Psychological Assessment 2013 American Psychological Association

2013, Vol. 25, No. 3, 679 691 1040-3590/13/$12.00 DOI: 10.1037/a0032297

The Basic Empathy Scale in Adults (BES-A):

Factor Structure of a Revised Form

Arnaud Carr Nicolas Stefaniak

Universit de Reims Champagne-Ardenne Universit de Reims Champagne-Ardenne and
University of Liege

Fanny DAmbrosio and Lela Bensalah Chrystel Besche-Richard

Universit de Reims Champagne-Ardenne Universit de Reims Champagne-Ardenne and Institut
Universitaire de France
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
This document is copyrighted by the American Psychological Association or one of its allied publishers.

Initially thought of as a unitary ability, empathy has been more recently considered to consist of 2
components (i.e., an affective and a cognitive component). The Basic Empathy Scale (BES) is a tool that
has been used to assess empathy in young people and adolescents on the basis of this dual-component
conception (Jolliffe & Farrington, 2006). Recent studies of empathy have led to it being defined as
underpinned by 3 components, namely, emotional contagion, emotional disconnection, and cognitive
empathy. The aims of this study were (a) to validate the BES in Adults and (b) to compare the different
conceptions of empathy. Three hundred seventy French adults took part in the study, and 160 of them
filled out complementary scales measuring empathy, alexithymia, and emotional consciousness. The
confirmatory factor analyses showed that the 3-factor model was the model that was best able to account
for the data. Complementary tools confirmed the relationships previously observed between empathy as
assessed with the BES and other scales assessing emotional processes. The results of this study make it
clear that empathy can be seen as process-dependent. This conception of empathy, which is based on 3
factors, is consistent with the current, more integrated view of empathy. The implications of this
conception and the opportunity to use the 2 or 3 factors of the BES in adults are presented in the
Discussion.

Keywords: empathy, emotion, contagion, emotional regulation, Basic Empathy Scale (BES)

Empathy is a heterogeneous construct that has received consid- cesses have also been found to be impaired in schizophrenia, in
erable attention during the last few years (Decety & Svetlova, which a negative correlation has been observed between negative
2012). Empathy has recently been accorded an important role due symptomatology and the automatic emotional processes involved
to the increasing interest shown by researchers in issues relating to during the early stages of empathy (Haker & Rssler, 2009).
social cognition. For instance, empathy has been shown to be Similarly, the automatic processes of emotion perception and
negatively correlated with aggressive behaviors (L. E. Marshall & identification as well as the cognitive strategies involved in the
Marshall, 2011; Mehrabian, 1997). This impairment in empathy processing of empathy are impaired in autism spectrum disorder
has not only been observed in aggressors. For example, depression (Clark, Winkielman, & McIntosh, 2008; Schulte-Rther et al.,
has also been associated with empathic disturbance (Thoma et al., 2011).
2011), with depressed patients usually being more sensitive to both Interest in empathy has not been limited solely to the field of
the distress and affective states of other people. Empathic pro- psychopathology, and the way it is conceived of has changed

This article was published Online First July 1, 2013. The research was supported in part by a scientific grant from the
Arnaud Carr, Department of Psychology, Research Unit Cognition, Champagne-Ardenne Regional Council (Arnaud Carr) and the Institut
Health, Socialization (C2S, EA 6291), University of Reims Champagne- Universitaire de France (Chrystel Besche-Richard). Arnaud Carr received
Ardenne, Reims, France; Nicolas Stefaniak, Department of Psychology, a Doctoral Research Fellowship from the Champagne-Ardenne Regional
Research Unit Cognition, Health, Socialization (C2S, EA 6291), Uni- Council (2009 2012). Fanny DAmbrosio received a Doctoral Research
versity of Reims Champagne-Ardenne, Reims, and Department of Cogni- Fellowship from the Champagne-Ardenne Regional Council (20072010).
tive Science, University of Liege, Lige, Belgium; Fanny DAmbrosio and Arnaud Carr is now ATER (Attach Temporaire dEnseignement et de
Lela Bensalah, Department of Psychology, Research Unit Cognition, Recherche/Teaching and Research Assistant) at the University of Reims
Health, Socialization (C2S, EA 6291), University of Reims Champagne- Champagne-Ardenne (20122013).
Ardenne; Chrystel Besche-Richard, Department of Psychology, Research Correspondence concerning this article should be addressed to Arnaud
Unit Cognition, Health, Socialization (C2S, EA 6291), University of Carr, Laboratoire C2S, Cognition Sant Socialisation (EA6291), Univer-
Reims Champagne-Ardenne, and Institut Universitaire de France (IUF), sit de Reims Champagne-Ardenne, Reims, France. E-mail: arnaud.carre@
Paris, France. univ-reims.fr

679
 680 CARR ET AL.

considerably since the early pioneering studies (Lipps, 1979; the ability to experience another persons emotion). Even though
Titchener, 1909). Rogers (1951) defined empathy as the concept the IRI has long been used to assess empathy, it has also been the
of the as if (p. 129), which means that empathy makes it object of considerable criticism. According to Jolliffe and Far-
possible to understand another persons views and his or her rington (2004, 2006), the perspective-taking component of the IRI
feelings. This includes the ability to feel similar emotions and is not limited to the understanding of an emotion but assesses a
understand their causes. Viewed within this perspective, empathy broader ability to adopt the other persons viewpoint even when
has been considered to be a unique ability. This initial conception emotions are not involved, as in IRI Item 25 (when I am upset at
highlights the main difference between empathy, in which a certain someone, I usually try to put myself in his shoes for a while;
distance and a distinction is maintained between self and oth- Davis, 1980). Moreover, Joliffe and Farrington also argued that the
ers, and a complete process of identification. Empathic Concern subscale confounds empathy with sympathy
However, the way in which empathy is thought of has changed because this subscale is designed to assess other-oriented feelings
in order, more specifically, to account for the processes involved in of sympathy and concern for unfortunate others (Davis, 1980, p.
both emotion processing and social interactions. This shift led to 114). This view is also supported by Batson, Early, and Salvaranis
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

the emergence of a conception of empathy based on two compo- (1997) results. Indeed, these authors showed that asking people to
This document is copyrighted by the American Psychological Association or one of its allied publishers.

