Single Study on Collaborative and Proactive Solutions

O R I G I N A L A R T I C L E

Emotional Abilities in Children with Oppositional Defiant Disorder (ODD): Impairments in Perspective-Taking and Understanding Mixed Emotions are Associated with High Callous–Unemotional Traits

Richard O’Kearney1 • Karen Salmon2 • Maria Liwag1 • Clare-Ann Fortune2 •

Amy Dawel 1,3

Published online: 21 April 2016

� Springer Science+Business Media New York 2016

Abstract Most studies of emotion abilities in disruptive

children focus on emotion expression recognition. This

study compared 74 children aged 4–8 years with ODD to

45 comparison children (33 healthy; 12 with an anxiety

disorder) on behaviourally assessed measures of emotion

perception, emotion perspective-taking, knowledge of

emotions causes and understanding ambivalent emotions

and on parent-reported cognitive and affective empathy.

Adjusting for child’s sex, age and expressive language

ODD children showed a paucity in attributing causes to

emotions but no other deficits relative to the comparison

groups. ODD boys with high levels of callous–unemotional

traits (CU) (n = 22) showed deficits relative to low CU

ODD boys (n = 25) in emotion perspective-taking and in

understanding ambivalent emotions. Low CU ODD boys

did not differ from the healthy typically developing boys

(n = 12). Impairments in emotion perceptive-taking and

understanding mixed emotions in ODD boys are associated

with the presence of a high level of CU.

Keywords Oppositional defiant disorder � Emotional competencies � Callous unemotional traits � Ambivalent emotions

Introduction

Children with oppositional defiant disorder (ODD) and

conduct disorder (CD) display symptomatic difficulties in

emotional functioning. These include ineffective self-reg-

ulation of negative emotions, a restricted range of emo-

tional expression and a failure to appreciate and to respond

appropriately to the emotions of others [1]. Some theories

[2–4] propose that these symptoms reflect deficits in

specific psychological competencies which underpin chil-

dren’s capacity to respond to emotional experiences, and to

understand that others also have private emotions. Prob-

lems in these competencies are seen as contributing

strongly to the development of the disturbed social inter-

actions including the anti-social and defiant behaviours

which are characteristic of children with disruptive beha-

viour disorders [2, 5].

There has been a recent emphasis on identifying deficits

in emotional abilities characteristic of the subgroup of

disruptive children with callous and unemotional (CU)

traits. These children show cruelty, a lack of concern for

others, an inability to feel guilt and impaired empathy and

in the context of conduct disorder are delineated as having

limited pro-social emotions [1]. Recent models of CU [6,

7] suggest that deficiencies in basic emotional abilities

profoundly impact on the development of empathy and

social emotions in this group. Blair et al. [6] proposes that

poor recognition of fear and distress cues is fundamental to

high CU disruptive children’s insensitivity to others’ feel-

ings and impoverished empathic concern. For Dadds et al.

[7, 8] deficits in high CU children’s distress recognition is

due to impaired attention during the perception of impor-

tant emotional stimuli such as other people’s eyes and

facial features. Impaired attention not only impacts emo-

tion recognition but also inhibits children’s capacity to

& Richard O’Kearney [email protected]

1 Research School of Psychology, Australian National

University, Canberra, ACT 0200, Australia

2 Victoria University of Wellington, Wellington, New Zealand

3 ARC Centre of Excellence in Cognition and its Disorders,

University of Western of Australia, Perth, Australia

123

Child Psychiatry Hum Dev (2017) 48:346–357

DOI 10.1007/s10578-016-0645-4

attribute mental states including emotions and intention to

the other person [9] and also, more broadly, disrupts the

quality of interactions between children and their parents,

caregivers and peers.

The ability to accurately recognise emotion expressions

contributes to the development of a number of more

advanced emotional competencies which underlie the

ability to identify and predict emotional reactions and

experiences and to appreciate that others also have private

emotional experiences. These abilities typically develop in

an expected pattern during early childhood and into the first

years of school continuing to advance in their complexity

and intricacy [10]. According to most developmental

models children move from understanding emotion ‘‘as

nothing more than certain instrumental (e.g. approach-

avoidance) or expressive (e.g. cry-smile) actions’’ [11,

p. 291] to situational understandings where they make

sense of emotions by knowing that certain situations or

circumstances lead regularly to particular instrumental or

expressive actions. Subsequently, children develop sub-

jective and experiential understandings where they have a

sense of the subjectivity of emotional experience and

reflective knowledge about the role of internal states such

as desire and belief in emotions [10, 12].

The competencies which reflect these stages include

emotion labelling (recognising and naming basic emotions

like anger, sadness, fear), emotion perspective-taking

(knowing how another person might feel in specific situa-

tions), understanding of the causes of emotions (knowledge

of the behavioural, situational and internal-state causes of

emotions) and appreciating more complex, mixed and

ambivalent emotions (knowing that different and con-

trasting emotions can co-occur) [10, 13–15]. As children

develop, these competencies support their experience and

reflective understanding of their own and others’ emotions

in increasingly complex and dynamic social contexts and

guide them in how to respond appropriately in these con-

texts. Children typically make considerable gains in emo-

tion perspective taking and understanding the causes of

emotions during the preschool years and this continues

once they enter school [10, 13]. The development of a

reflective understanding of mixed or ambivalent emotions

continues during the primary school years so that by age 8

most children at least understand that positive and negative

emotions can be elicited in the same situation [16].

The more advanced emotion competencies are consid-

ered to have critical roles in the development of moral

reasoning and pro and antisocial decision making and

behaviour. They are necessary for properly integrating

expressive and situational information when understanding

other’s emotional perspective because both types of

information are necessary for appreciating that another

person can display one emotion but really feel a different

way, that one person can feel different emotions to another

person in the same situation and also that the same context

can elicit competing and conflicting emotional responses

within the same person [17–19]. Reflective understandings

of emotions are also prerequisites for knowledge of self-

focused, social emotions [18] such as embarrassment,

pride, guilt and shame because these emotions are initial

evoked by congruence or incongruence between the child’s

emotional response and the emotional response of others to

the child’s behaviour. By age 8 most children are begin-

ning to experience and understand the social meaning of

these later emotions as arising from violation or conformity

with social standards and these emotions increasingly

become the motivational drivers of children’s developing

prosocial responses [18, 20].

