Annu. Rev. Psychol. 2000.

51:665697
Copyright q 2000 by Annual Reviews. All rights reserved

EMOTION, REGULATION, AND MORAL

DEVELOPMENT
Nancy Eisenberg
Department of Psychology, Arizona State University, Tempe, Arizona 852871107;
e-mail: [email protected]

Key Words aggression, empathy, guilt, prosocial behavior, shame

Abstract Research and theory on the role of emotion and regulation in morality
have received considerable attention in the last decade. Much relevant work has concerned the role of moral emotions in moral behavior. Research on differences between
embarrassment, guilt, and shame and their relations to moral behavior is reviewed, as
is research on the association of these emotions with negative emotionality and
regulation.
Recent issues concerning the role of such empathy-related responses as sympathy
and personal distress to prosocial and antisocial behavior are discussed, as is the
relation of empathy-related responding to situational and dispositional emotionality
and regulation. The development and socialization of guilt, shame, and empathy also
are discussed briefly. In addition, the role of nonmoral emotions (e.g. anger and sadness), including moods and dispositional differences in negative emotionality and its
regulation, in morally relevant behavior, is reviewed.

CONTENTS
The Moral Emotions....................................................................... 666
The Self-Conscious Moral Emotions........................................................ 666
Empathy-Related Responding................................................................ 671
The Development of Guilt, Shame, and Empathy......................................... 678
The Socialization of Guilt, Shame, and Empathy......................................... 680
Relations of Nonmoral Emotions to Morally Relevant Behavior ................. 682
Mood ............................................................................................ 683
Individual Differences in Emotionality and Regulation.................................. 684
Summary.......................................................................................... 688

For millennia, philosophers have debated whether emotions can be moral and
whether emotion contributes to higher-level moral judgment and behavior. Emotions, by their very nature, express a personal, polarized, and biased perspective.
Thus, emotion has been viewed as biasing ones evaluations and cognitions and
as disrupting rational, moral thought. More recently, philosophers have argued
that biased emotional reactions are justified and that emotions help people
00846570/00/02010665$12.00

665

666

EISENBERG

to distinguish moral features in specific contexts, to motivate moral behavior, and

to undercut immoral behavior. In addition, emotions can play a communicative
role by revealing our moral values and concerns to others and ourselves (BenZeev 1997, Blum 1980).
Philosophers changing view of the role of emotion in morality is consistent
with the predominant view of emotion in psychology today. As is discussed,
higher-order emotions such as guilt and sympathy are believed to motivate moral
behavior and to play a role in its development and in moral character (e.g. Hoffman 1998, Walker & Pitts 1998). Moreover, in the 1990s there has been considerable interest in the role of basic emotions (i.e. those that are probably universal
and involve less cognitive complexity), such as anger and fearfulness, in moral
behavior (Eisenberg et al 1999a, Kochanska 1997).
In this chapter, recent issues and findings concerning the role of emotion and
emotion-related regulation in moral functioning are reviewed. Behaviors of moral
relevance as viewed by others, regardless of their motivation, are the primary foci
of interest; the difference between behaviors that are truly moral and those motivated by nonmoral factors is discussed primarily in the context of differentiating
between moral and nonmoral emotional reactions (e.g. sympathy vs personal
distress). First, issues and findings pertaining to several moral emotions (guilt,
shame, and empathy-related responding) are reviewed. Then the role of situational
emotion (mood) in moral behavior is discussed briefly. Next, the relations of
individual differences in emotionality and regulation to morally relevant behaviors are examined. Finally, research on the prediction of morally relevant behavior
from the combination of emotionality and regulation is considered.

THE MORAL EMOTIONS

Several emotions, including guilt, shame, and empathy, have been viewed as
playing a fundamental role in morality. Although pride is a self-evaluative emotion that can stem from moral behavior, research on pride usually has concerned
achievement. Thus, pride is not discussed further.

The Self-Conscious Moral Emotions

Guilt and shame frequently have been implicated in theories of morality, with
guilt being a quintessent moral emotion. Both are considered self-conscious
emotions, as is embarrassment. These emotions are labeled self-conscious
because the individuals understanding and evaluation of the self are fundamental
to these emotions.
Embarrassment Keltner & Buswell (1997) argued that embarrassment is an
emotion distinct from guilt and shame in that it involves antecedents, experience,
and nonverbal displays that are different from those of other emotions. Recent
research is consistent with the conclusion that embarrassment, in comparison to
shame and guilt, is the least negative, least serious, and most fleeting emotion; it

MORAL DEVELOPMENT

667

is the least related to moral implications and moral transgressions; it involves less
anger at the self and less interest in making amends; and it tends to involve
surprising and accidental events for which people feel less responsible (Miller &
Tangney 1994; JP Tangney, D Marschall, K Rosenberg, DH Barlow & P Wagner,
unpublished data). Rather than playing a role in morality, embarrassment may
serve to appease others for ones transgressions of social convention by eliciting
light-hearted emotion (Keltner 1995) or may prevent loss of face and serve to
assure adherence to important social norms (Leary et al 1996, Miller & Leary
1992). Thus, there generally is consensus that embarrassment plays at most a
minor role in moral behavior.
Guilt and Shame Guilt has been defined in a variety of ways. In classic psychoanalytic theory, it is viewed as a superego response to ones own unacceptable
impulses, often based on anxiety caused by childhood conflicts over such issues
as abandonment and punishment by parents. This type of guilt generally is seen
as causing psychological distress and problems with adjustment, and today it is
not viewed as playing much of a role in moral behavior. In contrast, in developmental and social psychology, guilt often refers to regret over wrongdoing. For
example, it has been defined as an agitation-based emotion or painful feeling of
regret that is aroused when the actor actually causes, anticipates causing, or is
associated with an aversive event (Ferguson & Stegge 1998:20). The guilty actor
accepts responsibility for a behavior that violates internalized standards or causes
anothers distress and desires to make amends or punish the self (Ferguson &
Stegge 1998, Hoffman 1998, Tangney 1991). It is this type of guilt that is most
relevant to a discussion of moral emotion.
Shame often has been used as a synonym for guilt and has received much less
theoretical attention in the past. More recently it has been defined as . . . a
dejection-based, passive, or helpless emotion aroused by self-related aversive
events. The ashamed person focuses more on devaluing or condemning the entire
self, experiences the self as fundamentally flawed, feels self-conscious about the
visibility of ones actions, fears scorn, and thus avoids or hides from others
(Ferguson & Stegge 1998:20).
The topic of guilt, although important in psychoanalytic theory and in early
discussions of socialization, was virtually ignored by social and developmental
psychologists in the 1970s and 1980s (Baumeister et al 1994). However, in the
1990s, there has been a flurry of research on these self-conscious emotions.
Differences Between Guilt and Shame Many researchers and theorists now
agree that guilt and shame (at least as defined above) are two distinct emotions
and that an important difference between them is in the degree of focus on the
self (Lewis 1971, Tangney 1998). When a person experiences shame, the entire
self feels exposed, inferior, and degraded. Adults report that shame experiences
are more painful and intense than are guilt experiences and are associated with a
preoccupation with others opinions. In contrast, guilt generally is less painful

668

EISENBERG

and devastating than shame because, when one experiences guilt, the primary
concern is with a particular behavior, somewhat distinct from the self (Ferguson
et al 1991, Tangney 1998). Guilt involves feelings of tension, remorse, and regret,
but does not affect ones core identity. Shame is associated with the desire to
undo aspects of the self, whereas guilt is reported to involve the desire to undo
aspects of behavior (Niedenthal et al 1994). Similarly, shame, but not guilt, is
related to discrepancies between ones beliefs about the self and beliefs about
what the self ought to be or what the ideal self would be (Tangney et al 1998).
It is important to note that guilt often has been operationalized as a response
that involves concern about others feelings and with reparation (e.g. Tangney
1991). It is likely that guilt that is less reparation oriented, based on irrational or
illogical assessments of responsibility, or that is not resolved can affect feelings
about the self over time and may have more maladaptive effects. Moreover, guilt
and shame often co-occur; children may be especially prone to the combination
(Ferguson et al 1999).

Guilt vs Shame: Links to Empathy and Moral Behavior Based on adults reports,
shame and guilt both involve a sense of responsibility and the feeling that one
has violated a moral standard (JP Tangney, D Marschall, K Rosenberg, DH Barlow & P Wagner, unpublished data). Moreover, both emotions can be responses
to the same situations, and both can arise from concerns about the effects of ones
behavior on others (Tangney 1992; JP Tangney, D Marschall, K Rosenberg, DH
Barlow & P Wagner, unpublished data). Nonetheless, guilt appears to be the more
moral emotion of the two. Shame, but not guilt, is likely to arise from nonmoral
situations and issues (e.g. failure in performance situations or socially inappropriate behavior), and only shame seems to involve concern about others evaluations (Ferguson et al 1991, Tangney 1992). Shamed people are relatively unlikely
to try to rectify their transgression. Probably because guilt is focused more on the
transgression than the self, guilt seems to motivate restitution, confession, and
apologizing rather than avoidance (Tangney 1998; JP Tangney, D Marschall, K
Rosenberg, DH Barlow & P Wagner, unpublished data). However, it should be
kept in mind that, in much of this work, guilt has been defined as a reparative
response, so these associations are not surprising.
Moreover, shame and guilt appear to be differentially related to empathyrelated responding. Tangney (1991) found that guilt was associated with adults
self-reported, other-oriented empathic responsiveness, whereas shame was negatively associated, especially when controlling for guilt. Shame was especially
associated with personal distress reactions (i.e. aversive, self-focused reactions to
others in need or distress). When providing autobiographical accounts of shame
and guilt experiences, people conveyed more empathy in guilt than in shame
descriptions, although this association was somewhat stronger among adults than
children (JP Tangney, D Marschall, K Barlow & DH Wagner, unpublished data).
Nonetheless, because shame and guilt are substantially correlated and these anal-

MORAL DEVELOPMENT

669

yses were part correlations controlling for one another, it is likely that the distinction between guilt and shame is not quite as clear cut as these findings suggest.
Guilt and Shame as Predictors of Externalizing Behavior Tangney also has
found that shame generally is more consistently and highly correlated than is guilt
with externalizing problems, including aggression. Findings of this sort, if consistent, are important because problems with aggression are viewed as a component of antisocial behavior. When part correlations were used to assess relations
of problem behavior with adults guilt-free shame and shame-free guilt, shame
still was consistently positively related to externalizing problems, whereas guilt
generally was unrelated (Tangney et al 1992; also see Tangney et al 1996a).
Similar results have been obtained for children in some studies (Ferguson et al
1997; JP Tangney, PE Wagner, SA Burggraf, R Gramzow & C Fletcher, unpublished data), although even shame-free guilt has been associated with externalizing problems for girls [but not for boys (Ferguson et al 1999)]. Among children,
adolescents, and adults, guilt-free shame has been linked with direct, indirect, and
displaced aggression, whereas shame-free guilt has been negatively related to
these types of responding (Tangney et al 1996b).
Tangneys work has been conducted with nonclinical populations and usually
has involved a measure of guilt in response to specific events (as assessed with
brief vignettes). Moreover, in this work guilt was defined as an adaptive response
such as taking responsibility or wanting to make reparations. It appears that dispositional guilt (or shame) that is more global, ruminative, and chronic and guilt
assessed with a projective measure (as well as lack of guilt) are positively related
to childrens and adults psychopathology, including externalizing problems (Ferguson et al 1996, 1999; Harder et al 1992; OConnor et al 1999; Sorenson et al
1997). In addition, it appears that girls high in shame-free guilt sometimes may
be prone to externalizing behavior, even when guilt is assessed by measures that
tap concern with adhering to standards, expressing empathy, and taking appropriate responsibility (Ferguson et al 1999; Zahn-Waxler & Kochanska 1990).
Ferguson et al (1999) suggested that the relation between guilt and externalizing
problem behavior may hold because girls experience anger at being held to stricter
standards of behavior than boys but also realize that failure to express guilt will
reap negative consequences.
In discussing the differences in findings across measures, Tangney (1996)
argued that her situational measure of guilt is the more valid way to assess the
construct of guilt, especially shame-free guilt. Other measures of guilt often (a)
rely on respondents abilities to differentiate verbally between guilt and shame,
(b) do not assess emotional reactions in specific contexts, and (c) likely tap a
combination of guilt and shame. However, it is possible that guilt often is not as
distinct from shame or as adaptive as operationalized in Tangneys work, especially in childhood. In any case, initial findings support the view that one gets
different results with measures that focus on specific behaviors (and are unlikely
to reflect ruminative guilt) and with measures of more global, chronic, and unre-