nents: an affective and a cognitive component (Davis, 1983a, imagine how they would feel in a particular situation leads not
1983b; Deutsch & Madle, 1975; Hoffman, 1977; Hogan, 1969; only to an empathic response but also to self-oriented distress and
Jolliffe & Farrington, 2006; Lawrence, Shaw, Baker, Baron- a compassionate response. In other words, this subscale assesses
Cohen, & David, 2004; Mehrabian & Epstein, 1972). According to sympathy, which is different from empathy. These results weaken
this view, empathy is an essential part of both emotional function- the Empathic Concern subscale of the IRI because the items of this
ing and interpersonal cognition, making individuals particularly subscale are formulated in a way that asks people to determine
attentive to both the mental states and emotions of other people. how they would feel in a particular situation. For instance, there is
Furthermore, it is necessary to distinguish between this concept nothing in IRI Item 2 (I often have tender, concerned feelings for
and a simple sensitivity to distress (Batson, Fultz, & Schoenrade, people less fortunate than me; Davis, 1980) to suggest that a
1987; Decety, 2010). According to this view, appropriate empathic person less fortunate than the responder is experiencing his or her
responses are the result of the efficient functioning of several situation in a negative way. Furthermore, affective empathy is not
processes. At a fundamental level, if people are to be responsive to assessed any better by the Personal Distress subscale because all
the emotional states of others needs, then they must also be the corresponding items focus on emergency situations (e.g., IRI
sufficiently attentive to their emotions (Decety & Jackson, 2004; Item 5 [In emergency situations, I feel apprehensive and ill-at-
W. L. Marshall, Hudson, Jones, & Fernandez, 1995). ease; Davis, 1980]). However, empathic response does not nec-
The empathic response requires the recognition of ones own essarily involve emergencies. These issues explain why the IRI is
and other peoples emotions. It also requires the ability to share not the best way to assess the complexity of the processes involved
and replicate other peoples emotional states while simultaneously in empathy.
being aware that these emotions are not ones own (i.e., affective In order to overcome the weaknesses of the IRI, Jolliffe and
responsiveness). In addition, it demands the ability to adopt an- Farrington (2006) developed the Basic Empathy Scale (BES),
other persons perspective while simultaneously preserving the which focuses on two factors (i.e., cognitive and affective factors
distinction between self and other (emotional perspective taking). of empathy) and four basic emotions (i.e., anger, fear, happiness,
Finally, it requires individuals to choose the best socioemotional and sadness). In the BES, affective empathy is defined as the
response (e.g., by soothing a sad person without being as sad as ability to feel an appropriate emotional response when one is
this person). Although the empathic response is often confounded confronted with the mental state attributed to another person
with that of sympathy, these two concepts are different. Indeed, (Bryant, 1982), and cognitive empathy is defined as the under-
according to Eisenberg (2010), sympathy frequently stems from standing of another persons affective state (Hogan, 1969). Unlike
empathy but can be distinguished from it in that it consists of in the IRI, the Cognitive subscale of the BES is limited to under-
feeling an emotion for the other person rather than feeling an standing why another person feels a specific emotion (e.g., I can
emotion as the other feels it or is expected to feel it. According to understand my friends happiness when she/he does well at some-
Joliffe and Farrington (2006), the emotion felt in sympathy is not thing), whereas the Affective Empathy subscale focuses on how
necessarily the same as the emotion felt by the other person. another persons emotions are felt without any reference to an
Although empathy can result in individuals feeling negative emo- emergency situation (e.g., After being with a friend who is sad
tions on behalf of another person, it is sufficiently modulated not about something, I usually feel sad).
to cause personal distress. However, the BES, which can be considered to be a two-factor
Curiously, the range of tools developed in order to assess scale, does not take account of the most recent conception of
empathy is very limited. Davis (1983b) was the first to contribute empathy. Indeed, a number of studies have suggested that empathy
by developing the Interpersonal Reactivity Index (IRI; Davis, depends on three components (e.g., Decety, 2011a; Decety &
1983b). This scale is itself further subdivided into two scales Michalska, 2010). First, emotional contagion is thought to corre-
(affective and cognitive components), and also includes several spond to the automatic replication of another persons emotions
cognitive (i.e., Fantasy and Perspective Taking) and affective (i.e., (Iacoboni & Dapretto, 2006; Lipps, 1979). Second, cognitive em-
Personal Distress and Empathic Concern) subscales. It makes it pathy is defined as the ability to understand and mentalize another
possible to describe the cognitive processes involved in empathy persons affects (Decety, 2011b). The mechanism of cognitive
(i.e., the ability to understand another persons emotion) and to empathy is therefore thought to be distinct from emotional conta-
characterize the style of an individuals emotional functioning (i.e., gion and automatic identification (Hoffman, 1977, 2000; W. L.
 EMPATHY AND THE BASIC EMPATHY SCALE IN ADULTS 681

Marshall et al., 1995). Third, emotional disconnection is seen as a empathy in healthy adults (BES-A). Second, to determine the best
regulatory factor that involves self-protection against distress, factor structure for the BES-A, we compared three conceptions of
pain, and extreme emotional impact (Batson et al., 1987; Lamm, empathy, namely, the single-factor model, the usual two-factor
Batson, & Decety, 2007). One argument suggesting that this three- model of the BES (Albiero, Matricardi, Speltri, & Toso, 2009;
factor model might be appropriate is its consistency with devel- Albiero, Matricardi, & Toso, 2010; DAmbrosio et al., 2009;
opmental and neuroimaging studies of empathy (Decety, 2010, Jolliffe & Farrington, 2006; Li, Lv, Liu, & Zhong, 2011), and the
2011a; Decety & Jackson, 2004; Decety & Svetlova, 2012), and three-factor model, which has emerged from the most recent
the need to see empathy as an active process based on functional developments in the study of empathy (Decety, 2010, 2011b). We
and dynamic mechanisms that are involved in social contexts and wanted to determine which of these models provides the best
account for the processes involved in empathy (Gerdes, Lietz, & account of the psychometric properties of the BES-A. Further-
Segal, 2011). Indeed, recent research suggests that empathy, as more, because empathy is related to emotional functioning, we
well as the processing and regulation of emotions, depends on both evaluated the relationship between empathy and the dimensions of
bottom-up and top-down processing (Decety & Svetlova, 2012; emotional consciousness assessed using the Emotional State Ques-
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

Gyurak, Goodkind, Kramer, Miller, & Levenson, 2012). Accord- tionnaire (ESQ; Cass-Perrot, Fakra, Jouve, & Blin, 2007). We
This document is copyrighted by the American Psychological Association or one of its allied publishers.

ing to this approach, the bottom-up component of empathy would expected that the BES-A subscales would be correlated with the
relate to affective sharing or contagion, whereas the top-down ESQ. Moreover, a number of previous studies have found rela-
processes, which involve the regulation of emotions and inten- tionships between empathy and alexithymia, indicating that a
tional mechanisms, would modulate empathic experience (i.e., the person who finds it difficult to express or identify his or her own
cognitive empathy component of the three-factor model) and could feelings is likely to experience similar difficulties when confronted
be regulated by emotional disconnection (i.e., the third component with another persons emotions (DAmbrosio et al., 2009; Gryn-
of the three-factor model), which could itself correspond to an berg, Luminet, Corneille, Grzes, & Berthoz, 2010; Jolliffe &
emotional suppression (Gross, 2002; Lamm et al., 2007). Farrington, 2006). We expected to find a link between impair-
Neuropsychological studies also seem to support the idea that ments in empathy and alexithymia as assessed using the Toronto
three components are involved in empathy. Emotional contagion, Alexithymia Scale (TAS-20; Bagby, Taylor, & Ryan, 1986). Fi-
for example, is thought to involve automatic processes that permit nally, in order to exclude the possibility that the score on the
the rapid evaluation of the nature of the emotion and whether it is BES-A, which is a self-report questionnaire, might be related to a
positive or negative, pleasant or aversive. Due to its automatic desirability bias, we also included the Social Desirability Scale
character, it involves subcortical structures such as the limbic lobe, (SDS; Crowne & Marlowe, 1960). We did not expect to find any
which is known to be involved in emotion processing (Derntl et al., correlation between the SDS and the BES-A.
2010; Hariri, Tessitore, Mattay, Fera, & Weinberger, 2002; Phil-
lips, Drevets, Rauch, & Lane, 2003). The second component,
which relates to cognitive empathy (Decety & Svetlova, 2012), Method
involves activations of the insular cortex, which promotes emo-
tional awareness, as well as of the ventromedial prefrontal cortex Participants
(PFC) and the medial PFC, which are responsible for the under-
standing of emotions (Decety, 2011a). The third component of Three hundred seventy participants (260 women and 110 men)
empathic functioning makes it possible to regulate emotions were recruited. The participants consisted of psychology or social
through the mechanism of emotional disconnection. This appears science students (248 participants) as well as working (118) (em-
to be related to executive functions that are implemented in a ployees) and retired people (four participants). According to the
top-down network based on the orbitofrontal cortex, medial and classification by the Institut National de la Statistique et des Etudes
dorsolateral prefrontal cortex, and the anterior cingulate cortex Economiques (French National Institute of Statistics and Eco-
(Decety, 2011a; Decety & Michalska, 2010). nomic Studies; INSEE), 26 of the working participants were con-
From a developmental viewpoint, the emotional contagion pro- sidered as belonging to Category 3 (i.e., senior executive), 18 to
cesses are the first component to appear. After this, more cognitive Category 4 (i.e., middle executive), 28 to Category 5 (employee)
(i.e., cognitive empathy) and regulatory (i.e., emotional disconnec- and 23 to Category 6 (i.e., manual worker). The other participants
tion) functions develop in parallel with cognitive and cerebral were unemployed persons (12), artisans (five), or agricultural
maturation. Indeed, both these components of empathy are closely workers (one). This information was not provided for six persons.
related to the development of both theory of mind and executive The participants were recruited on a voluntary basis. A brief
functions, which evolve later (Zelazo, Carlson, & Kesek, 2008). screening questionnaire was used at the start of the study in order
Among the different tools used to assess empathy, the BES to collect information about gender, age, level of education, and
appears to avoid some of the weaknesses present in the IRI (Joliffe mother tongue. These characteristics are detailed in Table 1. A
& Farrington, 2004) and is therefore a suitable instrument to use self-report questionnaire was then administered to identify past or
when seeking to account for the characteristics of empathy. Al- present anxious, depressive, neurological or somatic disorders,
though the BES has been translated and support for a two-factor addictions, and drug consumption. Participants with current or past
structure has been found in French adolescents (DAmbrosio, disorders were excluded from the analyses. Sixty percent of the
Olivier, Didon, & Besche, 2009), its factor structure has not yet sample (n 222) completed the scale twice at an interval of 7
been examined in French adults. The first aim of the present study weeks in order to establish testretest reliability. The participants
was to verify that the score on the BES, which was initially were volunteers who completed and signed a consent form. The
developed for young people, is a valid measure for assessing study was designed in accordance with the Declaration of Helsinki.
 682 CARR ET AL.