These consideration suggests that ODD children who

display low pro-social emotions and behaviour (those with

high CU traits) will show deficits in advanced emotional

competencies. Beyond the basic recognition deficits,

however, we know relatively little about the performance

of children with disruptive disorders, including those with

high CU, across the range of emotional competencies

which develop during childhood. While there has been

considerable investigation of the association between

emotion competencies and aggressive behaviour [21], rel-

atively few studies provide evidence about the extent and

nature of emotional deficits in children diagnosed with a

disruptive disorder (ODD or CD) [see 21, 22]. Overall,

there is evidence for ODD related impairments in chil-

dren’s emotion labelling [23, 24] and for biases in inter-

preting ambiguous emotions as indicating hostile intent in

others [3, 25]. Consistent with the impaired emotion

expression recognition models [4, 5], the strongest and

most consistent evidence is for an impairment in recogni-

tion of emotion expressions in faces, voices and body

posture in ODD and CD children and youth with high CU

[4, 8, 22] which may be strongest for fear [4, 22]. There is

also some evidence that high CU ODD school-aged chil-

dren have low levels of both cognitive and affective

empathy as observed by parents and teachers [7] suggesting

that emotion deficits may have a continuing impact on how

well high CU disruptive children show understanding of

others’ emotions.

It is unclear whether ODD children also have beha-

vioural deficits in emotion labelling, understanding emo-

tion causes, perspective taking and understanding mixed

emotions or whether any deficits in these abilities are

associated with presence of high CU. As children engage in

more diverse interpersonal and emotional situations, such

as in school, they are provided with opportunities to learn

contextual information to emotional understanding and

perspective-taking. There is evidence that as children

develop they increasingly take this information in account

Child Psychiatry Hum Dev (2017) 48:346–357 347

123

in their empathic judgements and pro-social responses,

successfully relying on it when emotion expressive infor-

mation is not available [26, 27]. This contextual informa-

tion includes the variety of terms and verbal expressions

for emotions, knowledge of the various causes of emotion

including internal-state causes such as desires, goals and

beliefs and how these might co-occur and compete, as well

as information about other’s perspective on the emotion

situation. While this route to the development of empathy

and advanced emotion understanding may offset the impact

of early emotion recognition deficits, it’s not clear how

well ODD children, particularly those with high CU traits,

learn and use contextual cues to other’s emotions. Evidence

about how disruptive children perform across the range of

emotional competencies could strengthen inferences about

the ongoing consequences of CU related emotion recog-

nition deficits and help identify important targets for

enhancing pro-social responding in disruptive children with

and without high CU.

The Current Study

These considerations suggest that while ODD is related to

biases in interpreting internal causes for others’ emotions,

deficits in emotion perspective taking, knowledge of causes

of emotion and understanding mixed emotions in children

with disruptive disorders may be associated with low pro-

social emotions (high CU traits). These later deficits will be

evident in comparisons between high CU ODD children

and low CU ODD children and also when compared to

other low CU children without ODD. It is unclear whether

there are other child characteristics common across child-

hood disorders which are related to these emotion com-

petencies deficits and which may influence differences

between high CU ODD children and other groups. The

current study considers the presence of persistent prob-

lematic difficulties in the regulation of emotions and child

sex as possible factors which could account for or moderate

these differences. Difficulties in regulation of emotions

may arise from deficits in emotional abilities and represent

common disordered processes which characterise child-

hood internalising problems such as anxiety and depression

as well as disruptive disorders. Boys are more strongly

represented than girls in ODD groups and also in the high

CU ODD subgroup. Boys and girls also show different

progress in the development of emotional competencies

across early childhood [28] suggesting that being male may

account for some of the emotional deficits observed in

ODD and in high CU disruptive children. The current study

considers these two factors alongside ODD status and high

and low CU traits in a between group design.

The current study examines the performance of young

children in the first years of school across emotional

competencies including (1) emotion perception, (2) emo-

tion perspective-taking, (3) understanding the causes of

emotions, and (4) understanding ambivalent or mixed

emotions as well as their parents’ ratings of their levels of

(5) cognitive and affective empathy. By age 8 most children

have progressed in each of these competencies and are

beginning to experience and understand the social meaning

of emotions such guilt and shame [18, 29]. This study

firstly compares children 4–8 years who have a diagnosis

of ODD to their non-ODD peers, including a subgroup with

internalising problems, across these four emotion compe-

tencies using behavioural measures, and on cognitive and

affective empathy using parent’s report. The study then

asks whether relative deficits in emotional competencies

are marked or specific for ODD children with high levels of

CU. This subgroup analysis compares high CU ODD

children with low CU ODD children and children without

ODD across the range of emotional competencies. The

effect of child sex and its interaction with ODD and CU

status are also examined.

Methods

Participants

Participants were 124 children aged from 4 to 8 years and

their parents recruited at two sites. Parents were referred

for assessment of their child’s difficult-to-manage beha-

viours, or recruited through schools as part of a study about

how parents can enhance their children’s ability to manage

their emotions. Ethics approval was obtained from the

relevant human ethics committees which give approval in

accordance with ethical standards comparable to the 1964

Helsinki declaration and its later amendments. Full, written

consent from parents was obtained.

Procedures

Screening

Parents were interviewed via telephone for suitability in

terms of child’s age, nature of any concerns, English spo-

ken at home and to exclude possible autism spectrum

disorder or developmental disability. An in-person assess-

ment session was scheduled for suitable children. A Parent

Assessment Kit with parent-report instruments, Informa-

tion letter and Informed Consent Form was posted to the

parents to complete and bring to assessment.

348 Child Psychiatry Hum Dev (2017) 48:346–357

123

Assessment Session

Separate interviews with the parent and with the child took

place in adjoining rooms. A semi-structured diagnostic

clinical interview, brief history-taking and developmental

assessment was administered to the parent regarding their

child. The child’s emotional competencies were assessed

using emotion competency measures. The assessment ses-

sion lasted an hour and a half to two hours and was

attended by mothers (n = 116), fathers (n = 2) or both

(n = 6).

Measures

Diagnosis

The clinician-administered Diagnostic Interview Schedule

for Children, Adolescents and Parents [30] was used to

establish the presence of ODD and any other Axis 1 DSM-

IV diagnosis. This semi-structured interview yields a multi-

axis DSM-IV diagnosis and clinical severity index and has

been shown to have high between clinician reliability for a

diagnosis of ODD [31].

Callous–Unemotional Traits

The Parent Report of the Inventory of Callous Unemotional

traits (ICU0) [32] was used to assess the child’s level of

CU traits. ICU items are rated on a 4 point Likert scale

from 0 (Not at all true) to 3 (Definitely true). This 24 item

measure assesses total CU and has been shown to have

good construct validity [33]. ICU has acceptable reliability

across different age ranges and genders [34] and was found

to good internal consistency in the present study (Cron-

bach’s alpha = 0.76). The total ICU score was used to

classify children as high and low CU (HCU, LCU). Con-

sistent with prevalence estimates of high CU in previous

work [35] and the cut-offs used to classify disruptive

children as high CU [36], ODD children were considered to

have high CU if they scored at or above the 67 % ile on the

ICU.

