and a victim statement. However, the emotion displays of the criminal and the victim were varied across summaries. The criminal and the victim were either depicted as sad or relaxed during the testimony. This resulted in four possible combinations: sad criminal, sad victim; sad criminal, relaxed victim; relaxed criminal, relaxed victim; and relaxed criminal, sad victim. Both the criminal's testimony and the victim's statement contained embedded nonverbal expression cues to indicate an emotion of sadness or unconcern/relaxation. Robinson et al. (1994) included other emotions, such as happiness. Results indicated that participants were confused by these emotions. Emotions of sadness and unconcern have been used extensively in past studies with manipulation checks demonstrating their effect (Robinson et al. 1994; Tsoudis and Smith-Lovin 1998, 2001).

The nonverbal cues embedded to represent the criminal's emotional displays were identical to those used in Robinson et al. (1994). A complete description of the vignette is presented there. The victim's vignette in the shooting criminal trial (with "sad" and "unconcern" condition cues in parentheses) read as follows:

The same emotion indicators are used for the sad criminal whether it is the shooting or the assault. The variation lies in the details of each crime.

Participants were told that this was a mock jury case in which they were to act like jurors in deciding punishment for the criminal. After reading the criminal's and victim's testimony for one of the two crime trials, each participant answered a questionnaire requesting judgments about the criminal, the criminal behavior, and the victim. Questions measured perceived emotion displays, empathy and identity inferences for the criminal and the victim, and recommended punishment for the criminal.

The participants were to focus on the case in order to decide punishment, not verdict. The criminal has confessed to the crime; his guilt is not in question. In order to avoid the influence of mistaken convictions and questions of guilt, the participants were told that the criminal has confessed. The focus of the study is on factors influencing judgments of punishment.

Measures
Most questions were measured with an eleven point Likert scale ranging from 0 to 10. The scale endpoints were "not at all" to "extremely".

Empathy for the criminal and the victim was measured through the question: To what extent do you feel compassionate towards the (criminal/victim)? This refers to the empathy definition previously discussed focusing on compassion. The observer with compassion for the criminal and/or victim has empathy for the criminal and/or victim.

Inferences regarding the criminal's and the victim's perceived identities were measured with similar questions: How likely is it that the (criminal/victim) will continue to engage in behaviors like the one described in his testimony? This measure of identity has been used in past affect control studies (Robinson et al.1994: Tsoudis and Smith-Lovin 1998; Tsoudis and Smith-Lovin 2001). The criminal has engaged in a negative behavior. An individual with a negative identity engages in negative behaviors. When an observer perceives that the criminal will continue to engage in this negative behavior in the future, he/she is indicating that the negative behavior is typical for the criminal. In other words, the criminal has a negative identity and typically engages in negative behaviors. When the observer perceives that the criminal will not engage in this behavior in the future, the criminal is not seen as typically engaging in negative behaviors. The negative identity is atypical; a more positive identity is seen as typical.

This is similar for the victim. This measure indicates whether being a victim (a negative identity) is a common identity for the victim. Most victims express a negative emotion (e.g., sadness) after a criminal act, because most victims have positive identities. If the victim displays a positive emotion, observers make more negative inferences about the victim's identity because of an inference that such emotions reflect routine involvement in events. The criminal behavior is seen as less negative, resulting in a less harsh punishment. How typical or atypical this behavior is for the victim will influence perceptions of the victim's identity. As previously stated, this measure has been used in past research to measure the victim's identity.

The choices for recommended punishment were death penalty, life imprisonment with no opportunity for parole, life imprisonment, 25 years, 20 years, 15 years, 10 years, 5 years with an opportunity for parole, and probation with no prison term. The alternatives were coded in a simple ordinal scale ranging from 8 (death penalty) to 0 (probation with no prison). There was no option of "not guilty" because the participants were told that the criminal had confessed to the crime.

Manipulation Checks
The manipulation checks indicate whether the emotion displays affect the participants as intended in the study. Reliability checks were incorporated in the questionnaire to verify the respondent's interpretation of the emotion displays. The relevant questions were: In your opinion, how sad was the (criminal/victim)? How unconcerned was the (criminal/victim)?
Respondents who read transcripts embedded with nonverbal cues suggesting sadness perceived the criminal to be significantly more sad (F=17.76, p=.0001) than respondents who read about a relaxed criminal. Similar results were found for the description of the victim's emotion display (F=32.06, p=.0001).