670

EISENBERG

solved guilt (Eyre & Ferguson 1997, Ferguson & Stegge 1998, Quiles & Bybee
1997, Tangney et al 1995). It is quite possible that scenario measures of guilt in
specific contextswhich are correlated with the personality trait of agreeableness
in adulthoodtap empathy-based guilt, whereas other commonly used measures
of guilt tap anxiety-based guilt (Einstein & Lanning 1998). Moreover, it is possible that there is a continuum of guilt proneness and that very low levels of guilt
are related to externalizing problems, whereas very high levels of guilt are related
to shame and irrational guilt.
The Relation of Guilt and Shame to Negative Emotionality and Regulation It
is likely that dispositional (personality or temperamental) characteristics of people
play a role in the proclivities to experience guilt and shame. Because of the
intrinsic role of emotion in these responses and the role of regulation in both
managing emotion and in moral behavior, dispositional emotionality and regulation are likely correlates of the tendencies to experience guilt and shame.
In fact, guilt and shame have been linked to fear, hostility, anxiety, and sadness
in adulthood (Forgas 1994, Harder et al 1992, OConnor et al 1999, Watson &
Clark 1992) and childhood (Zahn-Waxler & Robinson 1995). The degree to which
shame and guilt are differentially related to negative emotion likely varies with
the measure used. In some studies with adults (JP Tangney, D Marschall, K Rosenberg, DH Barlow & P Wagner, unpublished data) or toddlers (Zahn-Waxler &
Robinson 1995), there were few differences in the patterns of relations for guilt
and shame. In contrast, in another study with adults, shame and anxious guilt
were positively correlated with negative emotionality (i.e. neuroticism on a measure of the Big Five personality factors), whereas situational guilt (which often
may be based on empathy) was not (Einstein & Lanning 1998). Moreover, some
researchers have found that guilt, especially situationally based guilt, is unrelated
or weakly related to various negative emotions, especially when the effects of
shame are controlled in the correlations, whereas shame is associated with anger
and anxiety, even when guilt is controlled in the correlations (Tangney et al 1992,
1996a; JP Tangney, PE Wagner, SA Burggraf, R Gramzow & C Fletcher, unpublished data). Until situational guilt based on situational empathy is differentiated
from chronic guilt, findings are likely to be inconsistent.
During the toddler and early-childhood years, the link between guilt or shame
and other negative emotions appears to occur primarily in girls (Kochanska et al
1994), with the exception that fear has been related to guilt in male, but not
female, toddlers (Zahn-Waxler & Robinson 1995). However, because mothers
sometimes provided the data on childrens guilt and emotionality, it is possible
that the sex difference is based on something related to mothers beliefs about
girls and boys emotions. Although there is some evidence that females show
shame more than do males, it is not clear that there are sex differences in guilt
(Ferguson & Eyre 1999). Moreover, it is quite possible that displays of guilt in
very young children actually reflect a combination of shame, guilt, and fear and
that guilt in the very young has a different significance than does guilt in older

MORAL DEVELOPMENT

671

children and adults (who better understand notions of responsibility and

causality).
The relation of guilt and shame to individual differences in dispositional regulation seldom has been examined. Emotionally well-regulated children would be
expected to manage their emotional arousal so that they are not overwhelmed by
feelings of shame; moreover, behavioral regulation would be expected to underlie
some markers of guilt such as reparation. Consistent with these expectations,
Rothbart et al (1994) found that mothers ratings of 7-year-olds regulatory capacities (effortful control, including the abilities to voluntarily shift and focus attention and inhibit behavior) were positively correlated with mothers reports of their
childrens guilt/shame (combined). In a study with 2- to 6-year-old children, regulatory capacities were associated with affective discomfort after wrongdoing,
but only for girls (Kochanska et al 1994). Although, as is discussed below, measures of conscience often are associated with dispositional regulation, these measures do not necessarily tap guilt. Thus, although well-regulated children might
be expected to experience relatively high levels of situationally and ageappropriate guilt, links between regulation and both guilt and shame have been
insufficiently examined.
Summary In the 1990s, there has been an increase in research on guilt and
shame, the difference between the two, their relation to morally relevant behaviors, and their socialization correlates. Findings often vary as a function of the
index of guilt. Salient issues to be addressed include the meaning of various
measures of guilt, gender differences in guilt and in the relation of guilt or shame
to moral behavior, and the role of regulatory capacities in guilt vs shame. In
addition, it is important to determine whether different negative emotions are
differentially related to chronic vs empathy-based guilt or shame, for example,
whether anxiety and anger are related more closely to shame than guilt.

Empathy-Related Responding
Empathy-related reactions can be other- or self-related or sometimes neither.
Eisenberg and colleagues (e.g. Eisenberg et al 1994a) have defined empathy as
an affective response that stems from the apprehension or comprehension of
anothers emotional state or condition and is similar to what the other person is
feeling or would be expected to feel. If a child views a sad person and consequently feels sad (even though the child differentiates his or her own and the
other persons emotional states or situations at a rudimentary level), that child is
experiencing empathy.
In Eisenbergs view, pure empathy is not other-oriented. However, with further
cognitive processing (assuming that the individual is old enough to differentiate
between ones own and others internal states), an empathic response usually turns
into either sympathy, personal distress, or some combination (perhaps alternating)
thereof. Sympathy is an emotional response stemming from the apprehension or

672

EISENBERG

comprehension of anothers emotional state or condition, which is not the same

as what the other person is feeling (or is expected to feel) but consists of feelings
of sorrow or concern for the other. Thus, if a girl sees a sad peer and feels concern
for the peer, she is experiencing sympathy. A sympathetic reaction often is based
upon empathic sadness, although sympathy also may be based on cognitive perspective taking or encoded cognitive information relevant to anothers situation
accessed from memory. Personal distress is a self-focused, aversive, affective
reaction to the apprehension of anothers emotion (e.g. discomfort or anxiety),
such as the distress of a person feeling anxious when viewing someone who is
sad.
About two decades ago, Batson (1998) proposed that sympathy (which he has
called empathy) is associated with other-oriented motivation, whereas personal
distress is associated with the motive to alleviate ones own aversive affective
state. Thus, sympathy is viewed as an other-oriented moral emotion fostering
altruism. In contrast, personal distress is hypothesized to lead to prosocial behavior only when that is the easiest way to reduce ones own aversive emotional state
(e.g. in a situation in which one cannot escape dealing with the person causing
ones distress). Thus, sympathy is viewed as a moral emotion, whereas personal
distress is believed to result in egoistically motivated behavior.
In the 1980s, there was considerable interest in why people sometimes help
others at a cost to themselves and whether truly selfless altruism exists. These
questions stimulated numerous empirical studies demonstrating a positive relation
between sympathy and prosocial behavior and a negative relationor sometimes
a lack of a relationbetween personal distress and prosocial behavior, both in
adults (Batson 1998) and in children (Eisenberg & Fabes 1991, 1998). Work on
this topic has continued into the 1990s. Another emerging issue in recent years
has been the role of emotionality and regulation in empathy-related responding.
Empathy-Related Responding and Prosocial/Antisocial Behavior Researchers
have continued to demonstrate empirical relations between prosocial behavior
and both situationally induced and dispositional empathy-related responding. In
the social psychological literature, sympathy and personal distress generally have
been elicited in laboratory situations and then examined in relation to prosocial
behavior directed toward the target of that emotion. In general, the positive relation between sympathy and prosocial behavior has been replicated (Batson 1998,
Batson et al 1997b, Trobst et al 1994). Among children, markers of empathy and
sympathy in specific situations, such as their facial, behavioral, and physiological
reactions to viewing others in need or distress, also have been associated with
situational or dispositional prosocial behavior (Denham et al 1994; Fabes et al
1993; Hastings & Zahn-Waxler 1998; Holmgren et al 1998; Zahn-Waxler et al
1992, 1995). In addition, situational markers of personal distress generally have
been negatively related or unrelated to childrens prosocial behavior (e.g. Holmgren et al 1998, Fabes et al 1993, Miller et al 1996), although self-distress has
been positively related to toddlers prosocial behavior when the toddlers caused

MORAL DEVELOPMENT

673

the other persons distress [so the self-distress may have reflected a rudimentary
guilt reaction (Zahn-Waxler et al 1992)]. The diversity of methods used in various
studies with children enhances ones confidence that the relation between situational sympathetic concern and prosocial behavior is real, albeit sometimes modest in magnitude.
In research with adults, investigators have demonstrated that sympathy may
not only motivate moral behavior in specific contexts (Batson et al 1997b), but
it may also cause enduring changes in an individuals concern about others welfare (Batson et al 1995). For example, people who are induced to experience
sympathy for a member of a stigmatized group actually develop more benign
attitudes toward those individuals weeks later (Batson et al 1997a). However,
sympathy, like egoism, also can undermine concern with the welfare of a group
if an individual has to choose between allocating resources to the group or to
someone with whom they were induced to sympathize (Batson et al 1999).
The central focus in much of the social psychological research on empathy is
an issue that has been debated fiercely for two decadeswhether prosocial behavior induced by empathy (or sympathy) is really motivated by altruism (a pure
other-orientation) rather than egoism. The most recent challenge to the notion of
true altruism is the argument that sympathy for another leads to a greater sense
of self-other overlap, with the consequence that helping is not selfless but is
directed toward both the other person and the self [i.e. to make oneself feel better
(Cialdini et al 1997)]. Empirical data both for and against this argument have
been published (Batson et al 1997b, Cialdini et al 1997), and the debate continues
(Batson 1997, Neuberg et al 1997). In this literature, Cialdini et al (1997) assessed
merging of self-other boundaries with a measure of oneness (i.e. adults reports
that they would use the term we to define their relationship with the target of
sympathy and their selection of circles drawn close to each other to indicate the
closeness of their relationship with the other person). However, it is possible that
their measures of oneness reflect the awareness that they feel concern for the
person or close to the person but not a merging of boundaries. It is difficult to
imagine actual merging of boundaries when the study participants were responding to hypothetical situations.
Other researchers have been concerned with the relation of the dispositional
tendency to experience sympathy and/or personal distress (rather than situationally induced empathy-related responding) to such prosocial behaviors as providing support, volunteering, or helping. In general, links between dispositional
sympathy and prosocial behavior have been demonstrated, albeit to various
degrees, in research with both adults (e.g. Carlo et al 1999, Penner & Finkelstein
1998, Trobst et al 1994) and children and adolescents (e.g. Carlo et al 1998;
Eisenberg et al 1991c, 1995a; also see Estrada 1995, Roberts & Strayer 1996,
Eisenberg & Fabes 1998). It is likely that sympathy is most closely linked to
modes of prosocial behavior that are other-oriented, such as spontaneously emitted sharing behaviors in preschoolers (Eisenberg et al 1999c). In addition, dispositional sympathy and empathy have been associated with low levels of

