Psychiatry Research 187 (2011) 234240

Contents lists available at ScienceDirect

Psychiatry Research
j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / p s yc h r e s

Emotional sensitivity in youth with borderline personality pathology

Martina Jovev a,, Andrew Chanen a,b, Melissa Green c,d, Sue Cotton a, Tina Proftt a,
Max Coltheart e, Henry Jackson a,f
a

Orygen Youth Health Research Centre, Centre for Youth Mental Health, The University of Melbourne, Melbourne, Australia
Orygen Youth Health Clinical Program, Northwestern Mental Health, Melbourne, Australia
School of Psychiatry, University of New South Wales, Sydney, Australia
d
Black Dog Institute, Prince of Wales Hospital, Randwick, Australia
e
Macquarie Centre for Cognitive Science, Macquarie University, Sydney, Australia
f
Department of Psychological Sciences, The University of Melbourne, Melbourne, Australia
b
c

a r t i c l e

i n f o

Article history:
Received 18 January 2010
Received in revised form 3 November 2010
Accepted 8 December 2010
Keywords:
Borderline personality disorder
Affect
Adolescence

a b s t r a c t
If Borderline Personality Disorder (BPD) is characterized by an underlying emotional sensitivity, individuals
with this disorder would be expected to demonstrate accurate identication of emotional expressions at
earlier stages of expression (i.e., lower thresholds of facial expressivity across all emotional valences).
Twenty-one outpatient youth (aged 1524 years) meeting 3 or more DSM-IV BPD criteria and 20 communityderived participants (aged 1524 years) with no history of psychiatric problems were tested on a measure of
emotional sensitivity, the Face Morph Task. In this test faces morph from neutral to each of the six basic
emotional expressions. The BPD group showed no evidence of heightened sensitivity to emotional facial
expressions compared to the community control group (all P N 0.05 and effect sizes ranging from 0 to 0.6).
They require comparable levels of emotional expressivity in order to correctly identify emotions. Therefore,
emotional sensitivity might not be apparent early in the course of BPD. Rather, it might develop later in the
course of the disorder or be present only in severe BPD.
2011 Elsevier Ireland Ltd. All rights reserved.

1. Introduction
Facial expressions of emotions are fundamental social emotional
signals with a strong evolutionary bias and immediate implications
for behaviour. Emotional expressions provide information about the
intentions of others, personal relations, evoke emotional responses in
others and serve as incentives for action (Keltner and Kring, 1998). For
instance, sad expressions have been linked with promoting nurturance and inhibiting aggression and hostility in others, while angry
expressions have been shown to reduce errant behaviour (e.g.,
breaking social rules and expectations) in observers. Dysfunction in
either their own emotional displays or perception of emotions in
others can disrupt or harm social relationships and self-management
by eliciting adverse responses from others and failing to elicit
supportive responses when needed. As such, aberrant processing
and responding to facial emotional expressions might result in
detrimental effects on social functioning.
Emotion dysregulation lies at the core of Linehan's biosocial theory of
BPD (Linehan, 1993), which is one of the most thoroughly delineated
etiological models of borderline pathology (for other models, see:
(Fonagy et al., 2000; Judd and McGlashan, 2003; Kernberg, 1967, 1975,
Corresponding author. Orygen Youth Health Research Centre, Locked Bag 10,
Parkville, Victoria, Australia 3052. Tel.: +61 3 9342 2800; fax: +61 3 9387 3003.
E-mail address: mjovev@unimelb.edu.au (M. Jovev).
0165-1781/$ see front matter 2011 Elsevier Ireland Ltd. All rights reserved.
doi:10.1016/j.psychres.2010.12.019