Child Emotional Competencies

Emotion competencies were assessed using the Denham

Affective Knowledge Tests [37, 38] which are widely used

and reliable (Cronbach alpha = 0.66 to 0.86) behavioural

measures of Emotion perception, Emotion perspective-

taking, Emotion causes, and Ambivalent or mixed emo-

tions. The Affective Knowledge tests are administered as a

semi-naturalistic child interview embedded within play

with the experimenter where the child views and talks

about dolls with happy, sad, angry, and fearful expressions,

or takes part in enacted emotion scenarios or stories. Two

raters blind to the child’s diagnosis judged the child’s

response for emotion causes and mixed emotions with

75 % of the interviews co-rated for reliability.

Emotion Perception

This test taps into children’s ability to recognize and label

facial expressions of emotion. Children are shown dolls

with detachable faces that depict happy, sad, angry, and

afraid expressions in a random order. Receptive recogni-

tion: Children are asked to point to the correct face as the

experimenter named each expression. Expressive recogni-

tion: Children are first asked to name each expression as

the experimenter points to the faces in turn. For each

labelling task, total scores ranged from 0 to 8 (2 points for

correct labelling/recognition of emotion; 1 if valence is

correct; 0 for totally incorrect). Interrater agreement for

both receptive and expressive recognition was 100 %.

Emotion Perspective-Taking

These tasks tests children’s knowledge of other people’s

feelings in two types of situations. In the Stereotypical task,

the emotions evoked are non-equivocal, such as happiness

at being given an ice cream cone or fear at having a

nightmare. The experimenter enacts 8 vignettes with the

dolls used in the emotion labelling tasks, accompanied by

standardized vocal and visual emotion cues. To indicate

how the doll felt, the child affixes to it one of the four

fabric emotion faces. The Non-Stereotypical task measures

how well children can identify feelings in situations where

the protagonist feels differently from how the child would

feel in the same event. Ten vignettes are enacted by the

experimenter using the same dolls as before. For each

vignette 2 points are given for a correct answer, 1 point for

identifying the correct valence and 0 for an incorrect

answer. Total scores for each task were used. Interater

agreement for both types of emotion perspective taking

was 100 %.

Emotion Causes Task

To assess children’s causal understanding of emotion in the

Denham Causes Task, the experimenter presented the four

dolls in turn and asked, ‘‘What made the doll feel this

way?’’ pointing at each doll in random order. Standard

probes were used to encourage children to give more than

one cause for each emotion (happy, sad, angry, afraid) and

to ensure that children’s meanings could be understood.

The procedure and child’s responses are audiotaped and

transcribed responses scored. Fluency of causes for each

emotion is measured as the number of accurate,

Child Psychiatry Hum Dev (2017) 48:346–357 349

123

independent reasons given. Interater correlations were:

overall (r = 0.84), for happy (r = 0.75), sad (r = 0.88),

angry (r = 0.89) and afraid (r = 0.85).

Ambivalent or Mixed Emotions Task

Three vignettes that provoke ambivalent emotional

responses, e.g. riding a bike for the first time, or antici-

pating the last day of school, are narrated and acted out

using the dolls [39]. The child is asked to explain the co-

occurrence of the two emotions. The emotion mixes are;

vignette 1 happy-sad; vignette 2 happy-angry; vignette 3

excited-scared. Total scores ranged from 0 to 6 (2 points

per vignette for explaining both emotions, 1 point for

explaining only one emotion, and 0 for explaining neither

emotion correctly). Interater correlations were: overall

(r = 0.90), for vignette 1 (r = 0.77), vignette 2 (r = 0.81)

and vignette 3 (r = 0.85).

Empathy

Children’s empathy was assessed by the parent-report

Griffith Empathy Measure (GEM) a 23-item measure with

good reliability and validity [40]. GEM contains subset

measures for affective and cognitive aspects of empathy

[40]. One example of an affective empathy item is ‘My

child gets upset when he/she sees an animal being hurt’.

For cognitive empathy, an example item is ‘When I get sad

my child doesn’t seem to notice’. Mothers read each

statement and rated how much they agree with it using a

nine-point Likert scale that ranged from strongly disagree

(-4) to strongly agree (?4) [40]. The Griffith Empathy

Measure has been recognised not only for its strength in

reliably capturing specific aspects of emotion understand-

ing and sharing (Total empathy, alpha = 0.81; Cognitive

empathy, alpha = 0.62; Affective empathy, alpha = 0.83),

but also for its validity across age and gender [40].

Child Emotional and Behavioural Problems

The Strengths and Difficulties Questionnaire (SDQ) [41], a

25-item parent report was used to characterise the general

levels of child psychopathology. The SDQ provides scores

on three clinical subscales (conduct problems; hyperactiv-

ity; emotional problems) and two interpersonal subscales

(peer problems; pro-social behavior) as well as a total

difficulties score. The questionnaire has good reliability

and the individual clinical subscales show sound external

validity in predicting risk to a related mental disorder and

clinician rated severity of the disorder based on structured

interview [42].

Child Verbal Ability

The Expressive Vocabulary Test (EVT0) [43] is an indi-

vidually-administered test of expressive vocabulary. It can

be completed quickly (ceiling is reached with 5 consecu-

tive errors) and does not require the child to read, write, or

give lengthy oral responses. A child’s score is norm-ref-

erenced against children of the same age in months.

Demographic Information

Parents completed a brief demographic questionnaire with

information about child age and education, parental age,

education, employment status, and income range, marital

status and number of children.

Parent Emotional Status

Parents answered the 42 item-version of the Depression,

Anxiety and Stress Scale or DASS [44]. DASS assesses the

severity of the symptoms of depression, anxiety or stress.

Respondents used a 4-point severity/frequency scale to rate

the extent to which they have experienced each state over

the past week.

Data Analysis Plan

Differences between the diagnostic groups and boys and

girls on each of the emotional competencies were exam-

ined using multivariate general linear models with group

and sex as fixed factors and child’s age and expressive

language (EVT) as covariates. Between-group differences

on the two empathy constructs (cognitive; affective) were

evaluated in group by sex models with child’s age and

expressive language ability as covariates. Our sampling

resulted in a notable proportion (26 %) of comparison

children having an axis 1 internalising diagnoses, so the

group factor included 3 groups [ODD; Anxiety diagnosis

(AD); healthy typically-developing (TD)]. We report the

simultaneous planned contrasts on each component of the

emotional competency domain when the between group

effect is significant. Planned contrasts compared ODD

children with TD children, ODD with AD; ODD children

with all other children (TD and AD combined) and TD

with AD children for each component of the domain.

A second set of analyses compared ODD children

classified into high and low CU groups and the two com-

parison non-ODD groups (AD; TD) using multivariate

general linear model with group (HCU; LCU; AD; TD) and

sex as fixed factors and child’s age and expressive lan-

guage ability as covariates. For significant between group

effects we examined planned contrasts comparing HCU

350 Child Psychiatry Hum Dev (2017) 48:346–357

123

with LCU, HCU with AD, HCU with TD, LCU with AD

and TD children.