Table 1 items to the definition of emotional disconnection (DIS; Items 1, 7,

Characteristics of the Participants 8, 13, 18, 19) (e.g., I am not usually aware of my friends
feelings). In order to ensure that the items were correctly assigned
Variable M (SD) to the corresponding factors, we also asked two independent per-
Age (years) (SD) 26.05 (12.41) sons to classify the items on the basis of the definitions used in the
Level of education 12.58 (1.29) model. Both persons classified 19 items correctly. The item that
Social Desirability Scale 16.49 (5.03) was classified in another scale was Item 6 and was interpreted as
IRI (n 160) corresponding to emotional disconnection. However, after being
FS 26.57 (5.17)
EC 26.63 (4.68) asked why they did not consider the item to correspond to cogni-
PT 23.25 (4.48) tive empathy, both of them spontaneously changed their decision
PD 22.63 (4.92) and classified the item as cognitive empathy. Items are shown in
ESQ (n 278) Appendix A.
Total score 24.17 (6.45)
The SDS. The SDS (Crowne & Marlowe, 1960) is one of the
REC 6.7 (1.54)
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

EXPRES 6.42 (1.53) tools that is most frequently used to assess the tendency of indi-
This document is copyrighted by the American Psychological Association or one of its allied publishers.

IEE 5.63 (1.69) viduals to respond by distorting their own self-representations in a

SC 6.1 (1.91) way that reflects social desirability and a need for approval (An-
TAS-20 (n 160) drews & Meyer, 2003; Leite & Beretvas, 2005). The SDS is a scale
Total score 57.13 (8.69)
DIF 17.02 (5.32) consisting of 33 truefalse items that cover two kinds of behaviors:
DDF 14.32 (3.15) acceptable but doubtful behavior or unacceptable but plausible
EOT 25.78 (3.19) behavior. Given that this scale reflects a social desirability bias
Note. IRI Interpersonal Reactivity Index; FS Fantasy; EC Em- (i.e., a lack of honesty in the responses), no significant correlations
pathic Concern; PT Perspective Taking; PD Personal Distress; were expected between the SDS scores and the BES.
ESQ Emotional State Questionnaire; REC Recognition; EXPRES The ESQ. The ESQ (Cass-Perrot et al., 2007) is a short
Expression; IEE Internal Emotional Experience; SC Social Context; self-report measure of emotional consciousness. The 33 items are
TAS-20 Toronto Alexithymia Scale; DIF Difficulties in Identifying
distributed into four dimensions, that is, recognition (REC), ex-
Feelings; DDF Difficulties in Describing Feelings; EOT Externally
Oriented Thinking. pression (EXPRES), internal emotional experience (IEE), and so-

gender differences with p .05 using a t test. gender differences with cial context (SC). This tool assesses participants impression of
p .001 using a t test. their own ability to encode and decode emotions. Each dimension
is rated on a 10-point scale. The items in the ESQ are based on
fundamental emotions (happiness, sadness, anger, fear, disgust,
One hundred sixty participants from the overall sample involved and surprise) and are mostly assessed on the basis of the perception
in this study completed three supplementary scales: the TAS-20 of familiar faces. This makes it possible to evaluate emotional
(Bagby et al., 1986; Loas, Otmani, Verrier, Fremaux, & Marchand, profiles and their variations during changes to emotional states
1996), the IRI (Davis, 1983b; Grynberg et al., 2010), and the ESQ (Besnier et al., 2010; Cass-Perrot et al., 2007).
(Cass-Perrot et al., 2007). These three scales were optional and To the best of our knowledge, no norms or cutoff values relating
were completed at the same time as the initial completion of the to the emotional consciousness assessed with the ESQ have as yet
BES. been published for the general population. However, these values
have been published in a study of changes in emotional conscious-
ness in psychiatrists and psychologists (Besnier et al., 2010). A
Measures
comparison of Besnier et al.s (2010) results with those obtained in
The BES. The BES is a 20-item scale developed by Jolliffe the present study suggests that the sample in the present study
and Farrington (2006). The items in the French version that we obtained slightly higher scores.
used in the present study were back-translated by DAmbrosio et The IRI. The IRI (Davis, 1983a, 1983b) is one of the most
al. (2009). The participants had to give their ratings on a 5-point widely used self-report questionnaires for the study of empathy. It
Likert type scale (1 Strongly Disagree, 2 Disagree, 3 is a 5-point Likert type scale consisting of 28 items. The IRI
Neither Agree nor Disagree, 4 Agree, 5 Strongly Agree). contains four subscales (Fantasy-Empathy, Perspective Taking,
In the two-factor model (Jolliffe & Farrington, 2006), nine items Empathic Concern, Personal Distress). Two subscales assess the
assess cognitive empathy (Items 3, 6, 9, 10, 12, 14, 16, 19, 20), and cognitive dimension: Perspective Taking (PT) and fantasy (FS).
11 items assess affective empathy (Items 1, 2, 4, 5, 7, 8, 11, 13, 15, The affective dimension is also assessed by two subscales: Per-
17, 18). In the two-factor conceptualization, the BES included sonal Distress (PD) and Empathic Concern (EC). More specifi-
seven reversed items and the scores could range from 20 (deficit in cally, Perspective Taking refers to the ability to take into account
empathy) to 100 (high level of empathy). One of the authors of the both the views and mental states of others. Fantasy measures
present study also classified the items according to the definitions fictional identification. Personal Distress assesses the tendency to
used in the three-factor model. According to this classification, six feel anxious in negative situations and experience a lack of control
items were thought to relate to the definition of emotional conta- in emotional conditions. Finally, the Empathic Concern subscale
gion (CONT; Items 2, 4, 5, 11, 15, 17) (e.g., I get caught up in evaluates feelings toward others and the ability to worry about
other peoples feeling easily ), eight items to the definition of them. An examination of norms for the IRI (Davis, 1980) revealed
cognitive empathy (EMP; Items 3, 6, 9, 10, 12, 14, 16, 20) (e.g., I that our French sample obtained lower scores than the original
have trouble figuring out when my friends are happy), and six sample, which consisted of students.
 EMPATHY AND THE BASIC EMPATHY SCALE IN ADULTS 683

The TAS-20. Alexithymia is defined as an impairment in the

processing of emotions (Taylor, Bagby, & Parker, 1997). The
relations between empathy (assessed with the BES) and alexithy-
mia (assessed by the TAS) were assessed because alexithymia
shares characteristics with empathy at the level of interpersonal
cognition (Grynberg et al., 2010). The 20-item version of the
TAS-20 (Bagby, Parker, & Taylor, 1994; Loas et al., 1996) is
composed of three subscales: Difficulties in Identifying Feelings
(DIF), Difficulties in Describing Feelings (DDF), and Externally
Oriented Thinking (or attention to external events). The maximum
score that can be obtained on the TAS-20 is 100. On the basis of
French norms (Guilbaud et al., 2002; Loas et al., 1996), our sample
appeared to obtain scores slightly above the mean in the TAS-20
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

(M 46.2; SD 10.52), with a few scores being above the

This document is copyrighted by the American Psychological Association or one of its allied publishers.

general cutoff for alexithymia (56). The scores obtained in the

present study are presented in Table 1.