RESULTS
Preliminary Analyses
Preliminary ANOVA analyses tested the effects of case type on emotion displays, perceptions of identity, and sentencing. There were no significant interactions between the type of criminal case and other variables. Thus, the 8 conditions were collapsed into 4 conditions differing by emotion displays for the criminal and the victim. ANOVA results also demonstrated no evidence of gender main effects or interactions, thus male and female participants were grouped together for subsequent analysis.

Initial Model
Using path analysis, Figure 3 shows the standardized coefficients and levels of significance for the hypothesized model, drawn from prior research, as shown in Figure 1. Consistent with previous affect control studies, emotion displays, perceived identities and sentencing were significantly related for both the criminal and the victim. The sadder the criminal, the more positive the mock jurors' perceptions were of the criminal's identity, and the less harsh the punishment recommended. The sadder the victim, the more positive the mock jurors' perceptions were of the victim's identity and the more harsh the punishment recommended.

Empathy Model
Figure 4 and Table 1 present the results for the hypothesized revised path model depicted in Figure 2 which includes empathy as an intervening variable between the effects of emotion display on perceived identity and sentencing. As shown, there is a significant relationship between the criminal's emotion display and mock jurors' empathy for the criminal. The sadder the criminal's emotion display, the more empathy mock jurors felt for the criminal. In turn, empathy for the criminal significantly influenced the juror' perceptions of the criminal's identity, perceiving the criminal in a more positive light. Similarly, the victim's emotion display significantly affected empathy the mock jurors felt for the victim. The sadder the victim, the more empathy for the victim. However, empathy for the victim did not significantly influence the victim's perceived identity, contrary to what was hypothesized. As in previous studies, both victim and criminal perceived identities influenced the recommended level of punishment.

Post Hoc Analysis
Empathy for the victim did not significantly influence the perception of the victim's identity. Thus, there was no significant relation between the victim's emotion display and the criminal's recommended punishment. The results did demonstrate a significant relation between the criminal's emotion display and the criminal's recommended punishment, with empathy for the criminal serving as an intervening variable.

Observers, who tend to be more similar to victims than criminals, typically have more empathy for the victim. The absence of a significant relation was puzzling. Perhaps the victim was significantly operating through an unpredicted avenue.

A post hoc analysis was conducted to examine the relationship between the victim's emotion display and the mock jurors' empathy for the criminal. The victim's emotion display had a significant direct effect on the empathy for the criminal. When the mock jurors perceived a sad victim, they had less empathy for the criminal (F=18.95, p<.0001).

DISCUSSION
Research consistently supports affect control theory as an explanation for relations between emotion displays, perceived identities, and judgments about a social actor (Robinson et al. 1994; Tsoudis and Smith-Lovin 1998; Smith, Matsuno and Umino 1994; MacKinnon 1994; Heise and MacKinnon 1987; Heise 1989). This study explores whether affect control theory should be modified by adding empathy for the social actor and empathy for the object of the behavior as two intervening variables. Furthermore, this study attempts to further understand legal decision making by applying this theory to the legal context. This study found a significant relationship between emotion display and empathy for both the criminal and the victim. The presence of empathy, in turn, significantly influenced the perceived identity of the criminal. However, empathy did not influence the perceived identity of the victim. Based on these results, a juror's empathy for a criminal, although not the victim, is a contributing variable in affect control theory.

The post hoc analysis demonstrated that the victim's emotion display significantly influenced empathy for the criminal. Since the mock juror's ultimate judgment focused on recommending a sentence for the criminal, not a punishment directly affecting the victim, the mock juror may have been more focused on the criminal. In other words, perhaps this finding indicates that punishment is highly influenced by the crime (act) and criminal (actor), not the victim. If the jurors had been asked to make a disposition judgment regarding the victim, empathy for the victim might have been found to be a significant variable within this model in this context. Future studies should include judgments for the victim such as victim restitution, specifically since victim restitution is becoming a more significant part of the criminal justice system (Davis, Smith and Hillenbard 1991; Smith, Davis and Hillenbard 1989). This would give more evidence on differences in factors based on the type of punishment: victim restitution, incarceration, rehabilitation, et cetera. Furthermore, varying these punishments will give us more information on the significance of the victim's empathy. If victim restitution is not influenced by variation in victim empathy, perhaps the victim, similar to the participant (observer), is already positively perceived.