674

EISENBERG

aggression and externalizing problems in adolescence (Carlo et al 1998, Cohen

& Strayer 1996, Hastings & Zahn-Waxler 1998, Tremblay et al 1994).
Relations between sympathy/empathy (situational or dispositional) and prosocial behavior generally have been modest to moderate, and sometimes, when
measures of the two constructs have not been obtained from the same reporter or
in the same setting, they have been weak. Thus, an important issue is to identify
factors that moderate the degree of this association. When predicting prosocial
behavior, Miller et al (1996) found an interaction between level of moral judgment
and situational sympathy such that childrens helping of a distressed peer (shown
in a videotape) was highest if children were high in both needs-oriented (rudimentary other-oriented) moral reasoning and in reported sympathy. Similarly,
Knight et al (1994) found that the combination of high dispositional sympathy,
high perspective taking, and the ability to understand units and the value of money
predicted high helping of a peer in need (although sympathy alone also was
associated with helping). Studies such as these support the need to identify dispositional and situational moderators of the strength of the relation between
empathy-related responding and prosocial (or antisocial) behavior.
The Relation of Empathy-Related Responding to Emotionality and Regulation The differing relations of personal distress and sympathy to prosocial
behavior are consistent with the conclusion that the subjective experiences of
sympathy and personal distress are different. Eisenberg et al (1994a) hypothesized
that empathic overarousal in situations involving negative affect results in an
aversive, overaroused emotional state, which leads to a focus on ones one needs
and, consequently, personal distress (also see Hoffmans 1982 discussion of
empathic overarousal). In support of this view, investigators have found that negative emotional arousal, especially for reflective affective states such as sadness
(Green & Sedikides 1999), is associated with a focus on the self (Wood et al
1990) and that people exhibit higher physiological arousal and sometimes report
more distress in situations likely to elicit person distress in contrast to sympathy
(Eisenberg & Fabes 1991; Eisenberg et al 1991a,b).
Regulation Conceptualizing sympathy and personal distress in the above manner
led to the prediction that people who can regulate their emotions and emotionrelated behavior should be relatively likely to experience sympathy rather than
personal distress. Empirical findings in studies of adults have been somewhat
consistent with this prediction. In several studies, various measures of behavioral
and attentional regulation (e.g. attention shifting) have been negatively correlated
with dispositional personal distress (Eisenberg et al 1994a, Eisenberg & Okun
1996, Okun et al 1999). Sometimes dispositional regulation has been positively
related to adults sympathy (Eisenberg & Okun 1996), although in two studies
this relation was significant only when individual differences in negative emotionality were controlled (Eisenberg et al 1994a, Okun et al 1999). In a longitudinal study with children, adults reports of childrens dispositional regulation

MORAL DEVELOPMENT

675

were positively related to teachers or childrens reports of childrens dispositional

sympathy, both in concurrent analyses and over time (Eisenberg et al 1996c, 1998;
Murphy et al 1999). In addition, Rothbart and colleagues (1994) found that mothers ratings of 7-year-old childrens empathy were related to childrens effortful
control (an index of regulation), as rated by mothers. Thus, there appears to be a
positive relation between regulation and sympathy/empathy, especially in childhood, and a consistent negative relation between personal distress and regulation
in adulthood (also see Davies et al 1998).
Relations between situational measures of empathy-related responding and
measures of dispositional regulation are considerably weaker than are analogous
relations for dispositional empathy-related responding. In a study with adults,
self-reported sympathy, sadness, and distress in response to empathy-inducing
films all were negatively related to a self-report measure of emotion regulation.
Facial reactions of sadness, distress, concern, and disgust to the empathy-inducing
films were unrelated to measures of regulation, and mens heart rate acceleration
during an evocative portion of the film (an index of personal distress) was negatively related to self-reported emotion regulation [but primarily for men exposed
to the relatively evocative film (Eisenberg et al 1994a)]. Thus, the relations of
measures of situational empathy-related responding to regulation varied with the
specific measure and sex of the individual.
Somewhat more consistent relation between situational sympathy and dispositional regulation have been found for children. Ungerer et al (1990) found that
4-month-olds who were low in self-regulation were prone to personal distress at
12 months of age. In two studies with preschoolers or school-aged children, positive relations were found between markers of sympathy (heart rate, facial, or selfreported) or empathic facial sadness and adults ratings of childrens attentional
and/or behavioral regulation (Eisenberg & Fabes 1995, Guthrie et al 1997). However, findings sometimes were obtained for only one sex, and often they were
weak. Thus, although situational measures of empathy-related responding tend to
be associated with regulation, especially for children, the findings are complex
and relatively modest in magnitude. Given that empathic responding in any particular context may not be a reliable index of general empathy-related dispositions, it is not really surprising that the relations between situational measures of
empathy-related responding and regulation are relatively small.
Cardiac vagal tone, which is substantially correlated with heart rate variance,
is considered to be an index of physiological emotion regulation and is believed
to promote calm and prosocial behavior [because of its inhibitory effect on sympathetic pathways to the heart (Porges 1997)]. Relations of these measures to
childrens empathy-related responding have been inconsistent. Some researchers
have found that high vagal tone is negatively associated with indices of sympathy
and prosocial behavior (Zahn-Waxler et al 1995); others have found positive
(Fabes et al 1993), mixed (Eisenberg et al 1996c), or no (Eisenberg et al 1998)
relations. The sample for which negative relations were found was selected to
include children at risk for externalizing problems, and the relation of vagal tone

676

EISENBERG

or heart rate variance [both of which are correlated with lack of behavioral inhibition for boys (Kagan 1998)] to empathy-related responding may vary with characteristics of the sample, as well as sex (Eisenberg et al 1996c, Fabes et al 1993).
Emotionality Because people prone to experience negative emotions would be
expected to be susceptible to vicariously induced emotion, Eisenberg & Fabes
(1992) and Eisenberg et al (1999a) hypothesized that people prone to intense
emotions (especially emotions such as sadness or distress rather than anger) are
prone to both sympathy and personal distress. In addition, because people who
are content may be less preoccupied with their own needs and better able to
respond to the needs of others, sympathy is hypothesized to be associated with
positive emotionality. In general, adults reports of dispositional sympathy and
personal distress, as well as empathy, have been positively related to intensity and
frequency of negative emotions (Davies et al 1998, Davis 1994, Eisenberg et al
1994a, Eisenberg & Okun 1996, Okun et al 1999). Moreover, adults self-reports
of sympathy, empathic sadness, and personal distress to an empathy-inducing film
(the latter of which may have reflected sympathy to some degree), as well as their
facial reactions and heart rate (for men), generally have been positively correlated
with intensity of self-reported dispositional emotionality and sadness, but not
frequency of negative emotionality (Eisenberg et al 1994a). In addition, positive
emotional intensity has been positively associated with sympathy and unrelated
or negatively related to personal distress (Eisenberg et al 1994a).
Findings for children differ from those for adults and vary with the measure
of empathy-related responding (i.e. empathy or sympathy) and emotion. Rothbart
et al (1994) found that mothers ratings of 7-year-olds dispositional empathy
(rather than sympathy) were uncorrelated with anger/frustration in infancy but
positively related to high fearfulness. Empathy was positively related to negative
emotionality (especially sadness) at age 7 when other aspects of temperament
(including regulation) were controlled in the analysis. Anger at age 7 was negatively related to empathy when regulation was controlled.
In a longitudinal study of childrens dispositional sympathy, Eisenberg and
colleagues (Eisenberg et al 1996c, 1998; Murphy et al 1999) found that parents
and teachers reports of school childrens intense and frequent negative emotions
tended to be negatively correlated with (or unrelated to) childrens dispositional
sympathy (as reported by teachers and sometimes the children). It is likely that
the adults reports of negative emotionality often reflected problematic negative
emotions such as anger or anxiety that might undermine sympathy over time. In
the same study, boys physiological arousal (heart rate and skin conductance)
when exposed to a relatively distressing film clip was related to low dispositional
sympathy. Thus, boys prone to physiological overarousal appeared to be low in
dispositional sympathy (Eisenberg et al 1996c; also see Strayer 1993).
Findings for situational measures of empathy/sympathy in children suggest an
association between empathy-related responding and both positive and negative
emotionality. In a study of toddlers, children who sustained a high level of empa-

MORAL DEVELOPMENT

677

thy and concern in response to simulated distress incidents (the measure appeared
to tap empathy, sympathy, and/or prosocial behavior) from 14 to 20 months of
age were observed to express more negative emotion and reported to express
more positive emotion at 14 months of age than did those who dropped in
empathy/sympathy. Those who increased in empathy/sympathy expressed more
positive emotions than those who remained low in empathy (Robinson et al 1994).
In studies of preschool or school-aged children, situational sympathy has been
negatively related to adults reports of childrens negative emotionality, whereas
facial expressions or self-reports of childrens situationally induced empathic sadness have been associated with adults reports of childrens emotional intensity
in general or intensity of negative emotions (Eisenberg & Fabes 1995, Guthrie et
al 1997). Childrens situational personal distress responses sometimes have been
positively correlated with negative emotionality (Guthrie et al 1997). Thus, it
appears that situational empathic distress and sadness tend to be positively related
with negative emotionality and emotional intensity in childhood (also see Roberts
& Strayer 1996), whereas situational measures of sympathy are related to low
negative emotionality, at least as rated by adults. It is likely that markers of
situational sympathy have been negatively related to negative emotionality
because the latter reflected adults perceptions of nonconstructive negative emotions (e.g. anger or anxiety) and sympathetic children are too regulated to express
high levels of such emotion.
In brief, there is some evidence that sympathy is positively related to intensity
of dispositional emotional responding and some kinds of negative emotions (sadness), especially among adults (who generally provide self-report data). This does
not mean that people who respond with sympathy necessarily react intensely to
empathy-inducing stimuli; as was discussed above, there is evidence that sympathetic individuals are relatively well regulated. Moreover, reports of frequent
negative emotionality tend to be associated with low levels of sympathy. Dispositional personal distress and empathy have been positively related to negative
emotional intensity and/or frequency of negative emotionality. It is likely that
empathy, personal distress, and sympathy relate somewhat differently to negative
emotionality and that it is important to differentiate among different types of
negative emotions (e.g. anger and sadness) and intensity vs frequency of negative
emotionality when examining these relations.

Interaction of Emotional Intensity and Regulation Eisenberg & Fabes (1992)

hypothesized that emotional intensity in general (i.e. with valence of emotion
unspecified) or for negative emotions such as sadness would be moderately associated with sympathy, although optimally regulated people were expected to be
somewhat sympathetic regardless of level of emotional intensity. If people can
modulate their emotions as needed, their dispositional emotionality should not be
an important contributor to empathy-related responding. In contrast, people high
in intensity of negative emotions would be expected to be prone to personal

678

EISENBERG

distress if they lack the ability to regulate themselves because they will become
overwhelmed by their vicariously induced negative feelings.
These predictions have been tested in moderational analyses. Although these
predictions have not been supported in studies of adults (Eisenberg et al 1994a,
Okun et al 1999), they have received some empirical support in research with
children. When participants in a longitudinal study were age 68, there was an
interaction between general emotional intensity and regulation when predicting
teacher-reported sympathy in children. Unregulated children were low in sympathy regardless of their general emotional intensity; such children were likely to
be overwhelmed by their vicarious emotion when it was experienced. In contrast,
for children who were moderate or relatively high in their regulation, sympathy
increased with the level of general emotional intensity. Thus, children who were
likely to be emotionally intense were sympathetic if they were at least moderately
regulated.
Two years later, childrens sympathy was predicted by a similar interaction
between behavioral regulation and general emotional intensity, but only for boys.
In addition, at this age, attention focusing was associated with sympathy for
children who were relatively low in emotional intensity. For children who are not
predisposed to experience intense emotions, the ability to focus on events outside
themselves may enhance sympathy by facilitating the intake of information about
others and, consequently, cognitive perspective taking (Eisenberg et al 1998).
Summary Research on empathy-related responding has remained a focus in the
study of prosocial behavior. Although it is clear that sympathy is associated with
prosocial behavior whereas personal distress reactions tend to be negatively or
unrelated to prosocial action, there still is debate regarding the nature of sympathetic motivation. Another focus of interest has been the relation of empathyrelated responding to emotionality and regulation, especially dispositional
differences in these aspects of temperament or personality. The emerging body
of research indicates that negative emotionality is related to empathy-related
responding, but that relations vary with the type of empathy-related response and
with the dimension (intensity or frequency) and type of negative emotion. More
work on the ways that individual differences in emotionality and regulation interact in predicting empathy-related responding will be necessary to understand the
role of emotion and its regulation in empathy-related responding.