Statistical Analysis
The confirmatory factor analyses were conducted using LISREL
(Jreskog, 1990; Jreskog & Srbom, 2004). Confirmatory factor
analyses were performed on the polychoric correlations. We used
Bravais-Pearson test to examine the correlations between the BES
and the other scales used in the study.

Results

Psychometric Properties of the BES-A

A confirmatory factor analysis was performed on the polychoric
correlations using the diagonally weighted least squares parameter
estimation method in order to assess the factor structure of the BES
in the French-speaking adult population. We tested the two-factor
structure that was originally proposed by Joliffe and Farrington Figure 1. Two-factor structure of the Basic Empathy Scale in Adults
(2006) and had previously been tested in French youths (BES-A).
(DAmbrosio et al., 2009). When performing the analyses, we
treated residual errors as uncorrelated. This analysis revealed that
the data were fitted to the model, 2(169) 510.65, p .001, 2010; Favre, Joly, Reynaud, & Salvador, 2005, 2009). In order to
root-mean-square error of approximation (RMSEA) .074 (90% assess this model, we redefined the different items of the BES-A
CI [.067, .81]), goodness-of-fit index (GFI) .95, adjusted in the light of the definition of each of these factors (see the
goodness-of-fit index (AGFI) .94. These different indices re- Method section). This model also fitted the data, 2(167)
vealed a reasonably good fit between the data and the model 460.10, p .001, RMSEA .069 (90% CI [.061, .077]), GFI
because the chi-square value was not greater than 3 times the .96, AGFI .95. The AIC was smaller for the three-factor model
degree of freedom, the RMSEA was less than .10, and both the (AIC 546.10) than for the two-factor model (AIC 592.65).
GFI and the AGFI were greater than 0.90 (see Figure 1). Chi-square was also significantly better in the three-factor model
We tested two other models of empathy in order to determine than in the two-factor model, 2(2) 50.55, p .001. These
whether they might provide a better account of the data. The first results suggest that the three-factor model can also be used to
of these was the single-factor model. The different adjustment account for the data. The loadings of the items on the correspond-
criteria suggested that this model did not account for the data as ing factors in this model are presented in Figure 2.
well as a model in which the affective and the cognitive factors are Given that the loading of Item 4 was weak both in the two-factor
dissociated, 2(170) 776.54, p .001, RMSEA .098 (90% CI model and in the three-factor model, we explored the possibility
[.091, .110]), GFI .93, AGFI .91. These results were con- that errors could covary. However, the addition of error covariance
firmed by the Akaikes information criterion (AIC), which was in the model did not improve the loading of Item 4. Given that we
smaller for the two-factor model (AIC 592.65) than for the could not improve the loading of Item 4, we excluded this item
single-factor model (AIC 856.54), as well as by the significant from the analyses. This exclusion improved both the two- and
improvement in chi-square between the unidimensional and the three-factor models. However, the three-factor model remained
affective cognitive model, 2(1) 265.89, p .001. statistically better than the two-factor model and yielded the fol-
The second alternative model that we examined was the three- lowing indices: 2(149) 372.28 p .001, RMSEA .064 (90%
factor model (e.g., Decety, 2011a, 2011b; Decety & Michalska, CI [.056, .072]), GFI .97, AGFI .96. The AIC was smaller for
 684 CARR ET AL.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
This document is copyrighted by the American Psychological Association or one of its allied publishers.

Figure 3. Three-factor structure of Basic Empathy Scale in Adults

(BES-A) with 19 items.
Figure 2. Three-factor structure of Basic Empathy Scale in Adults
(BES-A).
tion was r .7980, r2 .6368, p .001, and the correlation for
cognitive empathy was r .6110, r2 .3733, p .001. In order
the three-factor model (AIC 454.28). The final model is pre- to determine whether testretest reliability was better for the
sented in Figure 3. two-factor model than for the three-factor model, we compared the
In order to determine the internal consistency of the BES-A, we correlations between cognitive empathy in the two-factor model
computed Cronbachs alpha for each factor in the three-factor and the empathy factor in the three-factor model. We also com-
model. Cronbachs alpha for EMP was .69. The corresponding pared both DIS and CONT with affective empathy. These analyses
value for CONT was .72 and .82 for DIS. In order to determine revealed that the sizes of the correlations were quite similar (p
whether reliability decreased for the three-factor model, we com- .12), except in the case of the comparison between DIS and
puted Cronbachs alpha for the two-factor model (without Item 4, affective empathy (p .02).
which was also problematic in the two-factor model). These anal- The fact that we found significant correlations means that the
yses showed that Cronbachs alpha for cognitive empathy was .71, three factors appeared, to some extent, to be related to each other.
whereas the corresponding value for affective empathy was .84.
These values were thus quite similar to the alpha values obtained
in the three-factor model for EMP (.69) and for DIS (.82). The Table 2
details of the item-total correlations are presented for the structures Mean (and Standard Deviation) for the Different Subscales of
with three and two factors in Table B1. the BES-A (Three Factors)
Finally, we also analyzed the testretest reliability in a sample of
222 participants. After a mean interval of 7 weeks, the correlations Scale M (SD) (n 370)
between the test and the retest were analyzed on the different
BES-A
factors. The correlation between EMP scores was r .56, r2 DIS 12.0 (3.95)
.3118, p .001, whereas that between CONT scores was r .74, CONT 16.6 (3.48)
r2 .5488, p .001, and, finally, between DIS scores, r .70, EMP 32.1 (3.24)
r2 .4761, p .001. The participants scores in the BES-A are Note. BES-A Basic Empathy Scale in Adults; DIS definition of
presented in Table 2. We also assessed the testretest reliability for emotional disconnection; CONT definition of emotional contagion;
the two-factor model. Concerning affective empathy, the correla- EMP definition of cognitive empathy.
 EMPATHY AND THE BASIC EMPATHY SCALE IN ADULTS 685

The CONT subscale was correlated with the DIS subscale (r ing feelings and emotional disconnection suggests that mislabeling
.64, r2 .42, p .001) and the EMP subscale (r .26, r2 .06, could lead to maladaptive behavior toward other peoples emo-
p .001). DIS was also correlated with EMP (r .41, r2 .17, tions. Detailed results for the correlations are also presented in
p .001). Table 3.

Bravais-Pearson Correlations Between the BES-A Gender Differences in Empathy

With Three Factors and Other Measures
Given that previous studies have demonstrated gender differ-
As expected, no significant correlation was found between the ences in empathy (DAmbrosio et al., 2009; Joliffe & Farrington,
BES-A subscales and the total SDS score. This absence of corre- 2006), we wanted to determine whether this gender difference
lation means that responses to the BES-A were not related to a appeared on the three factors. Because there were different num-
desirability bias. Detailed results are presented in Table 3. With bers of men and women in the sample, we resampled the women
regard to the relationship between the BES-A and emotional so that the groups were of the same size. We then performed a
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

consciousness, we found positive correlations between the ESQ two-tailed t test on each factor (using the Satterthwaite approxi-
This document is copyrighted by the American Psychological Association or one of its allied publishers.