Results

Demographic and Clinical Characteristics

Of the 124 children formally assessed, 74 had a DSM-IV

diagnosis of ODD, 33 had no diagnosis (TD) and 17 had a

diagnosis other than ODD. The primary diagnoses of the

Non-ODD group were anxiety disorders (12; 4 specific

phobia, 3 overanxious disorder and 3 another anxiety dis-

order) and ADHD (5). Forty-one (55.4 %) ODD children

had a comorbid diagnosis of ADHD. Other secondary

diagnoses for the ODD group included specific phobia (4),

another anxiety disorders (7) and sleep terror (1), while the

other group had secondary diagnosis of social phobia (2),

overanxious disorder (1) and enuresis (2). Clinician relia-

bility of a diagnosis based on the DISCAP interview was

strong (Kappa = 0.96). The majority of children come

from two parent families (89.9 %) and the diagnostic

groups did not differ in the proportion of children from

single parent families (v2 = 0.481; p = 0.49). Table 1 provides demographic and clinical characteris-

tics for the ODD and comparison groups and reports the

results for the group effect and the pair-wise between group

contrasts. There was a trend for ODD to score more poorly

on expressive vocabulary than TD (t (105) = 1.79;

p = 0.07).

Comparison Between ODD, TD, AD

Table 2 presents emotional competency and empathy

scores by group and sex as well as the results for the group

MANOVA and planned contrasts. There were no signifi-

cant sex effects, or group and sex interactions controlling

for child’s age and expressive language ability. There was

a significant between group effect only for emotion causes

(F (8224) = 2.42, p = 0.016; partial g2 = 0.08). Both ODD and AD children produced significantly fewer causes

for angry than the TD children (contrast esti-

mate = -1.269; p = 0.008; 95 % CI -2.199 to -0.399;

est. = -1.782; p = 0.001; 95 % CI -3.123 to -0.441

respectively) but did not differ from one another

(est. = -0.513; p = 0.405; 95 % CI -1.727 to 0.701).

ODD children produced fewer causes for afraid compared

to the AD children (est. = -1.224; p = 0.037; 95 % CI

-2.374 to -0.074) but their poorer performance relative to

TD children failed to reach significance (est. = -0.834;

p = 0.063; 95 % CI -1.715 to 0.046). AD children did not

differ from TD children (est. = -0.390, p = 0.545; 95 %

CI -1.66 to 0.88). There were no significance between

group differences for emotion recognition, emotion per-

spective taking, mixed emotions or empathy.

Comparison Between High CU ODD, Low CU ODD,

and Comparison Groups

As only 5 girls were classified HCU these comparisons

were for boys only. Twenty-two ODD boys were classified

as HCU and 25 LCU. Because only 7 boys were in the AD

group the analyses were undertaken using three groups

(HCU, n = 22; LCU, n = 25; TD, n = 17). HCU had

higher severity of conduct problems than LCU on the SDQ

conduct problem subscale (t(46) = 2.091, p = 0.041).

SDQ conduct problem score was included as an additional

covariate in the analyses.

The scores on emotional competencies and for empathy

for the three groups are presented in Table 3 adjusted for

the age, language ability and conduct problem severity.

There were no significant between group differences for

emotion perception or non-stereotypical perspective-tak-

ing. HCU showed stereotypical perspective-taking deficits

relative to LCU (est = -0.976; p = 0.042; 95 % CI

-1.914 to -0.038) but not relative to TD (est = -0.801;

p = 0.256; 95 % CI -2.20 to 0.597) while LCU did not

differ from the TD children (est = 0.175; p = 0.769; 95 %

CI -1.538 to 1.009). HCU and LCU did not differ on the

emotions causes task for any emotions while both groups

showed deficits relative to TD on emotion causes for anger

(HCU vs TD; est. = -2.141; p = 002; 95 % CI -3.785 to

-0.497; LCU vs TD est. = -1.698; p = 0.003 95 %CI

-3.098 to -0.292). HCU was poorer on the mixed emo-

tions tasks compared to LCU (est = -0.642; p = 0.005;

95 % CI -1.081 to -0.203 happy/sad; est. = -0.453;

p = 0.028; 95 % CI -0.860 to -0.046 happy/angry;

est = -0.692; p = 0.001; 95 % CI -1.095 to -0.289

excited/scared). HCU did not differ from the TD group on

any of the 3 mixed emotions tasks (est. = 0.221,

p = 0.548 95 %CI -0.954 to 0.512; est. = 341;

p = 0.315 95 % CI -1.107 to 0.334; est. = 0.068;

p = 0.844 95 % CI 0.752–0.657 respectively). While LCU

and TD did not differ on the happy/sad or happy/angry

mixed emotions task (est. = 0.390; p = 0.228 95 %CI

-1.031 to 0.251; est. = 0.122; p = 0.680 95 %CI -0.713

to 0.468) LCU scored significantly higher than TD on the

excited/scared task (est. = 0.619; p = 0.043 95 % CI

1.218–0.020). HCU had significantly lower scores on

cognitive empathy compared to LCU (est = -7.631;

p = 0.002; 95 % CI -12.227 to -3.035) but not on

affective empathy (est = -9.297; p = 0.068; 95 % CI

-18.87 to 0.722) while LCU did not differ from TD

(est. = -0.092; p = 0.979 95 % CI -6.87 to 7.053;

est. = -5.099; p = 0.495 95 %CI -20.98 to 9.901). HCU

did not differ significantly from TD on cognitive empathy

Child Psychiatry Hum Dev (2017) 48:346–357 351

123

(est = -7.771; p = 0.06; 95 % CI -15.758 to 0.336) or

affective empathy (est. = -4.045; p = 0.639; 95 % CI

-21.38 to 13.295).

Discussion

This study is the first to compare well defined groups of

children with ODD to other children on behavioural mea-

sures of emotional competencies beyond basic emotion

expression recognition abilities and to examine the specific

influence of children’s level of CU traits. Our findings take

in account variability in emotional competencies related to

sex, age and expressive language abilities. We found,

firstly, that children with ODD were less fluent than typi-

cally developing children in generating causes for angry,

and for afraid when compared to children with an anxiety

disorder but showed no relative deficiencies in emotion

perception, emotion perspective-taking or in the ability to

appreciate mixed emotions. Secondly, we found that ODD

boys with high levels of CU traits have marked deficits

relative to low CU ODD boys in emotion perspective-

taking and understanding mixed emotions.

There are two important inferences from the between

group findings for ODD children overall. First, when age

and expressive language abilities are taken into account

emotional competencies deficits are not characteristic of

ODD children as a group. They are no less proficient at

emotion perception, emotion perspective-taking and

understanding mixed emotions than other children using

these behaviourally based measures of these competencies.