1976). The dysfunction proposed by Linehan is one of broad dysregulation across all aspects of emotional responding. Individuals with BPD
display greater emotional sensitivity (low threshold for recognition of
emotional stimuli), greater emotional reactivity (high amplitude of
emotional responses), and longer duration of emotional responses
(slower return to baseline arousal). BPD emerges from transactions
between individuals with biological vulnerabilities and specic environmental inuences (e.g., family environment). More specically,
impulsivity is seen as a predisposing vulnerability for both current and
future difculties with emotion regulation, resulting in dysregulation
across cognitive processes, neurochemistry and physiology, facial and
muscle reactions and emotion-linked actions (Crowell et al., 2009).
There is some empirical support for this hypothesis. Individuals with
BPD process information in a negatively biased way (Veen and Arntz,
2000; Meyer et al., 2004), remember more negatively salient words
(and perhaps more negative memories) (Korne and Hooley, 2000), are
more sensitive to emotional stimuli (Frank and Hoffman, 1986; Ladisch
and Feil, 1988), tend to rate faces as less friendly and more rejecting
(Meyer et al., 2004) and have reduced facial emotional expressiveness
to positive as well as negative stimuli (Renneberg et al., 2005). This is
despite having autonomic arousal and startle responses comparable to
the non-BPD control group (Herpertz et al., 1999; Herpertz et al., 2000;
Herpertz et al., 2001) and better mental state discrimination based on
the eye region of the face compared to community controls (Fertuck
et al., 2009).

M. Jovev et al. / Psychiatry Research 187 (2011) 234240

Recently, Domes et al. (2009) systematically reviewed six studies

focusing on BPD patients' recognition and responding to facial stimuli.
Four of these studies used static emotional expressions with somewhat
varying results. In a predominantly female sample (66%), Levine et al.
(1997) reported lower accuracy in recognising facial expression of anger,
fear and disgust in the BPD group (n=30), compared to the nonpsychiatric comparison (n=40) group matched on age and education
level. Similarly, Bland et al. (2004) found impaired recognition accuracy
for fearful angry and sad faces in women with BPD. Minzenberg et al.
(2006) compared outpatients with BPD (n=43, 88% female) to healthy
controls (n=26, 89% female) on emotion recognition tasks across visual
and auditory modalities (facial, prosodic, and integrated facial/prosodic).
Patients with BPD showed normal ability to recognize isolated facial or
prosodic emotions. They did, however, have impaired recognition of
emotions in integrated facial/prosodic stimuli. Impaired discrimination of
non-emotional facial features was also evident and might be related to
some of the more serious symptoms of the disorder (e.g., interpersonal
antagonism). The above studies used a multiple-choice answer format,
where participants had to choose an answer from a list of alternatives. In
contrast, Wagner and Linehan (1999) asked participants to describe the
emotional state of the presented person. They examined recognition of
emotional facial expressions among women diagnosed with BPD
(n=21), compared to a group of women diagnosed with histories of
childhood sexual abuse (n=21) and a group of women with no history of
sexual abuse or BPD (n=20). Unlike the other studies, they found that the
BPD group were accurate in recognising emotions, but had signicantly
heightened sensitivity towards recognition of fear. Nevertheless, the
above studies using static emotional facial expressions have not claried
whether BPD patients show differences in accuracy of labelling specic
emotions or have a different detection threshold for more ecologically
valid ambiguous stimuli (i.e., facial stimuli not at 100% expression or facial
stimuli with ambiguous blends of emotion).
In order to simultaneously examine detection threshold and recognition accuracy for emotional faces, dynamic facial expressions need to be
utilised. Moreover, recognition of emotions relies upon anatomically
separable brain regions, depending on whether the stimuli were static or
dynamic displays of facial affect and there is evidence that different types
of knowledge about emotions may draw upon different neural systems
(Adolphs, 2002; Adolphs et al., 2003). Temporal and limbic-related
cortices may be important in retrieving information about emotions
signalled by static stimuli, while parietal and frontal lobe areas may be
more involved in retrieving knowledge about emotions signalled through
dynamic movement (Adolphs et al., 2003). In addition, the role of brain
areas mediating attention (e.g., inferior medial prefrontal cortex) are
implicated for longer presentation times since multiple cognitive
processes compete when a stimulus is presented for more than three
seconds, such that the observer divides attention between emotions
evoked by the stimulus and their secondary cognitive associations (Geday
et al., 2007). Thus the difference between static and dynamic tasks may
not only be in terms of stimulus presentation but also in the brain regions
activated during the experimental task.
Two studies of emotion processing in BPD to date have used
dynamic facial expressions in their design. Lynch et al. (2006)
demonstrated that participants with BPD (n = 20, 85% female) were
signicantly more likely to correctly identify facial affect (regardless
of valence) at an earlier stage than healthy controls (n = 20, 85%
female), implicating heightened sensitivity to facial emotion in BPD.
Conversely, Domes et al. (2008) found that female BPD patients had
detection thresholds comparable to a female non-clinical control
group. BPD patients (N = 25) showed a signicant reduction in
detection threshold over the course of the experiment, compared to a
control group (N = 25; matched on age, IQ and education), suggesting
that enhanced sensitivity might be particular to emotional expression
of familiar faces. Moreover, they found that BPD patients signicantly
over-reported anger when evaluating ambiguous blends of anger/
sadness and anger/happiness. Some of the differences in ndings