social scale and the cognitive empathy factor of the BES-A, which mation due to heteroscedasticity when necessary). These analyses
indicate that emotional consciousness about others increases in revealed a significantly higher score for women (M 17.28, SD
line with the empathic process (see Table 3). As far as the discon- 3.38) than for men (M 14.93, SD 3.34) on the emotional
nection and the contagion factors are concerned, we found a contagion factor, t(218) 5.17, p .001, d 0.70, and a
negative and a positive correlation with the expression factor of significantly lower score for women (M 11.35, SD 3.55) than
emotion (ESQ-EXPRES), respectively. Difficulties in emotion for men (M 13.88, SD 4.29) on the emotional disconnection
recognition were related to contagion. Contrary to previous results factor, t(218) 4.78, p .001, d 0.64. Concerning the cogni-
(e.g., Joliffe & Farrington, 2006), no correlation was found be- tive empathy factor, the mean score (M 32.24, SD 2.76)
tween the BES-A and the TAS-20 total scores, except in the case achieved by the women was marginally higher than that of the men
of the Cognitive Empathy subscale. Moreover, significant negative (M 31.46, SD 3.92), t(195.55) 1.71, p .09, d 0.64.
correlations were found between the EMP subscale of the BES-A These results are consistent with those reported by Joliffe and
and the DIF and DDF subscales of the TAS-20. These results make Farrington (2006) because they showed that the size of the effect
it clear that the misidentification and mislabeling of emotions was much greater for affective empathy (d 1.33) than for
might be related to difficulties in representing affective states. The cognitive empathy (d 0.63).
positive correlation between the Contagion subscale of the BES-A Importantly, although no significant difference was found for
and the DIF subscale of the TAS-20 suggests that contagion by the Cognitive Empathy subscale of the three-factor structure, such
other peoples emotion is related to difficulties in identifying ones a difference was observed when the analysis was performed on the
own emotions. Finally, the link between the difficulties in describ- two-factor structure, t(195.66) 2.34, p .05, with a higher mean
for women (36.62) than for men (35.44). This discrepancy can be
explained by the fact that one cognitive empathy item of the
Table 3 two-factor structure was interpreted as emotional disconnection in
Correlations of the BES-A and the Other Scales the three-factor structure.
BES-A DIS CONT EMP Conversely, factor analyses were not performed by gender
because there were not enough men in the sample. Indeed,
SDS .16 .01 .09 several authors suggest that the sample size should be at least
IRI-FS .28 .48 .26 200, or even more (Cattell, 1978; Comrey & Lee, 1992; Guil-
IRI-EC .66 .44 .45
IRI-PT .23 .05 .36 ford, 1954).
IRI-PD .31 .50 .10
ESQ-REC .07 .15 .09
ESQ-EXPRES .24 .18 .14 Discussion
ESQ-IEE .10 .07 .04
ESQ-SC .14 .10 .19 In the present study, we investigated the psychometric proper-
ESQ-total .16 .11 .15 ties of the French version of the BES in a sample of healthy adults
TAS-DIF .01 .19 .18 because these properties of the BES have already been examined
TAS-DDF .19 .01 .21 in various countries in populations consisting of youths and teen-
TAS-EOT .09 .03 .06 agers (Albiero et al., 2009; DAmbrosio et al., 2009; Jolliffe &
TAS-total .11 .13 .17
Farrington, 2006). Our aim, on the one hand, was to validate an
Note. BES-A Basic Empathy Scale in Adults; DIS definition of adult version of the BES and, on the other, to identify the model of
emotional disconnection; CONT definition of emotional contagion; empathy that is best able to explain the factor structure of the BES.
EMP definition of cognitive empathy; SDS Social Desirability Scale;
IRI Interpersonal Reactivity Index; FS Fantasy; EC Empathic In line with this objective, we tested three models: the unidimen-
Concern; PT Perspective Taking; PD Personal Distress; ESQ sional model, the two-factor model (i.e., affective and cognitive
Emotional State Questionnaire; REC recognition; EXPRES expres- empathy), and the three-factor model of empathy (i.e., emotional
sion; IEE internal emotional experience; SC social context; TAS contagion, emotional disconnection, and cognitive empathy). A
Toronto Alexithymia Scale; DIF Difficulties in Identifying Feelings;
DDF Difficulties in Describing Feelings; EOT Externally Oriented
confirmatory factor analysis indicated that both the two-factor and
Thinking. three-factor models could better account for the data than the

p .05. p .001. unidimensional model. These results further confirm both the
 686 CARR ET AL.

cognitive and neuroscientific data relating to emotional and socio- link between emotional disconnection and both PT and PD could
cognitive processing (Decety, 2011a; Decety & Svetlova, 2012), be due to the presence of a conflict between the understanding of
which suggest that empathy could be based on three components. other peoples emotional situations and ones own lack of emo-
Nevertheless, even though our results suggest that both the two- tional control, which would lead to emotional suppression. This
factor and the three-factor structure are appropriate, this does not emotional suppression might explain the impulsive behaviors ad-
constitute a validation of the scale because the finding is due to the opted in emergency situations (Billieux, Gay, Rochat, & Van Der
unequal proportions of men and women in the sample. Indeed, our Linden, 2010; Billieux et al., 2012), on the one hand, and mal-
sample contained twice as many women as men. adaptive behaviors, such as violence among young people (Favre
Empathy in social contexts therefore seems to be underpinned et al., 2005, 2009; Mehrabian, 1997), on the other. A positive
by (a) emotional contagion by another persons emotion, (b) emo- correlation was also found between emotional disconnection and
tional disconnection, and (c) cognitive empathy. Emotional con- DDF, as measured by the TAS-20. This might be related to
tagion is thought to be an automatic and unconscious process difficulties affecting the processes involved in empathy and the
(Dimberg & Thunberg, 2000; Papousek, Harald Freudenthaler, & regulation of emotion.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

Schulter, 2011), which involves subcortical structures known to be As far as cognitive empathy (i.e., BES-A EMP) is concerned,
This document is copyrighted by the American Psychological Association or one of its allied publishers.

involved in emotion processing (Derntl et al., 2010; Hariri et al., this factor appears to assess not only several components of the IRI
2002; Phillips et al., 2003). This is consistent with recent studies such as FS, EC, and PT but also experience with social context
that suggest that emotional contagion involves a bottom-up com- (i.e., SC) as measured by the ESQ. This finding supports the idea
ponent (Decety & Svetlova, 2012; Gyurak et al., 2012) and that it that empathy goes beyond basic skills such as recognizing emo-
constitutes the first step in the empathic functioning that is thought tions or understanding another persons point of view and could
to develop during the preverbal period (Lamm, Porges, Cacioppo, reflect specific aspects involved in cognitive empathy (Davis,
& Decety, 2008). Emotional disconnection is defined as a response 1983b; Grynberg et al., 2010). The absence of any significant
that is thought to be based on a mechanism of disconnection from correlation between cognitive empathy (EMP) and the PD measure
emotion that protects individuals from excessive emotions (i.e., of the IRI suggests that the process underpinning this component
emotions considered to be unsustainable). Emotional disconnec- provides a way of instantly distancing oneself from another per-
tion is considered to be a top-down process that has a regulatory sons emotions instead of being overwhelmed by them at the cost
function, in particular in terms of the inhibitory control exercised of personal pain, as is the case for the two previous factors (i.e.,
by the prefrontal and the cingulate cortex (Decety & Lamm, 2006; emotional contagion and emotional disconnection). Moreover, a
Singer & Lamm, 2009). Alternatively, this emotion regulation negative correlation was found between cognitive empathy and the
strategy could be considered to be a partially efficient way to react TAS-20 scales (i.e., DIF and DDF) and provides support for the
to emotional situations (Gross, 2002) when compared with com- idea that empathy impairment increases in alexithymia (Grynberg
plete emotional appraisal. Finally, cognitive empathy relates to the et al., 2010). One could argue that the links between alexithymia
ability to understand and mentalize other peoples affects and is and the Empathy scale might simply constitute an artifact due to
underpinned by the PFC (Decety, 2011a). It also includes emo- the difficulties in describing ones own feelings experienced by
tional awareness, which stems from the insular cortex (Decety & individuals with a high level of alexithymia. Indeed, one could
Svetlova, 2012). It corresponds to the concept of empathy as an hypothesize that people who find it difficult to describe their own
orientation towards others coupled with an ability to understand emotions might also have more general difficulties with regard to
others views. emotions and might therefore find it difficult to respond to an
The emotional contagion factor (i.e., BES-A CONT) was pos- empathic scale. However, this view would not be consistent with
itively related to the different subscales of the IRI (i.e., FS), EC, the studies that have shown that the ability of high- and low-
and PD but was not linked to the PT subscale. This suggests that alexithymic participants to recognize facial emotions is similar to
emotional contagion is due to a sensitivity to emotions as well as that of normal controls (Berembaum & Prince, 1994; Mayer,
to a tendency toward a lack of control (i.e., PD), which would be DiPaolo, & Salovery, 1990; McDonald & Prkachin, 1990; Mon-
consistent with a lack of PT. Emotional contagion was also posi- tebarocci, Surcinelli, Rossi, & Baldaro, 2011). Indeed, there is no
tively associated with the expression of emotions (i.e., EXPRES) reason to think that more highly alexithymic participants would be
and negatively linked to emotional recognition (i.e., REC) as less capable of understanding words related to emotions.
measured by the ESQ scale. These results might mean that emo- Taken together, these results suggest that the better delineation
tional contagion is also associated with a deficit in the ability to of factors offered by this new three-factor model of the BES-A
recognize other peoples emotions. This result is consistent with could contribute to our understanding of the mental impairment of
the correlation found between emotional contagion and DIF, as emotional components and might provide a better account of how
measured by the TAS-20 scale. More specifically, our results make empathy is deployed in social and emotional contexts. The rela-
it clear that the misidentification and mislabeling of emotions may tions with other tools that assess emotional functioning provide
be related to difficulties in representing affective states and that support for the external validity of the BES-A score. There is a
contagion by other peoples emotions is related to difficulties in growing body of evidence showing that empathy is a key element
identifying ones own emotions. in emotional and interpersonal functioning and that it should be
Emotional disconnection (i.e., BES-A DIS) was negatively cor- considered to be a multidimensional concept (Decety & Jackson,
related with all the subscales of the IRI (i.e., FS, EC, PT, PD) as 2004; Gerdes et al., 2011; Grynberg et al., 2010). Initially consid-
well as with EXPRESS of the ESQ scale. This would seem to ered as based on two factors, namely, the cognitive and affective
indicate that disconnection in empathic functioning is the inverse dimensions of empathy, the BES (Jolliffe & Farrington, 2006) can
of well-adapted empathic functioning. Interestingly, the negative now be thought of as a tool that is based on three factors. Further-
 EMPATHY AND THE BASIC EMPATHY SCALE IN ADULTS 687