Second, expressive language predicted variability in the

Table 1 Demographic and clinical characteristics of the ODD, Non-ODD and TD groups

ODD (n = 74) AD (n = 12) TD (n = 33) df v2 or F1 Contrasts results

Child gender (%)

Male 63.5 (n = 43) 64.7 (n = 11) 51.5 (n = 17) 2 1.52 ns

Female 36.5 (n = 31) 35.3 (n = 6) 48.5 (n = 14)

Age

Child (months) 70.9 (16.8) 80.8 (17.7) 70.8 (18.9) 121 1.74 ns

Mother (years) 39.3 (4.4) 39.9 (5.1) 41.5 (4.3) 115 2.77 ns

Family annual income (%)

$31,000–65,000 14.9 12.5 3.1 12 10.59 ns

$66,000–99,000 23.0 18.8 21.9

$99,000? 55.4 68.8 71.9

Not reported 6.8 0.0 3.1

Child psychopathology

SDQ emotional symptoms 3.33 (2.68) 3.92 (2.35) 1.72 (1.37) 118 6.31** ODD = AD [ TD2

SDQ conduct problems 4.81 (1.75) 1.33 (1.37) 1.41 (1.39) 118 60.61** ODD [ AD = TD3

SDQ hyperactivity 6.32 (2.9) 4.17 (2.55) 3.13 (2.22) 118 16.76** ODD [ AD = TD4

ICU total 27.11 (10.16) 16.92 (4.29) 16.53 (6.79) 116 18.21*** ODD [ AD = TD5

EVT-2 107.89 (11.73) 111.92 (11.3) 111.82 (6.84) 121 1.68 ns

Mother DASS

Depression 4.24 (7.34) 4.58 (4.98) 2.56 (3.57) 116 1.11 ns

Anxiety 2.99 (4.7) 2.08 (2.46) 1.66 (2.8) 116 1.27 ns

Stress 12.01 (7.44) 10.42 (5.90) 8.28 (4.91) 116 3.45* ODD = Non-ODD [ TD6

Means are presented with SDs in parentheses

SDQ Strengths and Difficulties Questionnaire, ICU inventory of callous–unemotional traits, EVT-2 expressive vocabulary test, DASS depression

anxiety stress scales

* p \ 0.05; ** p \ 0.01; *** p \ 0.001 1 v2 for analyses of sex and family income. F for all other analyses 2

ODD vs TD t (113) = 3.24; p = 0.002; ODD vs AD t (113) = 0.79; p = 0.43; AD vs TD; t (113) = 2.75; p = 0.007 3

ODD vs TD t (113) = 9.86; p \ 0.001; ODD vs AD t (113) = 6.86; p \ 0.001; AD vs TD t (113) = 0.13; p = 0.89 4

ODD vs TD t (113) = 5.47; p \ 0.001; ODD vs AD t (113) = 2.56; p = 0.012; AD vs TD t (113) = 1.14; p = 0.26 5

ODD vs TD t (113) = 5.44; p \ 0.001; ODD vs AD t (113) = 3.66; p \ 0.001; AD vs TD t (113) = 0.13; p = 0.90 6

ODD vs TD t (113) = 2.62; p = 0.01; AD vs TD t (113) = 0.94; p = 0.35; ODD vs AD t (113) = 0.77; p = 0.45

352 Child Psychiatry Hum Dev (2017) 48:346–357

123

simple and complex emotional perspective-taking tasks

independently of age indicating the importance of consid-

ering language ability when examining emotional compe-

tences in relation to behavioural and emotional difficulties.

As children with disruptive behaviour often have language

difficulties [45] any observed association between ODD

and emotional competencies [21] might be explained by

uncontrolled variability in language abilities.

Both ODD and clinically anxious children showed def-

icits in generating causes for anger with the capacity to

identify causes for afraid differentiating the two groups.

The difference for afraid is due not only to a paucity for

ODD children in generating causes but also to enhanced

performance by the children who had an anxiety disorder.

This latter finding reflects the hyper-cognising about fear

situations that is typical of children with anxiety disorders

[23]. Taken together the results indicate that ODD children

show some general deficit in generating emotion causes

which is strongest for anger. While we did not examine the

attributional character of the causes generated, the results

are consistent with the proposal of ODD-related attribu-

tional bias for anger and hostility which constrains the

Table 2 Means (SD) of emotional competencies and empathy for ODD, AD and TD groups by sex and MANOVA and contrast results

ODD AD TD df Wilk’s k F

Male

(n = 47)

Female

(n = 27)

Total

(n = 74)

Male

(n = 7)

Female

(n = 5)

Total

(n = 12)

Male

(n = 17)

Female

(n = 16)

Total

(n = 33)

Emotion perception 4, 220 0.937 1.82

Expressive 7.02

(1.13)

7.11

(1.34)

7.05

(1.20)

6.71

(1.50)

6.60

(0.55)

6.67

(1.16)

7.41

(0.80)

7.19

(0.98)

7.30

(0.88)

2, 111 2.71

Receptive 7.91

(0.35)

8.15

(1.63)

8.00

(1.02)

8.00

(0.01)

7.40

(0.90)

7.75

(0.62)

7.82

(0.53)

7.94

(0.25)

7.88

(0.42)

2, 111 1.09

Emotion perspective-taking 4, 216 0.995 0.13

Stereotypic 14.54

(1.89)

14.89

(3.24)

14.89

(2.46)

15,29

(0.96)

15.80

(0.46)

15.50

(0.80)

15.19

(0.99)

15.19

(1.17)

15.19

(1.06)

2, 109 0.24

Non-

stereotypic

17.43

(2.00)

17.59

(5.06)

17.49

(3.42)

17.29

(0.76)

19.40

(0.55)

18.17

(1.27)

18.19

(1.56)

18.00

(1.86)

18.09

(1.69)

2, 109 0.13

Emotion causes 8, 212 0.861* 2.05*

Happy 2.91

(2.00)

4.30

(3.55)

3.42

(2.73)

4.00

(1.73)

6.60

(6.69)

5.08

(4.44)

4.44

(3.03)

5.20

(4.13)

4.81

(3.56)

2, 109 1.80

Angry 2.17

(1.59)

3.44

(2.59)

2.64 a

(2.09)

2.86

(0.69)

3.40

(2.79)

3.08 b

(1.78)

4.13

(2.80)

4.13

(3.36)

4.13 a,b

(3.03)

2, 109 4.00**

Sad 2.17

(2.01)

3.89

(2.67)

2.80

(2.40)

3.14

(1.35)

3.00

(2.24)

3.08

(1.68)

3.44

(2.63)

3.80

(2.83)

3.62

(2.69)

2, 109 1.06

Afraid 1.66

(1.65)

2.63

(1.76)

2.01 a

(1.74)

2.71

(0.95)

6.00

(7.65)

4.08 a

(4.96)

2.75

(2.21)

3.47

(2.36)

3.10

(2.27)