The Development of Guilt, Shame, and Empathy

There has been some disagreement about the age at which guilt emerges (and
relatively little discussion about the emergence of shame). M Lewis (1998) has
argued that shame and guilt emerge at about age three, once children (a) can
clearly recognize the self as different from other people, (b) have developed some
standards of behavior, and (c) are able to use these standards to evaluate their
own behavior. Others (e.g. Barrett 1998) have suggested that the precursors of

MORAL DEVELOPMENT

679

guilt or shame are observed during the second and third years of life. Hoffman
(1998) described a developmental sequence in which prosocial actions and reparative behaviors (which often are viewed as evidence of guilt) both emerge from
early empathic capacities. Based on his theory, in the second year of life, as
children increasingly develop the ability to differentiate between their own and
others internal states, they are capable of becoming empathically involved in
others distress. In Hoffmans view, young children initially respond to others
distress with self-oriented distress, but they are increasingly able in the early years
to respond with other-oriented sympathetic concern. Empathy for a victim, combined with an awareness that one has caused anothers distress, is believed to
result in guilt, which motivates attempts at reparative behavior. Moreover, empathy or sympathy often motivates prosocial actions, even if the child did not cause
anothers distress or needy condition.
Consistent with the hypothesizing of Barrett and Hoffman, there is evidence
that 2-year-olds have some awareness of right and wrong and that they engage
in reparative behaviors (see Barrett 1995). Children 34 months old have some
understanding of the difference between moral and social conventional transgressions (Smetana & Braeges 1990). Empathic responding is observed in the
second year of life (Zahn-Waxler et al 1979, 1992), and children 2 to 3 years old
frequently show emotional reactions indicative of empathy and engage in reparative behavior in response to mishaps (Cole et al 1992). Moreover, parents report
that guilt increases from 14 to 24 months of age (Zahn-Waxler & Robinson 1995),
that remorse increases from 14 18 to 30 40 months (Stipek et al 1990), and that
discomfort about wrongdoing, apologizing, compliance with standards of conduct, and concern about others wrongdoing increase from 2133 months to 34
46 months (Kochanska et al 1994). Thus, it appears that precursors or rudimentary
forms of guilt are evident before age three and that guilt increases with age in the
early years. Empathy continues to increase with age in childhood (see Eisenberg
& Fabes 1998), but it is unclear whether these age-related changes are reflected
in developmental changes in guilt past the early years.
There also is evidence of a difference between shame and guilt responses in
2-year-olds. Barrett and colleagues (1993) observed toddlers reactions when they
were playing alone with an experimenters rag doll and a leg fell off. Some
toddlers (avoiders) displayed a shame-relevant pattern; they avoided the experimenter and delayed telling the experimenter about the mishap. Other children
(amenders) showed a guilt-relevant pattern of behavior. They repaired the doll
quickly, told the experimenter about the mishap shortly after the experimenter
returned, and showed relatively little avoidance of the experimenter (e.g. gaze
avoidance or active avoidance). The parents of amenders reported that their toddlers showed more guilt relative to shame at home than did parents of avoiders.
As might be expected, the development of conscience is associated with moral
behavior. For example, Kochanska et al (1994) found that children 26 41 months
old who exhibited evidence of a conscience (i.e. were reported by mothers to feel
affective discomfort over transgressions and to display evidence of spontaneous

680

EISENBERG

reparation, confession, attempts to regulate their behavior, and concern over others wrongdoing) transgressed less than their peers in an experimental context.
Moreover, during contrived mishaps, these childrens violations of standards were
associated with behavioral and affective responses indicative of guilt [e.g. acceptance of responsibility, apologies, focus on reparation, and distress (Kochanska
et al 1995)]. This association between evidence of conscience and moral behavior
also has been found in studies with preschoolers and elementary school children
(e.g. Lake et al 1995).

The Socialization of Guilt, Shame, and Empathy

Recently there has been a small burst of activity in assessing parental practices
and parenting styles associated with childrens guilt or conscience. In early work
on guilt, childrens guilt was linked to parental use of inductive-reasoning techniques [i.e. reasoning with the child about his/her behavior, e.g. You made Doug
cry. Its not nice to bite (Zahn-Waxler et al 1979)] and relatively low power
assertion [e.g. use of punishment or threats thereof (Hoffman 1977)]. Parental use
of induction is believed to foster sympathy, an other-orientation, and optimal
levels of attention to and learning about parental expectations and the reasons for
behaving in a moral manner (Hoffman 1983), especially if inductions are delivered with emotion and are used by loving parents (Hoffman 1977). Recently
researchers have replicated the findings pertaining to power assertion and inductions (Ferguson & Stegge 1995, Kochanska et al 1996b, Krevans & Gibbs 1996).
For example, Krevans & Gibbs (1996) found that children tended to be high on
empathy/sympathy and on the combination of empathy and guiltwhich is likely
to reflect other-oriented, empathy-based guiltwhen their parents used relatively
high amounts of inductive discipline. Moreover, it was found (DJ Laible & RA
Thompson, submitted for publication) that mothers references to feelings, needs,
or intentions and moral evaluative statements (e.g. good boy; this was a nice
thing to do) during conversations with their 4-year-olds were associated with
mothers reports of childrens guilt, remorse, and related reactions to a transgression or mishap, as well as with internalized compliance.
Findings concerning the relation of empathy and guilt to love withdrawal are
somewhat less consistent, both in the past (Hoffman 1983) and in recent work.
For example, Krevans & Gibbs (1996) found no relation between empathy/
sympathy (combined) or guilt and parental use of love withdrawal as discipline,
whereas Ferguson & Stegge (1995) found that love withdrawal was associated
with high loadings in a canonical correlation on both guilt and shame reactions.
It is likely that Ferguson & Stegges index of guilt reflected the general tendency
to evaluate oneself rather than empathy-based guilt. Moreover, these researchers
found that guilt, controlling for shame, was associated with the presence of parental anger in negative situations and parental pride reactions in positive encounters.
Kochanska and colleagues demonstrated that the relation between parental
socialization and the development of conscience often is moderated by charac-

MORAL DEVELOPMENT

681

teristics of the child. They typically have operationalized conscience as some

combination of guilt-related affect, an orientation toward reparation, and internalized compliance (which conceptually is less clearly linked to guilt). Kochanska
(1991) found that 8- to 10-year-olds affective/moral orientation (reflecting, in
part, their report of empathy and guilt when completing vignettes about transgressions) and childrens concern with reparation were positively related to maternal behavior deemphasizing the use of power assertion (based on both self-reports
and observed maternal behavior) when the children were toddlers. However, these
relations held primarily for children with a fearful/anxious temperament (Kochanska 1991). This finding was replicated in another sample at two ages (Kochanska
1995, 1997). Kochanska (1995) argued that, for fearful/anxious children, gentle
maternal discipline deemphasizing power results in an optimal, moderate level of
anxious arousal. A moderate level of arousal during disciplinary encounters is
viewed as motivating and optimal for the processing of information and not so
overwhelming that the child cannot attend to the information provided in the
disciplinary encounter (Hoffman 1983).
Investigators also have found an association between mutual positive affect or
a secure attachment between mother and child and childrens conscience or guilt
(Kochanska & Aksan 1995; DJ Labile & RA Thompson, submitted for publication) and empathy or sympathy (Kestenbaum et al 1989, Waters et al 1986; see
Eisenberg & Fabes 1998). This pattern of findings is consistent with the view that
a mutual interpersonal orientation between parent and child enhances the socialization process. However, a positive cooperative interactive set, as reflected in a
secure attachment between parent and child and maternal responsiveness, seems
to be especially important for the development of guilt in relatively fearless children (Kochanska 1995, 1997), a finding consistent with the notion that childrens
temperament moderates the association between parental socialization-related
behaviors and the development of conscience.
Moreover, the development of sympathy in children has been associated with
(a) parents being high in sympathy, (b) parents allowing their children to express
negative emotions that do not harm others, (c) low levels of hostile emotion in
the home, (d) parental practices that help children to cope with negative emotions,
and (e) parental practices that help children to focus on and understand others
emotions (Eisenberg et al 1991a, 1992; Fabes et al 1994; see Eisenberg & Fabes
1998). It is unclear whether parental practices that are supportive and help children to understand and deal with their emotions also foster empathy-based guilt
past the early years of life.
Although researchers seldom have differentiated between shame and guilt in
research on parental socialization, relatively recent findings suggest that the two
may be differentially related to parental socialization practices. Ferguson &
Stegge (1995) found that shame (when guilt was low) was predicted by high
parental anger and the absence of any discipline, including the absence of parental
induction, love withdrawal, and power assertion. Shame also was associated with
parents not responding positively to appropriate behavior. The combination of

682

EISENBERG

shame and guilt was predicted by an array of socialization responses by parents

(especially induction, but also including love withdrawal and, to a lesser extent,
power assertion). In contrast, in a study of childrens parenting experiences at age
5 and self-criticism at age 12, Koestner et al (1991) found that reports by 12year-olds of feeling guilty, of perceived incompetence, and of not living up to
their own standards were associated with same-sex parents restrictiveness and
rejection at age 5. Given the focus on chronic and global deficiencies of the self
in the measure of self-criticism, it is likely that these authors tapped shame as
much as guilt and that there is a positive relation between the development of
shame in children and parental anger, rejection, or the lack of appropriate
discipline.
Other work suggests that chronic and unjustified guilt can develop in children,
especially girls, in families with depressed mothers. For example, Zahn-Waxler
et al (1990) found that 5- to 9-year-old children of depressed mothers expressed
aberrant, distorted, and unresolved themes when responding to a semiprojective
procedure involving vignettes developed to elicit childrens narratives about interpersonal conflict and distress. Guilt responses are likely to be fused with shame
and may represent misplaced assignment of responsibility to the self when childrens guilt is assessed with responses to vignettes about negative events in which
the child is not unambiguously responsible (Ferguson et al 1999). Misplaced
responsibility may be based on a merging of guilt and empathy in young children,
especially daughters of depressed mothers, which makes them particularly vulnerable to false beliefs about their responsibility and blameworthiness for others
problems (Zahn-Waxler & Robinson 1995). Depressed mothers, in comparison
with well mothers, experience more guilt and irritability in their relationships
with their young children, so their children frequently are exposed to these emotions. Moreover, repeated exposure to a sad caregiver may increase the likelihood
that children will feel responsible for negative events simply because they are
there. In addition, depressed mothers may model a negative attributional style
(its my fault), and their children may experience more love withdrawal when
their mothers become less involved and emotionally unavailable because of their
depression (Zahn-Waxler & Kochanska 1990). However, there is little research
in which the reasons for guilt in children of depressed mothers have been tested
directly.

RELATIONS OF NONMORAL EMOTIONS TO MORALLY

RELEVANT BEHAVIOR
In recent research, a variety of primary, nonmoral emotions such as happiness,
sadness, and anger have been examined as predictors or correlates of moral behavior. Some of this research has pertained to situationally induced emotion, whereas
other research concerns dispositional emotion.