According to affect control theory, emotion displays of the criminal and victim will influence the criminal's and victim's identities, in turn, influencing punishment. Both the victim and criminal are significant in the affect control model. If there is a sad victim and sad criminal, affect control theorists predict that both emotion displays will influence the outcome of punishment. With a sad criminal and sad victim, there will be more empathy for both in contrast to the other conditions (for example, a condition with a relaxed criminal and a sad victim). It is interesting that the post hoc analysis demonstrated the victim's emotion display directly influencing empathy for the criminal. Perhaps a future study can further explore the observer's focus on the criminal and victim.

Affect control theory seeks to explain relations among several variables from an observer's perspective. Empathy is a relevant variable to be included in its application to the legal context. Legal studies have explored the importance of empathy in the legal system with regards to the criminal and victim (Fontaine and Kiger 1978; Finkel and Handel; 1989; Jenkins 1996; Ho and Venus 1995; Weir and Wrightsman 1990; Jenkins 1996; Engel 1990; Mugford, Mugford and Easterl 1989). However, there has been no theoretical explanation, explaining the significance of empathy. In order to ensure fairness and equality in legal decision making, the legal system stresses the legal factors specific to each case. However, individuals have emotions and these emotions will be evoked by emotion displays of the criminal and victim. Past research demonstrates that legal decision making is influenced by perceptions of the criminal's identity and victim's identity. Perhaps these results can be applied in jury selection and case preparation.

The results of this study further establish the importance of the empathy variable in understanding the relationship between the social actor and the observer. An observer makes decisions/judgments based on perceptions of the social actors' emotion displays and identities. As affect control theory studies have shown, the emotions of the actor and object of behavior are significant factors in identity perceptions and judgments. An individual observing another individual's emotion display, which may be an emotion of distress, will also have his/her own emotions influenced by these emotion displays. We cannot ignore that factors more specific to the observer will influence the observer's decision making. The inclusion of the empathy variable incorporates a measure of a factor directly related to the observer; a factor important for other contexts as well.

Even though this study focuses on the application of affect control within the legal arena, empathy's inclusion in affect control theory is significant in other contexts as well, in other areas of decision making. For example, parents make decisions on punishment for their children. Children typically display an emotion after engaging in a negative behavior perhaps with another child. For example, a child (actor) displays an emotion after hitting another child (object). The child who was the object of the hitting behavior also displays an emotion. The emotion displays of the two children (actor and object) will influence the parent's empathy, in turn, affecting a judgment for punishment. The relationships among emotion displays, empathy, identities, and judgment will be the same as predicted in the legal arena. Past research on parenting has continuously focused on the significant relation between parenting and empathy (Michelson 1987; Goldberg 1997; Rosenstein 1995; Gray 1978); however, there has not been a theoretical explanation connecting several variables.

Past research finds empathy to be significant in other contexts including therapists' decisions with regards to female offenders (Scott 1977), placing moral behavior in children (Misra 1991), the understanding of delinquent behaviors (Gray 1997), and correctional officers' interactions with inmates (Menard 1977). Thus, there are several areas to which affect control theory's addition of empathy can be applied in explaining social interactions.

In addition to these different contexts, empathy may vary in significance depending on whether the juror makes a decision as an individual (as in this study), in contrast to jurors deliberating as a group. What role would empathy play in a group setting? Would it have the same significance? Affect control theory has been applied to group processes (Smith-Lovin and Okamoto 2001; Smith-Lovin and Rashotte 1997; Smith-Lovin and Robinson 1990,1992). Perhaps affect control theory and empathy can be applied to explain group decision making.

As previously discussed, affect control theory was tested using students in an experimental situation. Even though this study is a theoretical piece, the argument can be made that the use of a student population is a limitation. In order to strengthen the application of affect control theory in legal decision making, this study will need to be replicated with a sample of community participants, eligible for jury participation. Nonetheless, it is important to note that studies have indicated that mock jury research with student participants are comparable to research with participant samples from the community (Bray et al. 1978; Roberts and Golding 1991; Finkel and Smith 1993).