2, 109 3.53*

Mixed emotions 6, 210 0.89 1.98

Happy/sad 0.87

(0.86)

1.35

(0.89)

1.04

(0.90)

1.57

(0.79)

1.60

(0.89)

1.58

(0.79)

1.06

(0.93)

1.40

(0.91)

1.23

(0.92)

2, 107 0.28

Happy/

angry

1.24

(0.85)

1.42

(0.9)

1.31

(0.87)

1.86

(0.38)

1.40

(0.89)

1.67

(0.65)

1.38

(0.72)

1.60

(0.74)

1.48

(0.72)

2, 107 0.39

Excited/

scared

1.43

(0.81)

1.65

(0.75)

1.51

(0.79)

2.00

(0.01)

1.40

(0.89)

1.75

(0.62)

1.31

(0.87)

1.47

(0.83)

1.39

(0.84)

2, 107 1.37

Empathy 4, 204 0.943 1.50

Cognitive 4.40

(7.68)

6.76

(6.89)

5.28

(7.43)

7.71

(8.18)

9.80

(7.43)

8.58

(7.60)

9.06

(4.99)

8.14

(3.23)

8.65

(4.25)

2, 102 2.25

Affective 0.36

(14.79)

5.76

(6.85)

2.37

(12.65)

11.43

(9.16)

9.60

(14.40)

10.67

(11.05)

5.88

(8.38)

7.36

(6.22)

6.55

(7.40)

2, 102 1.98

For all MANOVAs, design is Intercept ? language ? age ? sex ? Diagnostic Group ? sex*Diagnostic Group a,b

Means with identical superscripts are significantly different in pair-wise between group contrasts

* p \ 0.05; ** p \ 0.01

Child Psychiatry Hum Dev (2017) 48:346–357 353

123

range of causes available for children with difficulties

regulating anger to consider [3, 4].

In the between group analyses for the boys, level of CU

traits did not influence the degree to which ODD boys had

deficits in generating causes for anger. The two subgroups

of ODD boys did not differ on this ability and both showed

a paucity relative to typical boys. The level of callous–

unemotional traits did, however, moderate the performance

of ODD boys on both stereotypical and more complex

emotion perspective-taking tasks. While ODD boys with

high CU traits were as accurate as their peers in recog-

nising and labelling basic emotions they fell significantly

below their low CU ODD peers in appreciating how

another child might routinely feel in relatively straight-

forward emotion contexts. This result is consistent with

inferences about the nature of emotion perspective-taking

in high CU children [2] and with the reports of parents and

teachers of low levels of empathy in these children which

are replicated here for cognitive empathy [8]. Our results

also show that high CU ODD boys display notable deficits

when the emotion perspective-taking task involves under-

standing competing emotional responses. In other words,

for the boys with ODD those with a high level of CU traits

showed a deficiency relative to their low level CU peers in

emotional perspective-taking for situations in which a

person could feel more than one emotion and those

Table 3 Means (SE) on measures of emotional competencies and empathy for the HCU, LCU, and TD and contrast results adjusted for child age, EVT and SDQ conduct problem severity

HCU (n = 22) LCU (n = 25) TD (n = 17) Planned contrast

HCU vs LCU HCU vs TD LCU vs TD

Emotion perception

Expressive 6.76

(0.66)

7.10

(0.22)

7.37

(0.29)

ns ns ns

Receptive 7.87

(0.10)

7.91

(0.08)

7.92

(0.11)

ns ns ns

Emotion perspective-taking

Stereotypic 14.15

(0.40)

15.12

(0.32)

14.95

(0.44)

p = 0.042 ns ns

Non-stereotypic 17.88

(0.43)

17.41

(0.34)

17.54

(0.48)

ns ns ns

Emotion causes

Happy 2.52

(0.53)

3.55

(0.43)

3.99

(0.60)

ns ns ns

Angry 1.71

(0.45)

2.23

(0.37)

4.66

(0.63)

ns p = 0.002 p = 0.003

Sad 1.98

(0.53)

2.79

(0.43)

2.86

(0.59)

ns ns ns

Afraid 1.45

(0.41)

1.87

(0.33)

2.71

(0.46)

ns ns ns

Mixed emotions

Happy/sad 0.63

(0.18)

1.24

(0.15)

0.85

(0.26)

p = 0.005 ns ns

Happy/angry 1.01

(0.17)

1.47

(0.14)

1.35

(0.24)

p = 0.028 ns ns

Excited/scared 1.12

(0.17)

1.81

(0.14)

1.19

(0.24)

p = 0.001 ns p = 0.043

Empathy

Cognitive -0.67

(1.80)

6.96

(1.50)

7.05

(2.93)

p = 0.002 ns ns

Affective -4.81

(3.87)

4.34

(3.24)

-0.76

(6.31)

ns ns ns

For all contrasts, design is Intercept ? language ? age ? CP severity ? Diagnostic Group

354 Child Psychiatry Hum Dev (2017) 48:346–357

123

emotion have competing valencies like happy and sad. On

average high CU ODD boys could not provide an account

of an alternate emotion in the mixed emotion contexts.

These deficits in simply and more complex affective per-

spective taking suggest that high levels of CU traits in

disruptive children are associated with a deficiency or

delay in properly understanding the subjective nature of

emotions and the role played by one own or other’s internal

states such as belief and desire in emotion experience [14].

The poorer comprehension of mixed emotions in high

CU boys is also consistent with evidence that better

understanding of ambivalent emotions in young children is

associated with higher moral orientation and more dis-

comfort with transgression [19, 46]. As high CU boys did

not show a relative deficit compared to low CU ODD boys

in recognising individual emotions or in generating causes

for specific emotions, our results suggest that high CU in

ODD boys is associated with poorer recognition of the co-

occurring and conflictual nature of salient goals of the

protagonist in these emotional scenarios. This may occur

either because certain type of goals described in these sit-

uations, e.g. sad because a friend is missing your party, are

simply not emotionally salient for high CU boys or alter-

natively, because once they identify with a particular goal

of the protagonist high CU ODD boys are less able to shift

attention towards appreciating alternate, emotion-relevant

goals during the course of the event. There is continuing

debate about whether ambivalent emotional situations elicit

a blended, complex emotion or result in a sequentially

experiencing of discrete but competing emotions [47].

Regardless, our results highlight that disruptive boys

described as showing low levels of pro-social behaviours

and emotions have marked deficiencies in dealing with

these more affectively ambivalent situations at a time when

such contexts are becoming increasingly prevalent in

children’s interpersonal and emotional experience [10, 11].