MORAL DEVELOPMENT

683

Mood
In the 1970s and 1980s, a popular topic of research was the relation between
temporary mood states (often experimentally induced) and such morally relevant
behavior as prosocial behavior and aggression. In this work, the focus was on the
typical effects of mood (across individuals) rather than on individual differences
in the effects of mood. Researchers found that positive mood is consistently
related to enhanced prosocial behavior and that a variety of mechanisms might
explain this association (Carlson et al 1988). In addition, in a meta-analytic
review, Carlson & Miller (1987) found an association between negative emotion
and helping, which varied with the degree to which attention was focused on the
self vs others, with helpers feelings of responsibility for the mood-lowering
event, and with a high level of objective self-awareness (i.e. the focusing of
attention on the self as an object).
In recent work, researchers focus has been primarily on the processes that
underlie the effects of positive and negative mood (e.g. Forgas 1995). For example, Wegener & Petty (1994) found that people in positive moods, in comparison
with those in sad or neutral states, tend to choose activities based on their affective
(hedonic) consequences for the self. This research has direct implications for the
type of helping behaviors people will engage in and their motives for assisting in
a positive mood. For example, such findings support the view that people in
positive moods often help to maintain their positive mood (see Carlson et al 1988).
Moreover, it appears that when adults experience negative emotional states elicited by threatening stimuli (aversive slides such as mutilation, starvation, a plane
crash, or a battered woman) or events (stress of impending exams), they make
decisions based on short-term outcomes regardless of possible long-term consequences. These findings can be interpreted as indicating that threat-related negative emotional states undermine the quality of cognitive processing and, as a
consequence, regulatory capacities (Gray 1999). Given the relation of regulation
to moral behavior (which is discussed shortly), negative moods owing to threatening stimuli likely predict impaired moral functioning.
Some recent work pertaining to temporary mood states and moral behavior or
cognition relates to feelings associated with perceived injustice. Anger and other
negative emotions (e.g. disgust and sadness) tend to be substantially linked with
the perception of injustice and immorality (Mikula et al 1998, Scher 1997).
Although appraisals of injustice often may elicit anger reactions, it also has been
argued that the experience of justice-related negative emotions such as guilt or
anger frequently leads to consideration of justice issues (Scher & Heise 1993;
also see Hoffman 1998, on empathic anger).
Situationally induced, directly experienced anger also has been associated with
morally relevant behavior and cognition. As is discussed below, situational and
trait anger predict externalizing problems (e.g. aggression). In addition, priming
anger increases adults punitive attributions and judgments of others in fictional
tort cases (Lerner et al 1998). People induced to feel anger also are likely to

684

EISENBERG

attribute responsibility or blame to others (Dix et al 1990, Keltner et al 1993),

which could increase the probability of aggressive behavior.
Thus, it is clear that temporary mood has some effect on a range of morally
relevant aspects of functioning. However, the role of mood in morally relevant
behavior has not received as much attention in the previous decade.

Individual Differences in Emotionality and Regulation

Emotionality Recently investigators interested in morality and emotion, especially developmentalists, have focused more on the role of individual differences
in emotionality in morality than on the effects of situational moods. Much of the
recent relevant work on the topic has pertained to aggression and externalizing
behavior. In general, children prone to intense and frequent negative emotions
(usually operationalized as a mix of different negative emotions such as dysphoria, anger, and anxiety) tend to exhibit relatively high levels of aggression and
externalizing problems (e.g. bullying, stealing, and lying) (Eisenberg et al 1996a,
Stice & Gonzales 1998). An association between temperamental negative emotionality and externalizing problems has been found across time and reporters
(Eisenberg et al 1995b, 1997a, 1999b), as well as when uncontaminated measures
of the constructs have been used [i.e. when overlapping items were removed
(Lengua et al 1998)]. Moreover, children prone to intense negative emotions tend
to deal nonconstructively with their anger (Eisenberg et al 1994b), and those prone
to intense externalizing and internalizing emotions (combined) may be low in
prosocial behavior (Eisenberg et al 1996b; see Eisenberg & Fabes 1998 for a
review). In addition, individual differences in intensity and frequency of negative
emotionality predict adolescent substance abuse/use, which sometimes is considered an externalizing behavior (Chassin et al 1993, Cooper et al 1995; cf. Stice
& Gonzales 1998).
Anger/frustration appears to be especially linked to externalizing problems.
Anger proneness in infancy as rated by mothers (Goldsmith 1996) or observed
in the laboratory (Rothbart et al 1994) has predicted aggression in the preschool
or early school years. Individual differences in typical intensity of anger reactions
have been related to the degree to which young childrens reactions to anger are
constructive (Eisenberg et al 1994b). Moreover, self-reported anger among high
school seniors predicted delinquency 9 months later, even controlling for earlier
levels of delinquency (Colder & Stice 1998). In another study, incarcerated juvenile offenders dispositional anger predicted aggressive behavior over the subsequent 3 months (Cornell et al 1999; also see Carlo et al 1998). In adulthood,
frustration in the workplace has been linked to antisocial behavior (Spector 1997).
Moreover, in situations involving provocation or harm, self-reported individual
differences in feelings of anger are associated with adults blaming others (Quigley & Tedeschi 1996) and have been found to mediate between attributions of
intentionality and nonconstructive aggression reactions to the provocateur (Graham et al 1997, 1992).

MORAL DEVELOPMENT

685

Some types of emotional and physiological reactivity may buffer children from
externalizing problems. Elevated levels of cortisol responding in novel settings
are, if associated at all, negatively related to externalizing problems in children
(Stansbury & Gunnar 1994). Cortisol reactions in these situations likely reflect
an emotional response to stress. Moreover, children and adolescents who are high
in their baseline physiological responding tend to be relatively low in delinquency
and other measures of externalizing problems (Mezzacappa et al 1997, Pine et al
1998, Raine 1993). In addition, children prone to internalizing emotions such as
fear are prone to low levels of aggression (e.g. Ladd & Profilet 1996, Rothbart
et al 1994) and are easily socialized on measures of conscience (e.g. Kochanska
1997). Thus, negative emotions such as fear or anxiety may serve to inhibit externalizing behavior, perhaps because individuals prone to these emotions are less
likely to find the stimulation and emotion associated with externalizing behavior
pleasurable and are likely to experience more negative emotion (anxiety, guilt,
and perhaps empathy) than other people when they engage in inappropriate behavior. Indeed, the tendency to be unemotional sometimes has been linked to antisocial behavior. For example, although children with psychopathic traits may be
prone to anger, they also are characterized as low in guilt and empathy, as callous,
and as shallow in their emotional responding (Frick 1998).
Regulation Attention to temperamental/personality regulation and its correlates
has increased substantially in recent years. Most measures of regulation pertain
to the control of overt (often emotionally induced) behaviors; others tap the regulation of attention or cognitions related to emotion or stress (see Eisenberg et al
1999a). Attentional modes of regulation are believed to be heavily involved in
the process of modulating emotional arousal, whereas capabilities such as the
ability to inhibit and activate behavior are believed to be particularly important
for modulating and regulating the behavioral expression of emotion.
An emerging body of work supports the assumption that individual differences
in regulatory behavior play a role in morally relevant behavior, as well as in social
competence more generally. In childhood, behavioral regulation has been associated with low externalizing problem behavior in numerous studies, sometimes
even when information about regulation and outcome variables was obtained from
different sources and when behavioral measures of regulation were obtained [e.g.
persistence on a task or delay of gratification (Eisenberg et al 1996a; Huey &
Weisz 1997; Krueger et al 1996; Lynam 1997; Oosterlaan & Sergeant 1996,
1998)]. In infancy and early childhood, the ability to inhibit and control ones
behavior has repeatedly been associated with a range of measures of conscience
and committed (internalized) compliance (e.g. following commands wholeheartedly, making reparation, cheating, and resistance to temptation), concurrently and
over time (Kochanska et al 1996a, 1997, 1998; Stifter et al 1999). Behavioral
regulation (including low impulsivity) also has been linked to low levels of adolescents substance abuse (e.g. Block et al 1988, Colder & Chassin 1997). In

686

EISENBERG

adulthood, disinhibition, which involves impulsivity and low behavioral control,

is associated with antisocial behavior, antisocial personality problems, and substance abuse (e.g. Clark & Watson 1999). Thus, temperamental or personality
traits such as impulsivity and voluntary behavioral inhibition appear to be intimately related to the development of conscience and antisocial behavior.
The ability to regulate attentional processes also seems to play an important
role in the development and enactment of morally relevant behavior. Attentional
regulation has been associated with high social competence and prosocial behavior (Eisenberg et al 1993, 1999a, 1997b; Ladd & Profilet 1996) and with low
problem behavior (Eisenberg et al 1996a, Ladd & Profilet 1996), as have composites of behavioral and attentional regulation (Eisenberg et al 1995b, Rothbart
et al 1994; also see Fabes et al 1997). Problems in attentional regulation, as tapped
by measures of executive cognitive functioning, have been linked to conduct
disorders (Moffitt 1993) and psychopathy (OBrien & Frick 1996, Patterson &
Newman 1993). Concentration problems in childhood also have been associated
with lower-level moral judgment in adolescence (Hart et al 1998).
Recently several groups of researchers identified personality types that reflect
undercontrolled, overcontrolled, and optimal styles of functioning in children
from Iceland (Hart et al 1997), the United States (Robins et al 1996), and New
Zealand (Newman et al 1997). In general, the well-adjusted (optimal) children
were resilient, self-assured, not emotionally labile, and in two samples (Hart et
al 1997, Newman et al 1997), attentionally and/or behaviorally regulated. Of most
relevance, the adjusted children were not prone to externalizing problems (Hart
et al 1997, Newman et al 1997). However, undercontrolled individuals, who
tended to be low in regulation and sometimes irritable and impulsive, were prone
to externalizing problem behaviors concurrently or later in adolescence or adulthood. In these three studies, the items used to classify children into the three
personality groups included ratings of a wide variety of social behaviors and/or
items pertaining to both regulation and emotionality. Thus, the investigators did
not examine individual differences in regulation and emotionality separate from
each other, from their social consequences, or from other temperamental or personality characteristics. Nonetheless, their findings converge with other research
in demonstrating an association between regulation and morally relevant
behaviors.
Prediction of Morally Relevant Behavior from the Combination of Emotionality
and Regulation The combination of negative emotionality and low regulation
may be especially problematic in regard to externalizing problems. In two major
longitudinal studies, researchers have used composite measures of combined negative emotionality and low regulation/impulsivity to predict externalizing problems over time. In the Dunedin, New Zealand, longitudinal sample, emotional
lability and negative emotionality at age three, when combined with lack of regulation (e.g. a short attention span and restlessness), predicted aggressive behavior

MORAL DEVELOPMENT

687

problems, criminality and conduct disorders, and antisocial behavior (but not
socialized delinquency) in adolescence (Caspi et al 1995) and antisocial behavior
(Newman et al 1997), criminality (Henry et al 1996), and antisocial personality
(Caspi et al 1996) in adulthood. For example, children identified as undercontrolled (i.e. emotionally labile, restless, with short attention spans, and high in
approach and negativism) at age 3 were 2.9 times as likely as adults to be diagnosed with antisocial personality disorder, 2.2 times as likely to be recidivistic
offenders, and 4.5 times as timely to be convicted for a violent offense (Caspi et
al 1996). Similarly, Pulkkinen (1996) and Pulkkinen & Hamalainen (1995)
assessed low self-control in childhood by using a measure that appeared to tap
emotionality as well as low regulation. Scores on low self-control tended to predict proactive aggression (aggression without provocation) in adolescence (particularly for boys) and criminal offenses in adulthood.
Often individual differences in emotionality and regulation predict unique,
additive variance in externalizing problem behavior, even though emotionality
and regulation obviously are correlated (Derryberry & Rothbart 1988, Eisenberg
et al 1993). For example, Eisenberg et al (1996a) found that both low regulation
and high negative emotionality provided significant, unique prediction of externalizing problem behavior (also see Eisenberg et al 1995b, 1997a). Rothbart and
colleagues (1994) obtained less evidence of additive effects, but they controlled
another aspect of temperament (surgency) in the regressions they used to assess
the unique effects of emotionality and regulation on aggression and defiant behavior (also see Lengua et al 1998).
In addition to main effects, both Rothbart & Bates (1998) and Eisenberg &
Fabes (1992) have emphasized the importance of examining moderational relations when assessing the prediction of adjustment from such aspects of temperament/personality as emotionality and regulation. Eisenberg and colleagues
(1999a) argued that emotionality, particularly negative emotionality, might have
fewer negative implications for behavior if the individual is well regulated.
Some research is consistent with this view, despite the difficulty of obtaining
interaction effects in small or moderately sized samples. For example, Eisenberg
et al (1996a) found that children who were low in negative emotionality (frequency and intensity) were low in externalizing problem behavior, regardless of
their level of regulation (attentional and behavioral regulation combined). However, for children who were more prone to negative emotion, higher regulation
often predicted less externalizing problems, sometimes even across reporters (i.e.
when reports of emotionality/regulation and externalizing behavior were obtained
from different adults). Similarly, Colder & Stice (1998) found that anger was
related to concurrent delinquency at higher but not lower levels of impulsivity
(although the relation was marginally significant even at lower levels of impulsivity). However, this interaction was not significant when predicting adolescents
delinquency 9 months later. In addition, somewhat similar interactions have been
obtained for both socially competent (Eisenberg et al 1995b) and prosocial behavior (Eisenberg et al 1996b) in other samples, although not in small samples for