235

between the two studies might be explained by differences in

experimental design. In Lynch et al.'s study, participants were allowed
to change their mind as often as they liked during the course of the
experiment, whereas in Domes et al.'s study the trial was stopped
following the rst response and participants were allowed to label the
facial stimuli without the time restraints. Factors other than
emotional sensitivity, such as the opportunity to guess, are thus likely
to vary between the two studies.
In summary, the majority of research to date does not support the
assumption of general hypersensitivity to facial emotions in BPD. Rather,
this collective evidence suggests more subtle impairments in labelling
accuracy, as well as a tendency to interpret ambiguous faces in a more
negative way (i.e., a bias toward negative emotional perceptions). It is also
of note that females, whether girls or women, perform affect recognition
tasks better than males (McClure, 2000; Thayer and Johnsen, 2000). The
majority of studies reviewed in this paper have either used samples
consisting entirely or mostly of females and thus reect the predominance
of the BPD diagnosis in females. Where mixed samples were used, the
ratio of males to females was closely matched between clinical and control
groups, suggesting that the observed pattern of results was not due to
different gender ratios between clinical and control groups. Some of the
differences in the ndings could be accounted for by the choice of
comparison group, as well as the choice of response format (free response
vs. forced choice). More importantly, group differences in accuracy
observed in the majority of previous studies do not necessarily reect
differences in sensitivity, as the majority have used facial stimuli at 100%
expression without examining stimuli at lower levels of intensity. The use
of dynamic emotional stimuli might be more appropriate, as they are
more effective than static stimuli in activating brain areas that process
emotion (LaBar et al., 2003; Sato et al., 2004).
The abovementioned studies have recruited adult samples (mean
age approximately 30 years or more). As BPD commonly has its onset
in adolescence or early adulthood (Chanen et al., 2008a), the
participants in the above studies are likely to have had long durations
of illness. Therefore, previous research has been confounded by the
effects of chronic BPD, as well as factors associated with the duration
of illness, such as recurrent co-occuring Axis I disorders including
substance use (Zanarini et al., 2004a), deliberate self-harm, cumulative traumatic events (Zanarini et al., 2005) and treatment (including
polypharmacy) (Zanarini et al., 2004b). Examining the processing of
social threat early in the course of BPD is likely to reduce the inuence
of such factors on brain morphology (Chanen et al., 2008b). However,
comparing a younger BPD sample to adults with BPD (N30 yrs of age)
may not be appropriate. This is because in younger participants brain
development may not be complete in the key areas (e.g., frontal lobe)
that are at least partially involved in recognising facial emotions
(Adolphs et al., 2003; Toga et al., 2006). The present study therefore
utilised a community control group in the same age range as the
clinical participants.
The aim of the present study was to explore the emotional sensitivity
hypothesis in a sample of youth with features of BPD. If BPD is characterized by an underlying emotional sensitivity it would be expected that
compared to healthy controls of the same age range, individuals with BPD
features would show accurate identication of emotional expressions at
earlier stages of expression across all presented emotional expressions.
Moreover, it was hypothesised that in comparison to the community
group of the same age range, the BPD group would have higher ratings of
threat and arousal, as well as lower ratings of valence and dominance/
control of the recognised facial expressions.
2. Method
2.1. Participants
Twenty-one outpatients meeting three or more DSM-IV BPD criteria and aged between
15 and 24 years (mean age=18.90, SD=3.10 years; 3 male, 18 female) were recruited from
the HYPE Clinic a specialized early intervention program for BPD (Chanen et al., 2009) at