Deficits in understanding ambivalent emotional context

have significant implications for the development of pro-

social behaviour. The experience of a conflict of goals in

emotional and interpersonal contexts together with the

corresponding mixed emotions and resultant negative

affect are considered by various models [17, 20, 46] to be

of critical importance to the development of the social

emotions such as embarrassment and guilt which predict

and may motivate pro-social responding [19]. While it is

not possible to make inferences about high CU ODD

children’s experience of negative affect in ambivalent

emotion contexts from this study, their failure to appreciate

the possibility of mixed and competing emotions increases

the chances that they don’t experience negative affect in

these contexts and are less likely to learn the incongruence

between their emotional experience and what is socially

accepted. The deficit also suggests that high CU boys will

respond inappropriately or with lack of concern to others

who are feeling ambivalent or mixed emotions in these

situations. As more complex social contexts become more

usual, the deficit in understanding ambivalent emotions in

others may play an increasing role in maintaining or

worsening ineffective social interactions. For the majority

of children the capacity to appreciate mixed emotions is

continuing to develop during the early school years [18,

29]. Our results suggest that efforts to enhance emotional

competence in these years could make use of these

opportunities to improve children’s understanding of mixed

emotional contexts. For many boys who have persistent

conduct problems enhancing their capacity to appreciate

the conflicting nature of these contexts and the competing

emotions which can arise may be an important step towards

modifying the enduring negative impact of high callous–

unemotional traits on social interactions.

While high CU ODD boys consistently scored the

lowest of the three groups, we did not find significant

differences in emotional competencies between them and

the typically developing boys except in generating causes

for anger. In these comparisons the number of boys in the

typically developing group was small and with adjustments

for variability in the emotional competencies associated

with severity of conduct problem, age and expressive

language there may not have been adequate power to detect

a significant effect. The decision to control for severity of

conduct problem enhanced inferences about CU specific

deficits in emotion competencies in the ODD boys.

Adjusting for the severity of conduct problem may have,

however, unnecessarily constrained the comparisons

between high CU ODD boys and typically developing

boys. A post hoc analysis of the contrasts between high CU

boys and typical boys without including severity of con-

duct problems as a covariate found statistically significant

deficits for high CU boys for the mixed emotion tasks and

for cognitive and affective empathy. At the same time,

ODD boys without high levels of CU showed no relative

deficits in the comparisons with typical boys except for the

mixed emotions task involving the emotions of excitement

and fear where they were significantly better than our

group of typically developing boys. Taken together our

results indicate that behavioural deficits in advanced

emotion competencies important for the development of

moral reasoning and pro-social responding are specific to

disruptive children with high levels of CU traits.

There are a number of limitations of the current study

which need to be considered. Because we recruited on the

basis of parent’s interest about their child’s emotional

development, there may be a bias in the selection of typi-

cally developing children to those with some difficulties in

emotional and behavioural regulation and therefore

towards those with weaker emotional competencies. At the

Child Psychiatry Hum Dev (2017) 48:346–357 355

123

same time identifying children with other mental disorders

allowed a control for clinical status on emotional compe-

tencies and firmer conclusions about specific ODD related

emotional deficiencies. This control is often absent in other

comparative studies of emotional competencies in exter-

nalising children [21]. Although the validity and usefulness

of the Denham tasks as behavioural measures of emotional

abilities is well established in this age group [37–39], they

may not have the sensitivity to detect all ODD and high CU

related differences. For example, performance on the

emotion perception and basic emotion perspective-taking

tasks were skewed towards high scores. All the Denham

emotion competency measures provide contextual and

expressive cues to the emotions so these results are not

directly comparable to studies which examine high CU

related deficits in recognising emotions solely from

expressive cues in face and posture without contextual

information. While the tasks used in the current study are

important in extending current knowledge of CU related

emotion abilities difficulties beyond emotion recognition,

the nature of these difficulties and their impact on high CU

disruptive children’s emotional and social functioning will

need to be replicated and clarified using a diverse range of

ecologically valid measures.

Summary

The focus of most empirical work examining emotional

competencies in ODD children, including those with high

callous–emotional traits, has been on emotion perception.

Emotion perception abilities contribute to the development

of a number of more advanced emotional competencies

which underlie children’s developing emotional under-

standing and their capacity to respond appropriately to their

own and other’s emotions. The aim of the current study

was to compare performance of ODD children on more

advanced emotional abilities involving emotion perspec-

tive-taking, understanding the causes of emotions and

appreciating mixed emotions. The results of the current

study show that a paucity in the attribution of causes to

emotional experiences is the only emotional deficit which

characterises ODD children as a group. Impairments in

emotion perspective-taking and appreciation that conflict-

ing emotions can co-occur is specific to ODD boys who

show high levels of callous–unemotional traits. The find-

ings have implications for theoretical understanding of the

nature of emotional abilities in disruptive children with low

pro-social emotions and behaviours. The findings may also

contribute to programs which aim to help young children

improve their understanding of emotions and their ability

to respond appropriately to their own and to other’s

feelings.

Funding The study was supported by an Australian Research Council grant (DP110101990) awarded to O’Kearney and Salmon. Dr

Dawel is supported in part by a grant from the Australian

Research Council Centre of Excellence for Cognition and its

Disorders.

Compliance with Ethical Standards

Conflict of Interest The authors declare that they have no conflict of interest.