688

EISENBERG

which the power to detect interaction effects was quite low (Eisenberg et al
1997a). Furthermore, there is some evidence, albeit not entirely consistent, that
emotionality and regulation sometimes interact when predicting substance abuse
problems in adolescence (Colder & Chassin 1997, Colder & Stice 1998; see
Eisenberg et al 1999a, for a review).
Few researchers have used longitudinal designs and statistics that optimize the
investigators ability to make inferences about causality from correlational data.
In a follow-up of the longitudinal sample of Eisenberg et al (1996b), similar
moderational relations were found 2 years later with structural equation modeling.
At two ages, negative emotionality moderated the relation of attentional regulation to childrens externalizing problems. The relation of attentional regulation to
low externalizing problems was stronger for children who were low rather than
high in regulation. Findings for behavioral regulation were in the same direction,
but the moderating effect was weak and nonsignificant in the best structural equation model (so behavioral regulation predicted low externalizing behavior for all
children). Of particular interest, the aforementioned pattern of relations in children
in grades 3 to 6 held even when the effects of externalizing behavior from 2 years
earlier were taken into consideration. Thus, consistency over time in externalizing
problems did not account for the relation between emotionality/regulation and
externalizing problems at the 2-year follow-up (Eisenberg et al 1999b).
Thus, it appears that behaviors of moral relevance are predicted not only by
regulation or emotionality (especially negative emotionality) in isolation, but also
by the combination of the two. Findings such as these suggest that individual
differences both in the tendency to experience negative emotions and in the ability
to modulate emotional arousal should be considered when theorizing about and
predicting moral development and behavior.

SUMMARY
Recent research highlights the importance of emotionality and emotion-related
regulation in the study of moral development and behavior. Currently, relevant
work is scattered throughout different bodies of literature and generally has not
been integrated. As the construct of emotion continues to permeate psychological
theory and research, knowledge about the role of emotion and its regulation in
morality is likely to increase. Moreover, empirical work in the field is starting to
move from attention to mere correlation to concern about moderating influences,
mediational processes, and the direction of causality between morally relevant
variables and emotionality and regulation. An important problem with the existing
literature is the confounding of measures in the research (i.e. overlap of items
measuring the various constructs), and this issue also is beginning to receive
attention (e.g. Lengua et al 1998, Sanson et al 1990). Thus, it is likely that research
on the contributions of emotionality and regulation to moral development and
behavior will be conceptually and methodologically stronger in the next decade

MORAL DEVELOPMENT

689

and will be increasingly integrated with our developing knowledge of the role of
emotion in human functioning.
ACKNOWLEDGMENTS
The writing of this chapter was supported by a Research Scientist Award (K05
M801321) and a grant (1 R01 HH55052) from the National Institutes of Mental
Health.
Visit the Annual Reviews home page at www.AnnualReviews.org.

LITERATURE CITED
Barrett KC. 1995. A functionalist approach to
shame and guilt. In Self-Conscious Emotions, ed. JP Tangney, KW Fischer, pp. 25
63. New York: Guilford
Barrett KC. 1998. The origins of guilt in early
childhood. In Guilt and Children, ed. J
Bybee, pp. 7590. San Diego: Academic
Barrett KC, Zahn-Waxler C, Cole PM. 1993.
Avoiders versus amendersimplication for
the investigation of guilt and shame during
toddlerhood? Cogn. Emot. 7:481505
Batson CD. 1997. Self-other merging and the
empathy-altruism hypothesis: reply to Neuberg et al. (1997). J. Pers. Soc. Psychol.
73:51722
Batson CD. 1998. Altruism and prosocial
behavior. In The Handbook of Social Psychology, ed. DT Gilbert, ST Fiske, G Lindzey, 2:282316. Boston: McGraw-Hill
Batson CD, Ahmad N, Yin J, Bedell SJ, Johnson JW, et al. 1999. Two threats to the common good: self-interested egoism and
empathy-induced altruism. Pers. Soc. Psychol. Bull. 25:316
Batson CD, Polycarpou MP, Harmon-Jones E,
Imhoff HJ, Mitchener EC, et al. 1997a.
Empathy and attitudes: Can feelings for a
member of a stigmatized group improve
feelings toward the group? J. Pers. Soc.
Psychol. 72:10518
Batson CD, Sager K, Garst E, Kang M, Rubchinsky K, Dawson K. 1997b. Is empathyinduced helping due to self-other merging?
J. Pers. Soc. Psychol. 73:495509

Batson CD, Turk CL, Shaw LL, Klein TR.

1995. Information function of empathic
emotion: learning that we value the others
welfare. J. Pers. Soc. Psychol. 68:30013
Baumeister RF, Stillwell AM, Heatherton TF.
1994. Guilt: an interpersonal approach.
Psychol. Bull. 115:24367
Ben-Zeev A. 1997. Emotions and morality. J.
Value Inq. 31:195212
Block J, Block JH, Keyes S. 1988. Longitudinally foretelling drug usage in adolescence: early childhood personality and
environmental precursors. Child Dev.
59:33655
Blum LA. 1980. Friendship, Altruism, and
Morality. London: Routledge & Kegan
Paul. 234 pp.
Carlo G, Allen JB, Buhman DC. 1999. Facilitating and disinhibiting prosocial behavior
behaviors: the nonlinear interaction of trait
perspective taking and trait personal distress on volunteering. Basic Appl. Soc. Psychol. In press
Carlo G, Roesch SC, Melby J. 1998. The multiplicative relations of parenting and temperament to prosocial and antisocial
behaviors in adolescence. J. Early Adolesc.
18:26690
Carlson M, Charlin V, Miller N. 1988. Positive
mood and helping behavior: a test of six
hypotheses. J. Pers. Soc. Psychol. 55:211
29
Carlson M, Miller N. 1987. Explanation of the
relation between negative mood and helping. Psychol. Bull. 102:91108

690

EISENBERG

Caspi A, Henry B, McGee RO, Moffitt TE,

Silva PA. 1995. Temperamental origins of
child and adolescent behavior problems:
from age three to age fifteen. Child Dev.
66:5568
Caspi A, Moffitt TE, Newman DL, Silva PA.
1996. Behavioral observations at age 3 predict adult psychiatric disorders. Arch. Gen.
Psychiatry 53:103339
Chassin L, Pillow DR, Curran PJ, Molina
BSG, Barrera M Jr. 1993. Relations of
parental alcoholism to early adolescent substance use: a test of three mediating mechanisms. J. Abnorm. Psychol. 102:319
Cialdini RB, Brown SL, Lewis BP, Luce C,
Neuberg SL. 1997. Reinterpreting the
empathy-altruism relationship: when one
into one equals oneness. J. Pers. Soc. Psychol. 73(3):48194
Clark LA, Watson D. 1999. Temperament: a
new paradigm for trait psychology. In
Handbook of Personality, ed. L Pervin, O
John. San Francisco: Guilford. In press. 2nd
ed.
Cohen D, Strayer J. 1996. Empathy in conductdisordered and comparison youth. Dev.
Psychol. 32:98898
Colder CR, Chassin L. 1997. Affectivity and
impulsivity: temperament risk for adolescent alcohol involvement. Psychol. Addict.
Behav. 11:8397
Colder CR, Stice E. 1998. The moderating
effect of impulsivity on the relationship
between anger and adolescent problem
behavior: cross-sectional and prospective
findings. J. Youth Adolesc. 27:25574
Cole PM, Barrett KC, Zahn-Waxler C. 1992.
Emotion displays in two-year-olds during
mishaps. Child Dev. 63:31424
Cooper ML, Frone MR, Russell M, Mudar P.
1995. Drinking to regulate positive and
negative emotions: a motivational model of
alcohol use. J. Pers. Soc. Psychol. 69:990
1005
Cornell DG, Peterson CS, Richards H. 1999.
Anger as a predictor of aggression among
incarcerated adolescents. J. Consult. Clin.
Psychol. 67:10815

Davies M, Stankov L, Roberts RD. 1998. Emotional intelligence: in search of an elusive

construct. J. Pers. Soc. Psychol. 75:989
1015
Davis MH. 1994. Empathy: A Social Psychological Approach. Madison, WI: Brown &
Benchmark
Denham SA, Renwick-DeBardi S, Hewes S.
1994. Emotional communication between
mothers and preschoolers: relations with
emotional competence. Merrill-Palmer Q.
40:488508
Derryberry D, Rothbart MK. 1988. Arousal,
affect, and attention as components of temperament. J. Pers. Soc. Psychol. 55:95866
Dix T, Reingold DP, Zambarano RJ. 1990.
Mothers judgments in moments of anger.
Merrill-Palmer Q. 36:46586
Einstein D, Lanning K. 1998. Shame, guilt, ego
development, and the five-factor model of
personality. J. Pers. 66: 55582
Eisenberg N, Carlo G, Murphy B, Van Court
P. 1995a. Prosocial development in late
adolescence: a longitudinal study. Child
Dev. 66:91136
Eisenberg N, Fabes RA. 1991. Prosocial
behavior and empathy: a multimethod,
developmental perspective. In Review of
Personality and Social Psychology, ed. M
Clark, 12:3461. Newbury Park, CA: Sage
Eisenberg N, Fabes RA. 1992. Emotion, regulation, and the development of social competence. In Review of Personality and
Social Psychology: Emotion and Social
Behavior, ed. MS Clark, 14:11950. Newbury Park, CA: Sage
Eisenberg N, Fabes RA. 1995. The relation of
young childrens vicarious emotional
responding to social competence, regulation, and emotionality. Cogn. Emot. 9:203
29
Eisenberg N, Fabes RA. 1998. Prosocial development. In Handbook of Child Psychology.
Social, Emotional, and Personality Development, ed. W Damon, N Eisenberg (ser.
ed). 3:70178. New York: Wiley & Sons
Eisenberg N, Fabes RA, Bernzweig J, Karbon
M, Poulin R, Hanish L. 1993. The relations

MORAL DEVELOPMENT
of emotionality and regulation to preschoolers social skills and sociometric status.
Child Dev. 64:141838
Eisenberg N, Fabes RA, Carlo G, Troyer D,
Speer AL, et al. 1992. The relations of
maternal practices and characteristics to
childrens vicarious emotional responsiveness. Child Dev. 63:583602
Eisenberg N, Fabes RA, Guthrie IK, Murphy
BC, Maszk P, et al. 1996a. The relations of
regulation and emotionality to problem
behavior in elementary school children.
Dev. Psychopathol. 8:14162
Eisenberg N, Fabes RA, Guthrie IK, Reiser M.
1999a. Dispositional emotionality and
regulation: their role in predicting quality
of social functioning. J. Pers. Soc. Psychol.
In press
Eisenberg N, Fabes RA, Karbon M, Murphy
BC, Wosinski M, et al. 1996b. The relations
of childrens dispositional prosocial behavior to emotionality, regulation, and social
functioning. Child Dev. 67:97492
Eisenberg N, Fabes RA, Murphy B, Karbon M,
Maszk P, et al. 1994a. The relations of emotionality and regulation to dispositional and
situational empathy-related responding. J.
Pers. Soc. Psychol. 66:77697
Eisenberg N, Fabes RA, Murphy B, Karbon M,
Smith M, Maszk P. 1996c. The relations of
childrens dispositional empathy-related
responding to their emotionality, regulation, and social functioning. Dev. Psychol.
32:195209
Eisenberg N, Fabes RA, Murphy M, Maszk P,
Smith M, Karbon M. 1995b. The role of
emotionality and regulation in childrens
social functioning: a longitudinal study.
Child Dev. 66:123961
Eisenberg N, Fabes RA, Nyman M, Bernzweig
J, Pinuelas A. 1994b. The relations of emotionality and regulation to childrens angerrelated reactions. Child Dev. 65:10928
Eisenberg N, Fabes RA, Schaller M, Carlo G,
Miller PA. 1991a. The relations of parental
characteristics and practices to childrens
vicarious emotional responding. Child Dev.
62:1393408