236

M. Jovev et al. / Psychiatry Research 187 (2011) 234240

Orygen Youth Health, Melbourne, Australia. Participants were recruited within a month of
their rst contact with their assigned therapist at the clinic. Participants had never received
specic treatment for BPD and were physically healthy, based upon medical history. Exclusion
criteria for this group included: visual impairment (e.g., uncorrected vision or colour
blindness); intellectual disability (as indicated in their clinical notes); a schizophrenia
spectrum or affective psychotic disorder; intoxication with alcohol and/or any other licit or
illicit drugs on the day of the procedure; or if they had a history of head injury, epilepsy,
meningitis, encephalitis or a brain infection, loss of consciousness for 10 min or more,
seizures, thyroid disorder or other signicant medical illness.
Twenty healthy controls aged between 16 and 24 years (mean age = 20.40,
SD=2.72 years; 7 male, 13 female) were recruited from the local community through
advertisements placed at public transport stops, libraries and community centres in the area.
Additional exclusion criteria to the BPD criteria were applied to this group: no BPD or
Antisocial PD features, and no history of psychiatric problems.
2.2. Measures
All participants were screened for medical, neurological and sensory disorders, as
well as demographic information, during a short phone interview.
2.2.1. Diagnostic measures for BPD participants
Structured Clinical Interview for DSM-IV Axis I disorders (SCID-I/P: (First et al.,
1997b) was used to assess for psychotic, affective, anxiety, and eating disorders.
For the purposes of this study, only the BPD and the Antisocial PD modules of the
Diagnostic Interview for DSM-IV Personality Disorders (DIPD) were used. This is a
reliable semi-structured interview designed to assess for DSM-IV Axis II disorders
(Zanarini et al., 1996). In keeping with previous research, a personality disorder
criterion was scored positive if it had been present for two (or more years) and did not
occur exclusively during an Axis I disorder.
2.2.2. Screening measures for control participants
The Structured Clinical Interview for DSM-IV Axis I Disorders Non-patient Edition
(SCID-I/NP(First et al., 1996) was administered over the telephone to assess for the
presence of Axis I disorders. Diagnostic modules of the SCID-I/NP are the same as those
of the SCID-I/P (with the psychotic screen). The only differences between the SCID-I/P
and the SCID-I/NP is the overview section in the latter and that in the SCID-NP, there is
no assumption of a chief complaint.
The McLean Screening Instrument for Borderline Personality Disorder (MSI-BPD:
(Zanarini et al., 2003) was used to screen for BPD symptoms in the community sample.
The MSI-BPD is a 10-item yes/no questionnaire that is based upon the BPD module of
the DIPD. It comprises one question for each of the rst eight DSM-IV BPD criteria, and
two questions for the ninth criterion (paranoia/dissociation). The MSI is one of the only
instruments designed specically for the screening of BPD and good psychometric
properties have been reported in non-clinical samples (Zanarini et al., 2003).
Screening for antisocial behaviour was conducted using the Structural Clinical
Interview for DSM-IV Axis II Disorders Personality Questionnaire (SCID-II PQ: (First et al.,
1997a). Only the 15 items corresponding to Criterion A (childhood conduct disorder) were
used, as these items are necessary (but not sufcient) for a diagnosis of antisocial PD.
2.2.3. Facial emotion task
The Face Morph Task was used to assess participants' ability to accurately assess
participants' ability to perceive facial emotional expressions presented at varying
degrees of intensity. This measure was based on the original task by Blair et al. (2001)
utilising Ekman images (Ekman and Friesen, 1976) and modied for use in the present
study at the Macquarie Centre for Cognitive Science. It was specically designed to
address the limitations of prior research while providing a more direct test of Linehan's
(1993) hypothesis regarding the emotional sensitivity of individuals with BPD.
The faces morph from neutral through to 100% expression of each of the six basic
emotions (happy, sad, anger, fear, disgust and surprise) in ve identities. The morphs
were made using 5% intervals (using MorphX program), such that there were 25
images used in the progression from 0 (neutral)-100% (prototypic expression). The
morphs were presented using DMDX 10 software, developed by Forster and Forster
(http://www.u.arizona.edu/~kforster/dmastr/), on a laptop computer. A total of 30
morphed faces were presented in a randomised order, plus two practice trials. An
example of the faces in the Face Morph task is presented in Fig. 1.
Performance on the Facial Morph Task is measured in terms of sensitivity and
impulsivity. Sensitivity was dened as the ability to recognise emotion at lower levels of
intensity. Thus, the earliest correct responses for the six emotions were examined.
Impulsivity was operationalized as the tendency to respond early and incorrectly to the
task of identifying facial emotions.
During the presentation of each morphed facial emotion sequence, participants were
asked to press the space-bar button on the keyboard once they recognised the emotion
shown on the face. The corresponding frame number (i.e., the threshold) was recorded for
the point at which the facial expression was identied as recognised. The recognition
accuracy was consequently measured using a forced-choice format: a) happy, b) sad,
c) angry, d) fear, e) disgust and f) surprise. In addition, the participants were asked to rate
their condence, threat, valence, arousal and dominance/control of the recognised facial
expression on the scale 09. The following questions were asked to obtain the ratings:
a) How condent are you about your choice of emotion?
b) Please rate how HAPPY you feel about this face on the following scale?