This suggestion for future research does not imply that the use of the student population is inappropriate in testing the application of affect control theory to legal decision making. Past studies have demonstrated the generalizability of affect control theory. The cognitive processing of information is similar across all different groups (cultures and subcultures), even though dimension measures may differ across different groups (Smith-Lovin 1987; MacKinnon 1994; Smith-Lovin and Heise 1998;Smith et al. 1994). For example, the Japanese culture and the American culture differed in meanings; however, the relations among behaviors, identities, and emotions were the same (Smith et al. 1994). Similar results were found in comparing data from college students in two American universities, high school students in Ireland, and Egyptian and Lebanese students studying in the United States (Smith-Lovin and Heise 1988).

Thus, the universality of this cognitive process allows researchers to gather results from one group, generalizing to other groups (keeping in mind that the dimension measures vary only; the process is the same). The relations between actor and object operate similarly for all social interactions. Any event with emotional reactions can be framed under affect control theory. The experimental results give a new direction in using affect control theory to understand and interpret social interactions in the legal context and other contexts.

REFERENCES
Aderman, David, Sharon Brehm and Lawrence Katz. 1974. "Empathic Observation of an Innocent Victim: The Just World Revisited." Journal of Personality and Social Behavior 29:342-347.

Anthony, Susan. 1976. "The Attribution of Responsibility for the Outcome of an Interpersonal Interaction." Journal of Psychology 93:85-91.

Archer, R.L., Foushee, H.C., Davis M.H. and Aderman, D. 1979. "Emotional empathy in a Courtroom simulation: A person-situation interaction." Journal of Applied Social Psychology 9:275-291.

Barrett-Lennard, G.T. 1962. "Dimensions of therapy response as causal factors in therapeutic change." Psychological Monographs 76:1-33.

Batson, Daniel. 1981. "Is Empathic Emotion a Source of Altruistic Motivation?" Journal of Personality and Social Psychology 40: 290-302.

Belli, M.M. 1956. Blood money: Ready for the plaintiff. New York, NY: Grosset and Dunlap.

Berkowitz, L. and Macaulay, Jacqueline. 1970. Altruism and Helping Behavior: Social Psychological Studies of Some Antecedents and Consequences. New York, NY: Academic Press.

Berkowitz, L. 1972. "Social Norms, Feelings and Other Factors Affecting Helping Behavior and Altruism." In Berkowitz, L. (Ed.) Advances in Experimental Social Psychology (Vol 6 pp. 63-106). New York, NY: Academic Press.

Berkowitz, L. 1987. "Mood, Self-Awareness and Willingness to Help." Journal of Personality and Social Psychology 52: 721-729.

Blumstein, Alfred, Jacqueline Cohen, Susan E. Martin and Michael H. Tonry, Eds. 1983. Research on Sentencing: Search for Reform. Washington D.C.: National Academy Press.

Bray, Robert M, Cindy Struckman-Johnson, Marshall Osborne, James McFarlane and Joanne Scott. 1978. "The Effects of Defendant Status on the Decisions of Student and Community Juries." Social Psychology 41(3): 256-260.

Casper, J.D. and Benedict, K.M. 1993. "The influence of case outcome information and attitudes on juror decision making in search and seizures." In R. Hastie (Ed.) Inside the Juror (pp.65-82) New York, NY: Cambridge University Press.

Chaikin, Alan and John Darley. 1973. "Victim or Criminal?: Defensive Attribution of Responsibility." Journal of Personality and Social Psychology 25:268-275.

Cohen, J.S. 1961. "Trial Tactics in Criminal Cases." In H. Toch (Ed.) Legal and Criminal Psychology. New York, NY: Holt, Rinehart and Winston.

Coller, Sarah and Patricia Resick. 1987. "Women's Attributions of Responsibility of Date Rape: The Influence of Empathy and Sex-Role Stereotyping." Violence and Victims 2:115-125.

Davis, R.C., B. Smith, and S. Hillenbard. 1991. "Increasing Offender Compliance with Restitution Orders." Judicature 74:245-248.

Deitz, S.R. 1980. "Double Jeopardy: The rape victim in court." Rocky Mountain Psychologist Fall: 1-17.

Deitz, Sheila and Lynne Byrnes. 1981. "Attribution of Responsibility for Sexual Assaults: The Influence of Observer Empathy and Criminal Occupation and Attractiveness." Journal of Psychology 108: 17-29.

Dunning, David. 1989. Situational Construals and Social Judgment: Predicting Divergent Responses to Death Penalty Crimes. Cornell University. Unpublished Manuscript.