691

Eisenberg N, Fabes RA, Schaller M, Miller PA,

Carlo et al. 1991b. Personality and socialization correlates of vicarious emotional
responding. J. Pers. Soc. Psychol. 61:459
71
Eisenberg N, Fabes RA, Shepard SA, Murphy
BC, Guthrie IK, et al. 1997a. Contemporaneous and longitudinal prediction of childrens social functioning from regulation
and emotionality. Child Dev. 68:64244
Eisenberg N, Fabes RA, Shepard SA, Murphy
BC, Jones J, Guthrie IK. 1998. Contemporaneous and longitudinal prediction of childrens sympathy from dispositional
regulation and emotionality. Dev. Psychol.
34:91024
Eisenberg N, Guthrie IK, Fabes RA, Reiser M,
Murphy BC, et al. 1997b. The relations of
regulation and emotionality to resiliency
and competent social functioning in elementary school children. Child Dev.
68:295311
Eisenberg, N, Guthrie, IK, Fabes RA, Shepard
S, Losoya S, et al. 1999b. Prediction of elementary school childrens externalizing
problem behaviors from attentional and
behavioral regulation and negative emotionality. Child Dev. In press
Eisenberg N, Guthrie IK, Murphy BC, Shepard
SA, Cumberland A, Carlo G. 1999c. Consistency and development of prosocial dispositions: a longitudinal study. Child Dev.
In press
Eisenberg N, Miller PA, Shell R, McNalley S,
Shea C. 1991c. Prosocial development in
adolescence: a longitudinal study. Dev. Psychol. 27:84957
Eisenberg N, Okun MA. 1996. The relations of
dispositional regulation and emotionality to
elders empathy-related responding and
affect while volunteering. J. Pers. 64:157
83
Estrada P. 1995. Adolescents self-reports of
prosocial responses to friends and acquaintances: the role of sympathy-related cognitive, affective, and motivational processes.
J. Res. Adolesc. 5:173200

692

EISENBERG

Eyre HL, Ferguson TJ. 1997. Do you see what

I see? Self- and other-reports of guilt and
shame. Presented at Am. Psychol. Assoc.,
Washington, DC
Fabes RA, Eisenberg N, Eisenbud L. 1993.
Behavioral and physiological correlates of
childrens reactions to others distress. Dev.
Psychol. 29:65563
Fabes RA, Eisenberg N, Karbon M, Bernzweig
J, Speer AL, Carlo G. 1994. Socialization
of childrens vicarious emotional responding and prosocial behavior: relations with
mothers perceptions of childrens emotional reactivity. Dev. Psychol. 30:4455
Fabes RA, Shepard S, Guthrie I, Martin CL.
1997. The roles of temperamental arousal
and same-sex play in childrens social
adjustment. Dev. Psychol. 33:693702
Ferguson T, Sorenson C, Bodrero R, Stegge H.
1996. (Dys)functional guilt and shame in
developmental perspective. Presented at
Bienn. Meet. Int. Soc. Study Behav. Dev.
Quebec City, Canada
Ferguson TJ, Eyre HL. 1999. Engendering
gender differences in shame and guilt: stereotypes, socialization, and situational pressures. In Gender and Emotion, ed. A
Fischer. Cambridge, UK: Cambridge Univ.
Press. In press
Ferguson TJ, Eyre HL, Stegge H, Sorenson
CB, Everton R. 1997. The distinct roles of
shame and guilt in childhood psychopathology. Presented at Soc. Res. Child Dev.,
Washington, DC
Ferguson TJ, Stegge H. 1995. In Self-Conscious Emotions, ed. JP Tangney, KW
Fischer, pp. 17497. New York: Guilford
Ferguson TJ, Stegge H. 1998. Measuring guilt
in children: a rose by any other name still
has thorns. In Guilt and Children, ed. J
Bybee, pp. 1974. San Diego: Academic
Ferguson TJ, Stegge H, Damhuis I. 1991. Childrens understanding of guilt and shame.
Child Dev. 62:82739
Ferguson TJ, Stegge H, Miller ER, Olsen ME.
1999. Guilt, shame, and symptoms in children. Dev. Psychol. 35:34757

Forgas JP. 1994. Sad and guilty? Affective

influences on the explanation of conflict in
close relationships. J. Pers. Soc. Psychol.
66:5668
Forgas JP. 1995. Mood and judgment: the
affect infusion model (AIM). Psychol. Bull.
117:3966
Frick PJ. 1998. Callous-unemotional traits and
conduct problems: applying the two-factor
model of psychopathy to children. In Psychopathy: Theory, Research and Implications for Society, ed. DJ Cooke, A Forth,
RD Hare, et al, pp. 16187. Amsterdam:
Kluwer Academic
Goldsmith HH. 1996. Studying temperament
via construction of the toddler behavior
assessment questionnaire. Child Dev.
67:21835
Graham S, Hudley C, Williams E. 1992. Attributional and emotional determinants of
aggression among African-American and
Latino young adolescents. Dev. Psychol.
28:73140
Graham S, Weiner B, Zucker, GS. 1997. An
attributional analysis of punishment goals
and public reactions to O.J. Simpson. Pers.
Soc. Psychol. Bull. 23:33146
Gray JR. 1999. A bias toward short-term thinking in threat-related negative emotional
states. Pers. Soc. Psychol. Bull. 25:6575
Green JD, Sedikides C. 1999. Affect and selffocused attention revisited: the role of affect
orientation. Pers. Soc. Psychol. Bull.
25:10419
Guthrie IK, Eisenberg N, Fabes RA, Murphy
BC, Holmgren R, et al. 1997. The relations
of regulation and emotionality to childrens
situational empathy-related responding.
Motiv. Emot. 21:87108
Harder DW, Cutler L, Rockart L. 1992. Assessment of shame and guilt and their relationships to psychopathology. J. Pers. Assess.
59:584604
Hart D, Hofmann V, Edelstein W, Keller M.
1997. The relation of childhood personality
type to adolescent behavior and development: a longitudinal study of Icelandic children. Dev. Psychol. 33:195205

MORAL DEVELOPMENT
Hart D, Keller M, Edelstein W, Hofmann V.
1998. Childhood personality influences on
social-cognitive development: a longitudinal study. J. Pers. Soc. Psychol. 74:1278
89
Hastings PD, Zahn-Waxler C. 1998. Psychophysiological and socialization predictors
of empathy and externalizing problems in
middle childhood. Presented at the annual
conference of the Am. Psychol. Assoc.
August 98, San Francisco
Henry B, Caspi A, Moffitt TE, Silva PA. 1996.
Temperamental and familial predictors of
violent and nonviolent criminal convictions: age 3 to age 18. Dev. Psychol.
32:61423
Hoffman ML. 1977. Moral internalization: current theory and research. In Advances in
Experimental Social Psychology, ed. L Berkowitz, 10:86135. New York: Academic
Hoffman ML. 1982. Development of prosocial
motivation: empathy and guilt. In The
Development of Prosocial Behavior, ed. N
Eisenberg, pp. 281313. New York:
Academic
Hoffman ML. 1983. Affective and cognitive
processes in moral internalization. In Social
Cognition and Social Development: A
Sociocultural Perspective, ed. ET Higgins,
DN Ruble, WW Hartup, pp. 23674. Cambridge, UK: Cambridge Univ. Press
Hoffman ML. 1998. Varieties of empathybased guilt. In Guilt and Children, ed. J
Bybee, 4:91112. New York: Academic
Holmgren RA, Eisenberg N, Fabes RA. 1998.
The relations of childrens situational
empathy-related emotions to dispositional
prosocial behavior. Int. J. Behav. Dev.
22:16993
Huey SJ, Weisz JR. 1997. Ego control, ego
resiliency, and the five-factor model as predictors of behavioral and emotional problems in clinic-referred children and
adolescents. J. Abnorm. Psychol. 106:404
15
Kagan J. 1998. Biology and the child. In
Social, Emotional and Personality Development. Handbook of Child Psychology, ed.

693

W Damon (ser. ed.), N Eisenberg (vol. ed.),

3:177235. New York: Wiley & Sons
Keltner D. 1995. The signs of appeasement:
evidence for the distinct displays of embarrassment, amusement, and shame. J. Pers.
Soc. Psychol. 68:44154
Keltner D, Buswell B. 1997. Embarrassment:
its distinct form and appeasement functions. Psychol. Bull. 122:25070
Keltner D, Ellsworth PC, Edwards K. 1993.
Beyond simple pessimism: effects of sadness and anger on social perception. J. Pers.
Soc. Psychol. 64:74052
Kestenbaum R, Farber EA, Sroufe LA. 1989.
Individual differences in empathy among
preschoolers: relation to attachment history.
In New Directions for Child Development,
Vol. 44. Empathy and Related Emotional
Responses, ed. N Eisenberg, pp. 5164. San
Francisco: Jossey-Bass
Knight GP, Johnson LG, Carlo G, Eisenberg
N. 1994. A multiplicative model of the dispositional antecedents of a prosocial behavior: predicting more of the people more of
the time. J. Pers. Soc. Psychol. 66:17883
Kochanska G. 1991. Socialization and temperament in the development of guilt and conscience. Child Dev. 62:137992
Kochanska G. 1995. Childrens temperament,
mothers discipline, and security of attachment: multiple pathways to emerging internalization. Child Dev. 66:597615
Kochanska G. 1997. Multiple pathways to conscience for children with different temperaments: from toddlerhood to age 5. Dev.
Psychol. 33:22840
Kochanska G, Aksan N. 1995. Mother-child
mutually positive affect, the quality of child
compliance to requests and prohibitions,
and maternal control as correlates of early
internalization. Child Dev. 66:23654
Kochanska G, Casey RJ, Fukumoto A. 1995.
Toddlers sensitivity to standard violations.
Child Dev. 66:64356
Kochanska G, DeVet K, Goldman M, Murray
K, Putnam SP. 1994. Maternal reports of
conscience development and temperament
in young children. Child Dev. 65:85268

694

EISENBERG

Kochanska G, Murray K, Coy KC. 1997.

Inhibitory control as a contributor to conscience in childhood: from toddler to early
school age. Child Dev. 68:22842
Kochanska G, Murray K, Jacques TY, Koenig
AL, Vandegeest KA. 1996a. Inhibitory control in young children and its role in emerging internalization. Child Dev. 67:490507
Kochanska G, Padavich DL, Koenig AL.
1996b. Childrens narratives about hypothetical moral dilemmas and objective measures of their conscience: mutual relations
and socialization antecedents. Child Dev.
67:142036
Kochanska G, Tjebkes TL, Forman DR. 1998.
Childrens emerging regulation of conduct:
restraint, compliance, and internalization
from infancy to the second year. Child Dev.
69:137889
Koestner R, Zuroff DC, Powers TA. 1991.
Family origins of adolescent self-criticism
and its continuity into adulthood. J.
Abnorm. Psychol. 100(2):19197
Krevans J, Gibbs JC. 1996. Parents use of
inductive discipline: relations to childrens
empathy and prosocial behavior. Child Dev.
67:326377
Krueger RF, Caspi A, Moffitt TE, White J,
Stouthamer-Loeber M. 1996. Delay of gratification, psychopathology, and personality:
Is low self-control specific to externalizing
problems? J. Pers. 64:10729
Ladd GW, Profilet SM. 1996. The Child
Behavior Scale: a teacher-report measure of
young childrens aggressive, withdrawn,
and prosocial behaviors. Dev. Psychol.
32:100824
Lake NL, Lane S, Harris PL. 1995. The expectation of guilt and resistance to temptation.
Early Dev. Parenting 4:6373
Leary MR, Landel JL, Patton KM. 1996. The
motivated expression of embarrassment
following a self-presentational predicament. J. Pers. 64:61936
Lengua LJ, West SG, Sandler IN. 1998. Temperament as a predictor of symptomatology
in children: addressing contamination of
measures. Child Dev. 69:16481

Lerner JS, Goldberg JH, Tetlock PE. 1998.