c) Please rate how CALM you feel about this face on the following scale?
d) Please rate how IN CONTROL you feel about this face on the following scale?
e) Please rate how THREATENED you feel about this face on the following scale?
2.3. Procedure
All procedures were approved by the Northwestern Mental Health Research and
Ethics Committee, Melbourne, Australia. After complete description of the study
procedure to potential participants, written informed consent was obtained from each
participant and/or from a parent or guardian where appropriate.
All tasks were completed in a quiet room at Orygen Youth Health. Participants were
seated at a desk in front of the laptop computer, orientated to the purpose of the study
and instructions of the task. They were encouraged to ask questions or directions if they
felt the instructions were unclear prior to starting the task.
Prior to commencing the experimental tasks, the DHQ was administered to all
participants. Community participants also completed the screening measures for Axis I
and II disorders. For the clinical participants, the diagnostic interview was completed as
part of the routine entry assessment for the HYPE clinic. Clinicians are trained to a
rigorous standard, using DSM-IV operational criteria, and this standard is maintained
via a consensus diagnosis process for each patient, based upon a modied Longitudinal
Expert All Data (LEAD) standard (Pilkonis et al., 1991). Inter-rater reliability is not
routinely collected. Participants were remunerated AU$30 upon completion of the
tasks.
2.4. Data analysis
Two-way (2 6) repeated measures general linear models were conducted to
determine group differences across the six emotion conditions in terms of sensitivity
and impulsivity. Derived from these models were main effect for group, the main effect
for emotion, as well as the interaction term. The interaction term was of interest for this
study, as we wished to determine whether patterns of response to facial emotion
expressions differed with respect to group membership. Independent samples t-tests
were also used to examine differences on participants' ratings of their condence,
threat, valence, arousal and dominance of the recognised facial expression. Due to the
exploratory nature of the study, Cohen's d values (effect size) were calculated for each
comparison. Cohen's (Cohen, 1992) guidelines for interpreting the magnitude of effect
sizes (ES; small medium and large ESs are d = 0.20, 0.50, and 0.80 respectively) were
used.