more, our results are not restricted to adult populations. Indeed, we Bagby, R. M., Parker, J. D. A., & Taylor, G. J. (1994). The twenty-item
extended our analysis to include empathy as initially modeled in a Toronto Alexithymia Scale: I. Item selection and cross-validation of the
teenage population in a French sample (in accordance with factor structure. Journal of Psychosomatic Research, 38, 2332. doi:
DAmbrosio et al., 2009) and found that the three-factor model 10.1016/0022-3999(94)90005-1
again provides a better account of the results. The large body of Bagby, R. M., Taylor, G. J., & Ryan, D. (1986). Toronto Alexithymia
Scale: Relationship with personality and psychopathology measures.
evidence also confirms that empathy can be conceived of within a
Psychotherapy and Psychosomatics, 45, 207215. doi:10.1159/
three-factor model. 000287950
To further delineate this conception of empathy, it would appear Batson, C. D., Early, S., & Salvarani, G. (1997). Perspective taking:
necessary to examine results obtained using other measures of Imagining how another feels versus imagining how you would feel.
emotional functioning (e.g., anxious or depressed states) or on the Personality and Social Psychology Bulletin, 23, 751758. doi:10.1177/
basis of personality dimensions, as in the case of the IRI (Moora- 0146167297237008
dian, Davis, & Matzler, 2011). Another limitation of this study Batson, C. D., Fultz, J., & Schoenrade, P. A. (1987). Distress and empathy:
relates to the fact that the analyses were based on a sample of Two qualitatively distinct vicarious emotions with different motivational
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

nonclinical participants. As explained above, the three components consequences. Journal of Personality, 55, 19 39. doi:10.1111/j.1467-
This document is copyrighted by the American Psychological Association or one of its allied publishers.

that have been described in order to construct the new factors of 6494.1987.tb00426.x
the BES are drawn primarily from data about social cognition Berembaum, H., & Prince, J. (1994). (1994). Alexithymia and the inter-
pretation of emotion-relevant information. Cognition & Emotion, 5,
obtained within developmental and psychopathological frame-
19 30.
works. In addition, the BES-A, defined on the basis of the three-
Besnier, N., Cass-Perrot, C., Jouve, E., Nguyen, N., Lanon, C., Falissard,
factor model, could help to better specify the impairments of B., & Blin, O. (2010). Effects of paroxetine on emotional functioning
empathic functioning observed in several psychopathologies such and treatment awareness: A 4-week randomized placebo-controlled
as anxiety and mood disorders, autism, or schizophrenia. Finally, study in healthy clinicians. Psychopharmacology, 207, 619 629. doi:
as empathy is known to involve several processes that are related 10.1007/s00213-009-1691-4
to different cerebral areas (Shamay-Tsoory, 2011), it seems appro- Billieux, J., Gay, P., Rochat, L., & Van Der Linden, M. (2010). The role
priate to explore the neural correlates of the BES-A in order to of urgency and its underlying psychological mechanisms in problematic
determine whether the three components of this model correspond behaviours. Behaviour Research and Therapy, 48, 10851096. doi:
to the neural activations described in previous studies. 10.1016/j.brat.2010.07.008
In conclusion, if the BES-A remains an appropriate tool for Billieux, J., Rochat, L., Ceschi, G., Carre, A., Offerlin-Meyer, I., Defeldre,
assessing empathy according to a two-factor structure, it is also A-C., . . . Van der Linden, M. (2012). Validation of a short French
version of the UPPS-P Impulsive Behavior Scale. Comprehensive Psy-
possible to take into account the recent definition of the processes
chiatry, 53, 609 615. doi:10.1016/j.comppsych.2011.09.001
involved in empathy and to implement them in the BES-A, such as
Bryant, B. K. (1982). An index of empathy for children and adolescents.
we did in this study. In other words, the BES-A can be used in Child Development, 53, 413 425. doi:10.2307/1128984
French either with a two-factor or a three-factor structure depend- Cass-Perrot, C., Fakra, E., Jouve, E., & Blin, O. (2007). Conceptualisation
ing on the needs of the study. It now provides a brief assessment and validation of the Emotional State Questionnaire (ESQ): Evaluation
of empathic functioning in teenagers and adults. This conception of an emotional profile. LEncphale, 33, 169 178. doi:10.1016/S0013-
of empathy based on three factors is consistent with an extended 7006(07)91547-X
and more integrated vision of empathy. Earlier two-factor struc- Cattell, R. B. (1978). The scientific use of factor analysis in behavioral and
tures (i.e., cognitive and affective processes) were limited in their life sciences. New York, NY: Plenum.
ability to describe the processes involved in empathy. The three- Clark, T. F., Winkielman, P., & McIntosh, D. N. (2008). Autism and the
factor structure (i.e., emotional contagion, emotional disconnec- extraction of emotion from briefly presented facial expressions: Stum-
tion, and cognitive empathy) constitutes a more recent perspective bling at the first step of empathy: Empathy. Emotion, 8, 803 809.
doi:10.1037/a0014124
of the functional and dysfunctional components of empathic pro-
Comrey, A. L., & Lee, H. B. (1992). A first course in factor analysis.
cesses and responses in both adults and teenagers. Further research
Hillsdale, NJ: Erlbaum.
will be needed in order to gain a better understanding of the Crowne, D. P., & Marlowe, D. (1960). A new scale of social desirability
usefulness of the BES-A in different contexts. independent of psychopathology. Journal of Consulting Psychology, 24,
349 354. doi:10.1037/h0047358
DAmbrosio, F., Olivier, M., Didon, D., & Besche, C. (2009). The basic
References empathy scale: A French validation of a measure of empathy in youth.
Albiero, P., Matricardi, G., Speltri, D., & Toso, D. (2009). The assessment Personality and Individual Differences, 46, 160 165. doi:10.1016/j.paid
of empathy in adolescence: A contribution to the Italian validation of the .2008.09.020
Basic Empathy Scale. Journal of Adolescence, 32, 393 408. doi: Davis, M. H. (1980). A multidimensional approach to individual differ-
10.1016/j.adolescence.2008.01.001 ences in empathy. JSAS Catalog of Selected Documents in Psychology,
Albiero, P., Matricardi, G., & Toso, D. (2010). La Basic Empathy Scale, 10, 85. Retrieved from http://www.uv.es/~friasnav/Davis_1980.pdf
uno strumento per la misura della responsivita` empatica negli adoles- Davis, M. H. (1983a). The effects of dispositional empathy on emotional
centi: Un contributo alla validazione Italiana [The Basic Empathy Scale, reactions and helping: A multidimensional approach. Journal of Per-
a measure of empathy in adolescence: A further contribution to the sonality, 51, 167184. doi:10.1111/j.1467-6494.1983.tb00860.x
Italian validation]. Psicologia Clinica dello Sviluppo, 14, 205218. Davis, M. H. (1983b). Measuring individual differences in empathy: Evi-
doi:10.1449/32011 dence for a multidimensional approach. Journal of Personality and
Andrews, P., & Meyer, R. G. (2003). MarloweCrowne Social Desirability Social Psychology, 44, 113126. doi:10.1037/0022-3514.44.1.113
Scale and short form C: Forensic norms. Journal of Clinical Psychology, Decety, J. (2010). The neurodevelopment of empathy in humans. Devel-
59, 483 492. doi:10.1002/jclp.10136 opmental Neuroscience, 32, 257267. doi:10.1159/000317771
 688 CARR ET AL.