Sober second thought: the effects of
accountability, anger, and authoritarianism
on attributions of responsibility. Pers. Soc.
Psychol. Bull. 24:56374
Lewis HB. 1971. Shame and Guilt in Neurosis.
New York: Int. Univ. Press
Lewis M. 1998. Emotional competence and
development. In Improving Competence
Across the Lifespan, ed. D Pushkar, WM
Bukowski, AE Schwartzman, DM Stack,
DR White, pp. 2736. New York: Plenum
Lynam DR. 1997. Pursuing the psychopathy:
capturing the fledgling psychopath in a
nomological net. J. Abnorm. Psychol.
106:42538
Mezzacappa E, Tremblay R, Kindlon D, Saul
J, Arseneault L, et al. 1997. Anxiety, antisocial behavior, and heart rate regulation in
adolescent males. J. Child Psychiatry Psychol. 38:45769
Mikula G, Scherer KR, Athenstaedt U. 1998.
The role of injustice in the elicitation of differential emotional reactions. Pers. Soc.
Psychol. Bull. 24:76983
Miller PA, Eisenberg N, Fabes RA, Shell R.
1996. Relations of moral reasoning and
vicarious emotion to young childrens prosocial behavior toward peers and adults.
Dev. Psychol. 32:21019
Miller RS, Leary MR. 1992. Social sources and
interactive functions of emotion: the case of
embarrassment. In Emotion and Social
Behavior, ed. MS Clark, 8:20221. Newbury Park, CA: Sage
Miller RS, Tangney JP. 1994. Differentiating
embarrassment and shame. J. Soc. Clin.
Psychol. 13:27387
Moffitt TE. 1993. The neuropsychology of
conduct disorder. Dev. Psychopathol.
5:13551
Murphy BC, Shepard SA, Eisenberg N, Fabes
RA, Guthrie IK. 1999. Contemporaneous
and longitudinal relations of young adolescents dispositional sympathy to their emotionality, regulation, and social functioning.
J. Early Adolesc. 29:6697
Neuberg SL, Cialdini RB, Brown SL, Luce C,
Sagarin BJ, Lewis BP. 1997. Does empathy

MORAL DEVELOPMENT
lead to anything more than superficial helping? Comment on Batson et al (1997). J.
Pers. Soc. Psychol. 73:51016
Newman DL, Caspi A, Moffitt TE, Silva PA.
1997. Antecedents of adult interpersonal
functioning: effects of individual differences in age 3 temperament. Dev. Psychol.
33:20617
Niedenthal PM, Tangney JP, Gavanski I. 1994.
If only I werent versus If only I
hadnt: distinguishing shame and guilt in
counterfactual thinking. J. Pers. Soc. Psychol. 67:58495
OBrien BS, Frick PJ. 1996. Reward dominance: associations with anxiety, conduct
problems, and psychopathy in children. J.
Abnorm. Child Psychol. 24:22339
OConnor LE, Berry JW, Weiss J. 1999. Interpersonal guilt, shame, and psychological
problems. J. Soc. Clin. Psychol. In press
Okun MA, Shepard SA, Eisenberg N. 1999.
The relations of emotionality and regulation
to dispositional empathy-related responding among volunteers-in-training. Ind. Differ. In press
Oosterlaan J, Sergeant JA. 1996. Inhibition in
ADHD, aggressive, and anxious children: a
biologically based model of child psychopathology. J. Abnorm. Child Psychol.
24:1936
Oosterlaan J, Sergeant JA. 1998. Effects of
reward and response cost on response inhibition in ADHD, disruptive, anxious, and
normal children. J. Abnorm. Child Psychol.
26:16174
Patterson CM, Newman JP. 1993. Reflectivity
and learning from aversive events: toward
a psychological mechanism for the syndromes of disinhibition. Psychol. Rev.
100:71636
Penner L, Finkelstein MA. 1998. Dispositional
and structural determinants of volunteerism. J. Pers. Soc. Psychol. 74:52537
Pine D, Wasserman G, Miller L, Coplan J,
Bagiella E, et al. 1998. Heart period variability and psychopathology in urban boys
at risk for delinquency. Psychophysiology
35:52129

695

Porges SW. 1997. Emotion: an evolutionary

by-product of the neural regulation of the
autonomic nervous system. Ann. NY Acad.
Sci. 807:6277
Pulkkinen L. 1996. Proactive and reactive
aggression in early adolescence as precursors to anti- and prosocial behavior in
young adults. Aggress. Behav. 22:24157
Pulkkinen L, Hamalainen M. 1995. Low selfcontrol as a precursor to crime and accidents in a Finnish longitudinal study. Crim.
Behav. Ment. Health 5:42438
Quigley BM, Tedeschi JT. 1996. Mediating
effects of blame attributions on feelings of
anger. Pers. Soc. Psychol. Bull. 22:128088
Quiles Z, Bybee J. 1997. Chronic and predispositional guilt: relations to mental health,
prosocial behavior, and religiosity. J. Pers.
Assess. 69:10426
Raine A. 1993. The Psychobiology of Crime.
New York: Academic
Roberts W, Strayer J. 1996. Empathy, emotional expressiveness, and prosocial behavior. Child Dev. 67:44970
Robins RW, John OP, Caspi A, Moffitt TE,
Stouthamer-Loeber M. 1996. Resilient,
overcontrolled, and undercontrolled boys:
three replicable personality types. J. Pers.
Soc. Psychol. 70:15771
Robinson JL, Zahn-Waxler C, Emde RN. 1994.
Patterns of development in early empathic
behavior: environmental and child constitutional influences. Soc. Dev. 3:12545
Rothbart MK, Ahadi SA, Hershey KL. 1994.
Temperament and social behavior in childhood. Merrill-Palmer Q. 40:2139
Rothbart MK, Bates JE. 1998. Temperament.
In Handbook of Child Psychology. Vol. 3.
Social, Emotional, Personality Development, ed. W Damon (ser. ed.), N Eisenberg
(vol. ed.), 3:10576. New York: Wiley &
Sons
Sanson A, Prior M, Kyrios M. 1990. Contamination of measures in temperament
research. Merrill-Palmer Q. 36:17992
Scher SJ. 1997. Measuring the consequences
of injustice. Pers. Soc. Psychol. Bull.
23:48297

696

EISENBERG

Scher SJ, Heise DR. 1993. Affect and the perception of injustice. Adv. Group Proc.
10:22352
Smetana JG, Braeges JL. 1990. The development of toddlers moral and conventional
judgments. Merrill-Palmer Q. 36:32946
Sorenson CB, Ferguson TJ, Eyre HL. 1997.
ASC and ye shall find: measuring shame
and guilt. Presented at the Eastern Psychol.
Assoc., Washington, DC
Spector PE. 1997. The role of frustration in
antisocial behavior at work. In Anti-Social
Behavior in Organizations, ed. RA Jiacalone, J Greenberg, pp. 117. Thousand
Oaks, CA: Sage
Stansbury K, Gunnar MR. 1994. Adrenocortical activity and emotion regulation. Mongr.
Soc. Res. Child Dev. 59(240):108134
Stice E, Gonzales N. 1998. Adolescent temperament moderates the relation of parenting to antisocial behavior and substance
use. J. Adolesc. Res. 13:531
Stifter CA, Spinrad TL, Braungart-Reiker JM.
1999. Toward a developmental model of
child compliance: the role of emotion regulation in infancy. Child Dev. 70:2132
Stipek D, Gralinski H, Kopp C. 1990. Selfconcept development in the toddler years.
Dev. Psychol. 26:97277
Strayer J. 1993. Childrens concordant emotions and cognitions in response to
observed emotions. Child Dev. 64:188201
Tangney JP. 1991. Moral affect: the good, the
bad, and the ugly. J. Pers. Soc. Psychol.
61:598607
Tangney JP. 1992. Situational determinants of
shame and guilt in young adulthood. Pers.
Soc. Psychol. Bull. 18:199206
Tangney JP. 1996. Conceptual and methodological issues in the assessment of shame
and guilt. Behav. Res. Ther. 34:74154
Tangney JP. 1998. How does guilt differ from
shame? In Guilt and Children, ed. J Bybee,
pp. 117. San Diego: Academic
Tangney JP, Burggraf SA, Wagner PE. 1995.
Shame-proneness, guilt-proneness, and
psychological symptoms. In Self-Conscious
Emotions: The Psychology of Shame, Guilt,

Embarrassment, and Pride, ed. JP Tangney,

KW Fischer, pp. 34367. New York:
Guilford
Tangney JP, Miller RS, Flicker L, Barlow DH.
1996a. Are shame, guilt, and embarrassment distinct emotions? J. Pers. Soc. Psychol. 70:125669
Tangney JP, Niedenthal PM, Covert MV, Barlow DH. 1998. Are shame and guilt related
to distinct self-discrepancies? A test of Higgins (1987) hypotheses. J. Pers. Soc. Psychol. 75:25668
Tangney JP, Wagner P, Gramzow R. 1992.
Proneness to shame, proneness to guilt, and
psychopathology. J. Abnorm. Psychol.
101:46978
Tangney JP, Wagner PE, Hill-Barlow D, Marschall DE, Gramzow R. 1996b. Relation of
shame and guilt to constructive versus
destructive responses to anger across the
lifespan. J. Pers. Soc. Psychol. 70:797809
Tremblay RE, Pihl RO, Vitaro F, Dobkin PL.
1994. Predicting early onset of male antisocial behavior from preschool behavior.
Arch. Gen. Psychiatry 51:73239
Trobst KK, Collins RL, Embree JM. 1994. The
role of emotion in social support provision:
gender, empathy and expressions of distress. J. Soc. Pers. Relat. 11:4562
Ungerer JA, Dolby R, Waters B, Barnett B,
Kelk N, Lewin V. 1990. The early development of empathy: self-regulation and
individual differences in the first year.
Motiv. Emot. 14:93106
Walker LJ, Pitts RC. 1998. Naturalistic conceptions of moral maturity. Dev. Psychol.
34:40319
Waters E, Hay D, Richters J. 1986. Infantparent attachment and the origins of prosocial and antisocial behavior. In
Development of Antisocial and Prosocial
Behavior: Research, Theories, and Issues,
ed. D Olweus, J Block, M Radke-Yarrow,
pp. 97125. Orlando, FL: Academic
Watson D, Clark LA. 1992. Affects separable
and inseparable: on the hierarchical
arrangement of the negative affects. J. Pers.
Soc. Psychol. 62:489505

MORAL DEVELOPMENT
Wegener DT, Petty RE. 1994. Mood management across affective states: the hedonic
contingency hypothesis. J. Pers. Soc. Psychol. 66:103448
Wood JV, Saltzberg JA, Goldsamt LA. 1990.
Does affect induce self-focused attention?
J. Pers. Soc. Psychol. 58:899908
Zahn-Waxler C, Cole PM, Welsh JD, Fox NA.
1995. Psychophysiological correlates of
empathy and prosocial behaviors in preschool children with problem behaviors.
Dev. Psychopathol. 7:2748
Zahn-Waxler C, Kochanska G. 1990. The origins of guilt. Annu. Nebr. Symp. Motiv.
Socioemot. Dev., 36th, Lincoln, pp. 183
258. Lincoln: Univ. Nebr. Press

697

Zahn-Waxler C, Kochanska G, Krupnick J,

McKnew D. 1990. Patterns of guilt in children of depressed and well mothers. Dev.
Psychol. 26:5159
Zahn-Waxler C, Radke-Yarrow M, King RA.
1979. Child rearing and childrens prosocial
initiations toward victims of distress. Child
Dev. 50:31930
Zahn-Waxler C, Radke-Yarrow M, Wagner E,
Chapman M. 1992. Development of concern for others. Dev. Psychol. 28:12636
Zahn-Waxler C, Robinson J. 1995. Empathy
and guilt: early origins of feelings of
responsibility. In Self-Conscious Emotions,
ed. JP Tangney, KW Fischer, pp. 14373.
New York: Guilford