References

1. American Psychiatry Association (2013) Diagnostic and statisti-

cal manual of mental disorders, 5th edn. American Psychiatric

Association, Arlington

2. Blair RJ (2005) Responding to the emotions of others: dissoci-

ating forms of empathy through the study of typical and psy-

chiatric populations. Conscious Cogn 14:698–718

3. Crick NR, Dodge KA (1994) A review and reformulation of

social information-processing mechanisms in children’s social

adjustment. Psychol Bull 115:74–101

4. Marsh AA, Blair RJ (2008) Deficits in facial affect recognition

among antisocial populations: a meta-analysis. Neurosci Biobe-

hav Rev 32:454–465

5. Frick PJ, White SF (2008) Research review: the importance of

callous–unemotional traits for developmental models of aggres-

sive and antisocial behavior. J Child Psychol Psychiatry

49:359–375

6. Blair RJ, Budhani S, Colledge E, Scott S (2005) Deafness to fear

in boys with psychopathic tendencies. J Child Psychol Psychiatry

46:327–336

7. Dadds MR, Hawes DJ, Perry Y, Merz S, Riddell AC, Haines DJ

et al (2006) Attention to the eyes and fear-recognition deficits in

child psychopathy. Br J Psychiatry 189:280–281

8. Dadds MR, El Masry Y, Wimalaweera S, Guastella AJ (2008)

Reduced eye gaze explains ‘‘fear blindness’’ in childhood psy-

chopathic traits. J Am Acad Child Adolesc Psychiatry

47:455–463

9. Rigato S, Farroni T (2013) The role of eye gaze in the processing

of emotional facial expressions. Emot Rev 5:36–40

10. Pons F, Harris P, de Rosnay M (2004) Emotion comprehension

between 3 and 11 years: developmental periods and hierarchical

organization. Eur J Dev Psychol 1:127–152

11. Wellman HM (1995) Young children’s conception of mind and

emotion: Evidence from English speakers. In: Russell JA, Fer-

nandez-Dols J, Manstead AS, Wellenkamp JC (eds) Everyday

conceptions of emotion. Kluwer, Dordecht, pp 289–313

12. Tenebaum HR, Alfieri L, Brooks PJ, Dunne G (2008) The effects

of explanatory conversations on children’s emotion understand-

ing. Br J Dev Psychol 26:249–263

13. Denham SA, Blair KA, DeMulder E, Levitas J, Sawyer K,

Auerbach-Major S, Queenan P (2003) Preschool emotion com-

petence: pathway to social competence. Child Dev 74:238–256

14. Lane JD, Wellman HM, Evans EM (2010) Children’s under-

standing of ordinary and extraordinary minds. Child Dev

81:1475–1489

15. Wellman HM, Liu D (2004) Scaling of theory-of-mind tasks.

Child Dev 75:523–541

16. Larsen JT, To Y, Fireman G (2007) Children’s understanding and

experience of mixed emotions. Psychol Sci 18:186–191

356 Child Psychiatry Hum Dev (2017) 48:346–357

123

17. Langdon R, Mackenzie C (eds) (2012) Emotion, imagination and

moral reasoning. Psychology Press, Hove

18. Lewis M (2008) Self-conscious emotions: embarrassment, pride,

shame, and guilt. In: Lewis M, Haviland-Jones JM, Barrett LF

(eds) Handbook of emotions. Guildford Press, New York,

pp 742–756

19. Malti T, Krettenauer T (2013) The relation of moral emotion

attributions to pro- and antisocial behavior: a meta-analysis.

Child Dev 84:397–412

20. Tangney JP, Miller RS, Flicker L, Barlow DH (1996) Are shame,

guilt, and embarrassment distinct emotions? J Personal Soc

Psychol 7:1256–1269

21. Trenacosta CJ, Fine SE (2010) Emotion knowledge, social

competence and behaviour problems in childhood and adoles-

cents: a meta-analytic review. Soc Dev 19:1–29

22. Dawel A, O’Kearney R, McKone E, Palermo R (2013) Not just

fear and sadness: meta-analytic evidence of pervasive emotion

recognition deficits for facial and vocal expressions in psy-

chopathy. Neurosci Biobehav Rev 36:2288–2304

23. O’Kearney R, Dadds MR (2005) Language for emotions in

adolescents with externalizing and internalizing disorders. Dev

Psychopathol 17:529–548

24. Speltz ML, McClellan J, DeKlyen M, Jones K (1999) Preschool

boys with oppositional defiant disorder: clinical presentation and

diagnostic change. J Am Acad Child Adolesc Psychiatry

38:838–845

25. Cadesky EB, Mota VL, Schachar RJ (2000) Beyond words: how

do children with ADHD and/or conduct problems process non-

verbal information about affect? J Am Acad Child Adolesc

Psychiatry 39:1160–1167

26. Ruby P, Decety J (2004) How would you feel versus how do you

think she would feel: a neuroimaging study of perspective-taking

with social emotions. J Cogn Neurosci 16:988–999

27. Vaish A, Carpenter M, Tomasello M (2009) Sympathy through

affective perspective-taking and its relation to pro-social beha-

viour in toddlers. Dev Psychol 45:534–543

28. O’Kearney R, Dadds MR (2004) Developmental changes in the

language for emotions across the early adolescent years. Cogn

Emot 8:913–938

29. Camras LA, Shuster MM (2013) Current emotion research in

developmental psychology. Emot Rev 5:321–329

30. Holland D, Dadds MR (1997) The diagnostic interview schedule

for children, adolescents, and parents (DISCAP). Griffith

University, Brisbane

31. Johnson S, Barrett P, Dadds MR, Fox T, Shortt A (1999) The

DISCAP: initial reliability and validity data. Behav Change

16:155–164

32. Frick PJ (2004) The inventory of callous–unemotional traits.

Unpublished rating scale, The University of New Orleans,

Orleans

33. Fanti KA, Frick PJ, Georgiou S (2009) Linking callous–unemo-

tional traits to instrumental and non-instrumental forms of

aggression. J Psychopathol Behav Assess 31:285–298

34. Kimonis ER, Ogg J, Fefer S (2014) The relevance of callous–

unemotional traits to working with youth with conduct problems.

Communiqué 42:16–18

35. Frick PJ, Moffitt TE (2010) A proposal to the DSM–V childhood

disorders and the ADHD and disruptive behavior disorders work

groups to include a specifier to the diagnosis of conduct disorder

based on the presence of callous–unemotional traits. American

Psychiatric Association, Washington

36. Dadds MR, Allen JL, McGregor K, Woolgar M, Viding E, Scott

S (2014) Callous–unemotional traits in children and mechanisms

of impaired eye contact during expressions of love: a treatment

target? J Child Psychol Psychiatry 55:771–780

37. Denham SA (1986) Social cognition, prosocial behavior and

emotion in preschoolers: contextual validation. Child Dev

57:194–201

38. Denham SA, Zoller D, Couchoud EA (1994) Socialization of

preschoolers’ emotion understanding. Dev Psychol 30:928–936

39. Denham SA, Kochanoff AT (2002) Parental contributions to

preschoolers’ understanding of emotion. Marriage Fam Rev

34:311–343

40. Dadds MR, Hunter K, Hawes DJ, Aaron DJ, Frost AD, Vassallo S

et al (2008) A measure of cognitive and affective empathy in

children using parent ratings. Child Psychiatry Hum Dev

39:111–122

41. Goodman R (2001) Psychometric properties of the Strengths and

Difficulties Questionnaire (SDQ). J Am Acad Child Adolesc

Psychiatry 40:1337–1345

42. Hawes DJ, Dadds MR (2004) Australian data and psychometric

properties of the Strengths and Difficulties Questionnaire. Aust

NZ J Psychiat, 38:644–651

43. Williams KT (2007) Expressive vocabulary test, manual, 2nd

edn. NCS Pearson, Minneapolis

44. Lovibond SH, Lovibond PF (1995) Manual for the depression

anxiety stress scales, 2nd edn. Psychology Foundation, Sydney

45. Yew SGK, O’Kearney R (2015) The role of early language dif-

ficulties in the trajectories of conduct problems across childhood.

J Abnorm Child Psychol 43:1515–1527

46. Dunn J, Brown JR, Maguire M (1995) The development of

children’s moral sensibility: individual differences and emotion

understanding. Dev Psychol 31:649–659

47. Berrios R, Totterdell P, Kellett S (2015) Investigating goal con-

flict as a source of mixed emotions. Cogn Emot 29:755–763

Child Psychiatry Hum Dev (2017) 48:346–357 357

123

Child Psychiatry & Human Development is a copyright of Springer, 2017. All Rights Reserved.