Decety, J. (2011a). Dissecting the neural mechanisms mediating empathy. Haker, H., & Rssler, W. (2009). Empathy in schizophrenia: Impaired
Emotion Review, 3, 92108. doi:10.1177/1754073910374662 resonance. European Archives of Psychiatry & Clinical Neuroscience,
Decety, J. (2011b). The neuroevolution of empathy. Annals of the New 259, 352361. doi:10.1007/s00406-009-0007-3
York Academy of Sciences, 1231, 35 45. doi:10.1111/j.1749-6632.2011 Hariri, A. R., Tessitore, A., Mattay, V. S., Fera, F., & Weinberger, D. R.
.06027.x (2002). The amygdala response to emotional stimuli: A comparison of
Decety, J., & Jackson, P. L. (2004). The functional architecture of human faces and scenes. NeuroImage, 17, 317323. doi:10.1006/nimg.2002
empathy. Behavioral and Cognitive Neuroscience Reviews, 3, 71100. .1179
doi:10.1177/1534582304267187 Hinnant, J. B., & OBrien, M. (2007). Cognitive and emotional control and
Decety, J., & Lamm, C. (2006). Human empathy through the lens of social perspective taking and their relations to empathy in 5-year-old children.
neuroscience. The Scientific World Journal, 6, 1146 1163. doi:10.1100/ The Journal of Genetic Psychology, 168, 301322. doi:10.3200/GNTP
tsw.2006.221 .168.3.301-322
Decety, J., & Michalska, K. J. (2010). Neurodevelopmental changes in the Hoffman, M. L. (1977). Personality and social development. Annual Re-
circuits underlying empathy and sympathy from childhood to adulthood. view of Psychology, 28, 295321. doi:10.1146/annurev.ps.28.020177
Developmental Science, 13, 886 899. doi:10.1111/j.1467-7687.2009 .001455
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

.00940.x Hoffman, M. L. (2000). Empathy and moral development: Implications for

This document is copyrighted by the American Psychological Association or one of its allied publishers.

Decety, J., & Svetlova, M. (2012). Putting together phylogenetic and caring and justice. New York, NY: Cambridge University Press. doi:
ontogenetic perspectives on empathy. Developmental Cognitive Neuro- 10.1017/CBO9780511805851
science, 2, 124. doi:10.1016/j.dcn.2011.05.003 Hogan, R. (1969). Development of an empathy scale. Journal of Consult-
Derntl, B., Finkelmeyer, A., Eickhoff, S., Kellermann, T., Falkenberg, ing and Clinical Psychology, 33, 307316. doi:10.1037/h0027580
D. I., Schneider, F., & Habel, U. (2010). Multidimensional assessment of Iacoboni, M., & Dapretto, M. (2006). The mirror neuron system and the
empathic abilities: Neural correlates and gender differences. Psychoneu- consequences of its dysfunction. Nature Reviews Neuroscience, 7, 942
roendocrinology, 35, 67 82. doi:10.1016/j.psyneuen.2009.10.006 951. doi:10.1038/nrn2024
Deutsch, F., & Madle, R. A. (1975). Empathy: Historic and current Jolliffe, D., & Farrington, D. P. (2004). Empathy and offending: A sys-
conceptualizations, measurement, and a cognitive theoretical perspec- tematic review and meta-analysis. Aggression and Violent Behavior, 9,
tive. Human Development, 18, 267287. doi:10.1159/000271488 441 476. doi:10.1016/j.avb.2003.03.001
Dimberg, U., & Thunberg, M. (2000). Unconscious facial reactions to
Jolliffe, D., & Farrington, D. P. (2006). Development and validation of the
emotional facial expressions. Psychological Science, 11, 86 89. doi:
Basic Empathy Scale. Journal of Adolescence, 29, 589 611. doi:
10.1111/1467-9280.00221
10.1016/j.adolescence.2005.08.010
Eisenberg, N. (2010). Empathy-related responding: Links with self-
Jreskog, K. G. (1990). New developments in LISREL: Analysis of ordinal
regulation, moral judgment, and moral behavior. In M. Mikulincer &
variables using polychoric correlations and weighted least squares.
P. R. Shaver (Eds.), Prosocial motives, emotions, and behavior: The
Quality and Quantity, 24, 387 404. doi:10.1007/BF00152012
better angels of our nature (pp. 129 148). Washington, DC: American
Jreskog, K. G., & Srbom, D. (2004). LISREL (Version 8.7). Lincoln-
Psychological Association. doi:10.1037/12061-007
wood, IL: Scientific Software International.
Favre, D., Joly, J., Reynaud, C., & Salvador, L. L. (2005). Empathy,
Lamm, C., Batson, D. C., & Decety, J. (2007). The neural substrate of
emotional contagion and emotion disconnection. Enfance, 57, 363382.
human empathy: Effects of perspective-taking and cognitive appraisal.
doi:10.3917/enf.574.0363
Journal of Cognitive Neuroscience, 19, 4258. doi:10.1162/jocn.2007
Favre, D., Joly, J., Reynaud, C., & Salvador, L. L. (2009). Empathy,
.19.1.42
emotional contagion and splitting with emotions: Validation of a test to
Lamm, C., Porges, E. C., Cacioppo, J. T., & Decety, J. (2008). Perspective
locate and help pupils at risk. European Review of Applied Psychology/
Revue Europenne de Psychologie Applique, 59, 211227. doi: taking is associated with specific facial responses during empathy for
10.1016/j.erap.2009.03.004 pain. Brain Research, 1227, 153161. doi:10.1016/j.brainres.2008.06
Gerdes, K. E., Lietz, C. A., & Segal, E. A. (2011). Measuring empathy in .066
the 21st century: Development of an empathy index rooted in social Lawrence, E. J., Shaw, P., Baker, D., Baron-Cohen, S., & David, A. S.
cognitive neuroscience and social justice. Social Work Research, 35, (2004). Measuring empathy: Reliability and validity of the empathy
8393. doi:10.1093/swr/35.2.83 quotient. Psychological Medicine, 34, 911920. doi:10.1017/
Gross, J. J. (2002). Emotion regulation: Affective, cognitive, and social S0033291703001624
consequences. Psychophysiology, 39, 281291. doi:10.1017 Leite, W. L., & Beretvas, S. N. (2005). Validation of scores on the
.S0048577201393198 Marlowe-Crowne Social Desirability Scale and the Balanced Inventory
Grynberg, D., Luminet, O., Corneille, O., Grzes, J., & Berthoz, S. (2010). of Desirable Responding (English). Educational and Psychological
Alexithymia in the interpersonal domain: A general deficit of empathy? Measurement, 65, 140 154. doi:10.1177/0013164404267285
Personality and Individual Differences, 49, 845 850. doi:10.1016/j.paid Li, C-F., Lv, R., Liu, J., & Zhong, J. (2011). The adaptation of the Basic
.2010.07.013 Empathy Scale among Chinese adolescents. Chinese Journal of Clinical
Guilbaud, O., Loas, M., Corcos, M., Speranza, M., Stephan, P., Perez-Diaz, Psychology, 19, 163166.
F., . . . Jeammet, P. (2002). Alexithymia in addictive behaviors and in Lipps, T. (1979). Empathy, inner imitation and sense-feelings. In M. Rader
healthy subjects: Results of a study in French-speaking subjects. Annales (Ed.), A modern book of esthetics (5th ed., pp. 374 382). New York,
Mdico-psychologiques, revue psychiatrique, 160, 77 85. doi:10.1016/ NY: Harcourt College.
S0003-4487(01)00134-2 Loas, G., Otmani, O., Verrier, A., Fremaux, D., & Marchand, M. P. (1996).
Guilford, J. P. (1998). Psychometric methods. New York, NY: McGraw Factor analysis of the French version of the 20-item Toronto Alexithy-
Hill. mia Scale (TAS-20). Psychopathology, 29, 139 144. doi:10.1159/
Gyurak, A., Goodkind, M. S., Kramer, J. H., Miller, B. L., & Levenson, 000284983
R. W. (2012). Executive functions and the down-regulation and up- Marshall, L. E., & Marshall, W. L. (2011). Empathy and antisocial behav-
regulation of emotion. Cognition & Emotion, 26, 103118. doi:10.1080/ iour. The Journal of Forensic Psychiatry & Psychology, 22, 742759.
02699931.2011.557291 doi:10.1080/14789949.2011.617544
 EMPATHY AND THE BASIC EMPATHY SCALE IN ADULTS 689