3. Results
3.1. Sample characteristics
Key demographic variables for both groups are reported in Table 1.
There were no signicant differences between the groups on age and
sex (all P N 0.05). There was a signicantly higher proportion of
tertiary educated students in the community group and lower
proportion of participants born outside of the Australia/New Zealand
region (all P b 0.05).
All clinical participants had a co-morbid mood and/or anxiety
disorder (57% MDE, 48% PTSD, 48% Panic Disorder, 33% Dysthymia,
24% GAD, 14% Social Phobia, 14% Bulimia, 10% Bipolar II, 10% Specic
Phobia). The mean number of DSM-IV BPD criteria met was 5.10
(SD = 1.58), with 12 (57.1%) participants meeting the full threshold
(5 or more criteria), 6 (29%) participants meeting 4 criteria and 3
(14%) participants meeting 3 criteria. Eleven (52%) participants were
taking antidepressant medication, 2 (10%) were on mood stabilisers
and 1 (5%) was on antipsychotic (atypical) medication.
3.2. Sensitivity and impulsivity to facial affect
3.2.1. Differences in earliest frame for correct response Sensitivity
Two participants from the community group incorrectly identied
disgust across all 5 identities and were therefore excluded from this
analysis. The interaction term (Group Emotion) from the two-way
general linear model was not signicant (P N 0.05). There was a
signicant main effect for Emotion, F(5, 185) = 25.611, p b 0.001,
partial 2 = 0.41; but not for Group (P N 0.05). There was an overall
difference in earliest response to different types of emotions. In
particular, happy emotion was identied earlier and angry emotion
was identied later in comparison to the other ve emotions (all
P b 0.05).

M. Jovev et al. / Psychiatry Research 187 (2011) 234240

237

Fig. 1. Example of morphed angry face at 0, 20, 40, 60, 80 and 100% intervals.

Table 2 shows that the ESs were mostly in the small range (Cohen's
d 0.00.4), except for fear and disgust which were in the moderate
ES range (Cohen, 1992), thus suggesting that the BPD group was
somewhat slower at correctly identifying these two emotions.
3.2.2. Differences in earliest incorrect response Impulsivity
Since only 7 participants identied all 6 emotions incorrectly, it
was not appropriate to conduct repeated measures ANOVA on the
incorrect responses data. Instead, between-group comparisons were
conducted to examine differences between groups on earliest
incorrect response to specic emotions across the six emotions (see

Table 1
Demographic characteristics for the BPD and community groups.

Age (mean SD years)

Sex (N(%) female)
Accommodation (N (%))
Alone
Family/partner/friends
Other
Country of birth (N (%))
Australia/NZ
Employment (%)
Student
Employed
Unemployed
Education (%)
Tertiary
Secondary
Lower-secondary

BPD

Community

18.90 3.10
18 (86)

20.40 2.72
13 (65)

t = 1.64
2.38

0.12
0.12

0
17 (81)
4 (19)

1 (5)
81 (95)
0

4.01

0.14

20 (95)

13 (68)

4.97

0.02a

11 (53)
7 (33)
3 (14)

14 (74)
4 (21)
1 (5)

2.08

0.35

0
13 (62)
8 (38)

7 (37)
12 (63)
0

14.98

0.001b

Note: N = 19 in community group for all analysis except Sex due to missing data.
a
Signicant at alpha = 0.05 level.
b
Signicant at alpha = 0.01 level.

Table 2). There were no signicant differences (all P N 0.05) between

the groups. The ESs were mostly in the small range (Cohen's d 0.00.4),
except for happiness which was in the moderate ES range (Cohen,
1992), thus suggesting the BPD group might be impulsive (respond
earlier with errors) in identifying happiness than the community
control group.
3.3. Condence, threat, valence, arousal and dominance ratings
Planned between-group comparisons were conducted to examine
differences between BPD and control groups on self-rated levels of
condence, threat, subjective valence, arousal, and dominance/control,
in response to each facial expression (see Table 3). Although there were
no signicant group differences for any of these subjective responses (all
P N 0.05), moderate effect sizes were observed for subjective valance
ratings, with the BPD group rating disgust and happy emotions as
less positive compared to the community group (Cohen's d 0.5 to 0.6).
Moreover, the BPD group rated reduced feelings of control in response
to anger, disgust, happy and sad emotions compared to the community group (Cohen's d 0.5 to 0.6).
4. Discussion
The present study found no evidence of heightened sensitivity
(operationalized as the ability to detect emotions at lower thresholds
of expressivity) for emotional facial expressions in a sample of young
people with features of BPD. Both groups showed similar accuracy in
recognising emotional expressions. The ndings of the present study
are consistent with those of Domes et al. (2008) who reported an
emotion detection threshold in BPD participants comparable to the
non-clinical control group. Conversely, Lynch et al. (2006) reported a
generally heightened sensitivity for recognition of facial emotions in