Engel, Frema. 1990. "Victims of Criminal Acts: A Professional Intervention: Victim." Criminologie. 23:5-22.

Erez, Edna and Pamela Tontodonato. 1990. "The Effect of Victim Participation in Sentencing on Sentence Outcome." Criminology 28:451-474.

Finkel, Norman J. and Sharon E. Handel. 1989. "How Jurors Construe Insanity." Law and Human Behavior 13: 41-59.

Finkel, Noman and Stefanie Smith. 1993. "Principals and Accessories in Capital Felony-Murder: The Proportionality Principle Reigns Supreme." Law and Society Review 27(1): 129-156.

Fontaine, Gary and Rick Kiger. 1978. "The Effects of Defendant Dress and Supervision on Judgments of Simulated Jurors: An Exploratory Study." Law and Human Behavior 2:63-71.

Gleason, James and Victor Harris. 1975. "Race, Socio-Economic Status, and Perceived Similarity as Determinants of Judgments by Simulated Jurors." Social Behavior and Personality 3(2): 175-180.

Gleason, James and Victor Harris. 1976. "Perceived Freedom, Accident Severity, and Empathic Values as Determinants of the Attribution of Responsibility." Social Behavior and Personality 4:171-176.

Gray, Charlene. 1978. "Empathy and Stress as Mediators in Child Abuse: Theory, Research and Proactive Implications." Dissertation. Maryland.

Gray, Lewis. 1997. "Differences between Violent and Nonviolent Delinquent Males on Measures of Empathy, Attachment, Parental Criminality and Parental Acceptance/ Rejection." Unpublished Dissertation.

Greenson, R.R. 1960. "Empathy and its vicissitudes." International Journal of Psychoanalysis 41:418-424.

Greenson, R.R. 1967. The Technique and Practice of Psychoanalysis Volume 1. New York, NY: International Universities Press.

Heise, David R. 1979. Understanding Events: Affect and the Construction of Social Action. New York, NY: Cambridge University Press.

Heise, David R. and Neil J. MacKinnon. 1987. "Affective Bases of Likelihood Judgments." Journal of Mathematical Sociology 13:133-151.

Heise, David R. and Elsa Lewis. 1988. Introduction to INTERACT. National Collegiate Software Clearinghouse, Box 8101 North Carolina State University, Raleigh, NC 27695.

Heise, David R. 1989. "Effects of Emotion Displays on the Assessment of Character." Social Psychology Quarterly 52: 10-21.

Ho, Robert and Marilyn Venus. 1995. "Reactions to a Battered Woman Who Kills her Abusive Spouse: An Attributional Analysis." Australian Journal of Psychology 47:153-159.

Hoffman, M.L. 1984. "The contribution of empathy to justice and moral judgment." In N. Eisenberg and J. Strayer (Eds.), Empathy and Its Development. New York, NY: Cambridge University Press.

Jenkins, Pamela. 1996. "Contested Knowledge: Battered Women as Agents" In Witnessing for Sociology: Sociologists in Court. Connecticut: Praeger.

Kassin, Saul M. and Lawrence S. Wrightsman. 1988. The American Jury on Trial. New York: NY: Hemisphere.

Krebs, Dennis. 1975. "Empathy and Altruism." Journal of Personality and Social Psychology 32: 1134-1146.

Lazowski, Linda E. 1987. "Speaker's Nonverbal Expressions of Emotion as Moderators of Listener's Reactions to Disclosure of Self Harm and Social Harm." Unpublished doctoral dissertation, University of California, Santa Barbara.

Lincoln, Alan and George Levinger. 1972. "Observer's Evaluations of the Victim and The Attacker in an Aggressive Incident." Journal of Personality and Social Psychology 22: 202-210.

MacKinnon, Neil J. 1994. Symbolic Interactionism as Affect Control. New York, NY: University of New York Press.

McGowen, Ramsey and Glen King. 1982. "Effects of Authoritarian, Anti-Authoritarian, and Egalitarian Legal Attitudes on Mock Juror and Jury Decisions." Psychological Reports 51(3): 1067-1074.

Menard, James. 1977. "Preparing Correctional Officers and Inmates of Their Interaction." Offender Rehabilitation 1:241-249.

Michelson, Larry. 1987. "Cognitive-behavioral strategies in the prevention and treatment of antisocial disorders in children and adolescent", Chapter in Prevention of Delinquent Behavior. California: Sage.