Marshall, W. L., Hudson, S. M., Jones, R., & Fernandez, Y. M. (1995). perception. Biological Psychiatry, 54, 504. doi:10.1016/S0006-
Empathy in sex offenders. Clinical Psychology Review, 15, 99 113. 3223(03)00168-9
doi:10.1016/0272-7358(95)00002-7 Rogers, C. (1951). Client-centred therapy. London, England: Constable.
Mayer, J. D., DiPaolo, M., & Salovey, P. (1990). Perceiving affective Schulte-Rther, M., Greimel, E., Markowitsch, H. J., Kamp-Becker, I.,
content in ambiguous visual stimuli: A component of emotional intelli- Remschmidt, H., Fink, G. R., & Piefke, M. (2011). Dysfunctions in brain
gence. Journal of Personality Assessment, 54, 772781. networks supporting empathy: An fMRI study in adults with autism
McDonald, P. W., & Prkachin, K. W. (1990). The expression and percep- spectrum disorders. Social Neuroscience, 6, 121. doi:10.1080/
tion of facial emotion in alexithymia: A pilot study. Psychosomatic 17470911003708032
Medicine, 52, 199 210. Shamay-Tsoory, S. G. (2011). The neural bases for empathy. The Neuro-
Mehrabian, A. (1997). Relations among personality scales of aggression, scientist: A Review Journal Bringing Neurobiology, Neurology and
violence and empathy: Validational evidence bearing on the risk of Psychiatry, 17, 18 24. doi:10.1177/1073858410379268
eruptive violence scale. Aggressive Behavior, 23, 433 445. doi:10.1002/
Singer, T., & Lamm, C. (2009). The social neuroscience of empathy.
(SICI)1098-2337(1997)23:6433::AID-AB33.0.CO;2-H
Annals of the New York Academy of Sciences, 1156, 8196. doi:10.1111/
Mehrabian, A., & Epstein, N. (1972). A measure of emotional empathy.
j.1749-6632.2009.04418.x
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

Journal of Personality, 40, 525543. doi:10.1111/j.1467-6494.1972

Taylor, G. J., Bagby, R. M., & Parker, J. D. A. (1997). Disorders of affect
.tb00078.x
This document is copyrighted by the American Psychological Association or one of its allied publishers.

regulation: Alexithymia in medical and psychiatric illness. New York,

Montebarocci, O., Surcinelli, P., Rossi, N., & Baldaro, B. (2011). Alex-
ithymia, verbal ability and emotion recognition. Psychiatric Quarterly, NY: Cambridge University Press. doi:10.1017/CBO9780511526831
82, 245252. Thoma, P., Zalewski, I., von Reventlow, H. G., Norra, C., Juckel, G., &
Mooradian, T. A., Davis, M., & Matzler, K. (2011). Dispositional empathy Daum, I. (2011). Cognitive and affective empathy in depression linked
and the hierarchical structure of personality. The American Journal of to executive control. Psychiatry Research, 189, 373378. doi:10.1016/j
Psychology, 124, 99 109. doi:10.5406/amerjpsyc.124.1.0099 .psychres.2011.07.030
Papousek, I., Harald Freudenthaler, H., & Schulter, G. (2011). Typical perfor- Titchener, E. (1909). Elementary psychology of the thought processes.
mance measures of emotion regulation and emotion perception and frontal EEG New York, NY: Macmillan. doi:10.1037/10877-000
asymmetry in an emotional contagion paradigm. Personality and Individual Zelazo, P., Carlson, S., & Kesek, A. (2008). The development of executive
Differences, 51, 1018 1022. doi:10.1016/j.paid.2011.08.013 function in childhood. In C. A. Nelson & M. Luciana (Eds.), Handbook
Phillips, M. L., Drevets, W. C., Rauch, S. L., & Lane, R. (2003). Neuro- of developmental cognitive neuroscience (pp. 553574). Cambridge,
biology of emotion perception I: The neural basis of normal emotion MA: MIT Press.

(Appendices follow)
 690 CARR ET AL.

Appendix A
Items of the Basic Empathy Scale (20 items)

1. My friends emotions dont affect me much. 11. I often become sad when watching sad things on TV or in
films.
2. After being with a friend who is sad about something, I
usually feel sad. 12. I can often understand how people are feeling even before
they tell me.
3. I can understand my friends happiness when she/he
does well at something. 13. Seeing a person who has been angered has no effect on my
feelings.
4. I get frightened when I watch characters in a good scary
movie. 14. I can usually work out when people are cheerful.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
This document is copyrighted by the American Psychological Association or one of its allied publishers.

5. I get caught up in other peoples feelings easily. 15. I tend to feel scared when I am with friends who are afraid.
6. I find it hard to know when my friends are frightened. 16. I can usually realize quickly when a friend is angry.
7. I dont become sad when I see other people crying. 17. I often get swept up in my friends feelings.
8. Other peoples feeling dont bother me at all. 18. My friends unhappiness doesnt make me feel anything.
9. When someone is feeling down I can usually understand
how they feel. 19. I am not usually aware of my friends feelings.

10. I can usually work out when my friends are scared. 20. I have trouble figuring out when my friends are happy.

(Appendices continue)
 EMPATHY AND THE BASIC EMPATHY SCALE IN ADULTS 691

Appendix B

Table B1
Details of Cronbachs Alpha Coefficients (BES--Three-Factor Model)

Items CONT EMP DIS

1 .61
2 .54
3 .29
4 deleted
5 .53
6 .40
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

7 .54
This document is copyrighted by the American Psychological Association or one of its allied publishers.

8 .70
9 .34
10 .48
11 .42
12 .40
13 .47
14 .38
15 .40
16 .37
17 .51
18 .75
19 .47
20 .44

Details of Cronbachs alpha coefficients (BESTwo-factor model)

Item Affective Cognitive

1 .56
2 .56
3 .33
4 deleted
5 .62
6 .39
7 .56
8 .64
9 .35
10 .46
11 .45
12 .40
13 .49
14 .37
15 .39
16 .37
17 .54
18 .68
19 .37
20 .47
Note. BES Basic Empathy Scale; CONT definition of emotional contagion; EMP definition of cognitive empathy;
DIS definition of emotional disconnection.

Received May 28, 2012

Revision received January 29, 2013
Accepted February 5, 2013