Misra, Girishwar. 1991. "Socio-Cultural Influences on Moral Behavior." Indian Journal of Social Work 52: 179-194.

Mugford, Jane, Stephen Mugford and Patricia Weiser Easteal. 1989. "Social Justice, Public Perceptions and Spouse Assault in Australia." Social Justice 16:103-123.

Roberts, Caton F. and Stephen L. Golding. 1991. "The Social Construction of Criminal Responsibility and Insanity." Law and Human Behavior 15(4): 349-376.

Robinson, Dawn, Lynn Smith-Lovin and Olga Tsoudis. 1994. "From a Heinous Crime to an Unfortunate Accident: The Effects of Remorse in Responses to Mock Criminal Confessions." Social Forces 73: 175-190.

Rosenstein, Paula. 1995. "Parental Leave of Empathy as Related to Risk Assessment in Child Protective Service." Child Abuse and Neglect 19:1349-1360.

Scott, Edward. 1977. "Therapy with Female Offenders." International Journal of Offender Therapy and Comparative Criminology. 21: 208-220.

Smith, Herman W., Takanori Matsuno and Michio Umino. 1994. "How Similar are Impression Formation Processes among Japanese and Americans." Social Psychology Quarterly 57: 124-139.

Smith, B.E., R.C. Davis and S. Hillenbard. 1989. Improving Enforcement of Court Ordered Restitution. American Bar Association.

Smith-Lovin, Lynn and David Heise. 1988. Affect Control Theory: Research Advances. New York: Gordon and Breach.

Smith-Lovin, Lynn. 1990. "Emotion as a Confirmation and Disconfirmation of Identity: An affect control model." In T.D. Kemper's (Ed.), Research Agendas in Emotions. New York: SUNY Press.

Smith-Lovin, Lynn and Dina Okamoto. 2001. "Status, Gender, and Topic Transitions in Task Group Discussions." American Sociological Review 66(6):852-873.

Smith-Lovin, Lynn and Lisa Slattery Rashotte. 1997. "Who Benefits form Being Bold: The Interactive Effects of Task Cues and Status Characteristics on Influence in Mock Jury Groups." Chapter in Advances in Group Processes, Volume 14. JAI Press.

Smith-Lovin, Lynn and Dawn Robinson. 1992. "Gender and Conversational Dynamics." Pp. 122-156 in Cecilia Ridgeway (ed.) Gender and Interaction. Springer-Verlag.

Smith-Lovin, Lynn and Dawn Robinson. 1990. "The Timing of Interruptions in Groups Discussions." Pp. 45-74 in Edward J. Lawler, Cecilia Ridgeway, Henry Walke and Barry Markovsky (eds.), Advances in Group Processes, Volume 7. JAI Press.

Stotland, E. 1969. Exploratory studies of empathy. In L. Berkowitz (Ed.) Advances in Experimental Social Psychology, Volume 4. New York: Academic Press.

Tsoudis, Olga and Lynn Smith-Lovin. 1998. "How Bad Was It? The Effects of Victim and Perpetrator Emotions on Responses to Criminal Court Vignettes." Social Forces 77(2) 695-722.

Tsoudis, Olga and Lynn Smith-Lovin. 2001. "Criminal Identity: The Key to Situational Construals in Mock Criminal Court Cases." Sociological Spectrum 21(1):3-31.

Von Hirsch, Andrew and Hyman Gross, Eds. 1981. Sentencing. New York: Oxford Press.

Wier, Julie and Lawrence Wrightsman. 1990."The eterminants of Mock Jurors' Verdicts in a Rape Case." Journal of Applied Social Psychology 20:901-919.

Wispe, Lauren. 1991. The Psychology of Sympathy. New York, NY: Plenum Press.

ABOUT THE AUTHOR
Dr. Tsoudis received her bachelor's degree in Human Development and Family Studies at Cornell University in 1991. She received her master's degree in sociology in 1993 and then her Ph.D. in 1995 at the University of Arizona. Her other research interests include studying the nation of Cyprus with regard to its low crime rates, understanding jury selection during voir dire, and the effects of criminal justice education. Direct all correspondence to Olga Tsoudis, Ph.D., Assistant Professor, Administration of Justice Department, Arizona State University West. 4701 W. Thunderbird Road, Phoenix, Arizona 85306, email: otsoudis@asu.edu. Ph: 602-543-6628 back