238

M. Jovev et al. / Psychiatry Research 187 (2011) 234240

Table 2
Means, standard deviations and effect sizes for the BPD and community groups on earliest correct and incorrect response to specic emotions.
Earliest correct response
BPD

Anger
Disgust
Fear
Happiness
Sadness
Surprise

Earliest incorrect response

Community

BPD

Community

Mean

SD

Mean

SD

Cohen's d

Mean

SD

Mean

SD

Cohen's d

34.30
29.36
29.63
21.40
29.63
25.51

8.04
4.40
7.49
6.95
7.02
5.41

31.47
25.68
26.50
19.79
27.76
24.65

8.24
7.61
5.58
5.39
7.67
7.58

0.4
0.6
0.5
0.3
0.3
0.1

27.77
25.10
27.65
14.33
27.94
24.33

10.09
7.38
11.85
15.95
13.50
11.57

29.98
28.10
25.09
19.86
30.00
24.21

14.58
13.19
12.60
7.99
8.12
8.24

-0.2
0.3
0.2
-0.5
0.2
0

Bold moderate effect size.

BPD. Methodological differences between studies might explain these

discrepancies. In Lynch et al.'s (2006) study, participants could change
their responses as often as they liked during the course of each trial. In
both the present study and that of Domes et al. (2008), each trial was
stopped following the rst response and the recorded response could
not be altered. Therefore, factors other than emotional sensitivity,
such as the propensity to engage in subsequent guessing, might
account for these inconsistencies. The present study also suggests that
the BPD group rated disgust and happiness as less positive than the
community group (effect sizes range 0.5 to 0.6), which is consistent

Table 3
Difference between groups on ratings of condence, threat, valence, arousal and
dominance.
BPD

Community

Mean

SD

Mean

SD

Cohen's d

Condence
Anger
Disgust
Fear
Happiness
Sadness
Surprise

6.25
6.51
6.67
7.15
6.64
6.11

2.21
2.21
1.88
1.89
2.40
2.00

6.43
6.93
6.87
6.98
6.91
6.80

1.50
1.22
1.00
1.41
1.44
1.62

0.312
0.732
0.414
0.334
0.437
1.201

0.76
0.47
0.68
0.74
0.66
0.24

0.1
0.2
0.1
0.1
0.1
0.4

Threat
Anger
Disgust
Fear
Happiness
Sadness
Surprise

3.53
3.72
2.80
2.30
2.71
2.20

2.91
3.15
2.31
1.96
2.68
1.95

3.21
3.03
2.77
1.59
2.32
2.44

2.43
2.47
2.33
1.79
2.02
1.94

0.38
0.78
0.04
1.21
0.53
0.38

0.71
0.44
0.97
0.23
0.60
0.71

0.1
0.3
0.0
0.4
0.2
0.1

Valance
Anger
Disgust
Fear
Happiness
Sadness
Surprise

3.19
3.14
4.05
4.72
3.38
4.10

2.34
2.10
2.14
2.44
2.21
2.28

3.96
4.03
4.10
5.98
3.99
4.84

1.20
1.47
1.20
1.63
1.23
1.15

1.31
1.56
0.10
1.92
1.08
1.29

0.20
0.13
0.92
0.06
0.29
0.21

0.4
0.5
0.0
0.6
0.3
0.4

Arousal
Anger
Disgust
Fear
Happiness
Sadness
Surprise

2.48
2.70
2.71
2.33
1.99
2.25

2.89
2.68
2.66
2.37
2.18
2.27

2.53
2.36
2.56
2.66
2.62
2.66

2.38
2.44
2.45
2.54
2.51
2.30

0.07
0.42
0.19
0.43
0.86
0.58

0.95
0.68
0.85
0.67
0.40
0.57

0.0
0.1
0.1
0.1
0.3
0.2

Dominance/control
Anger
4.18
Disgust
4.10
Fear
5.01
Happiness
4.66
Sadness
4.63
Surprise
4.93

3.02
2.86
3.19
3.17
2.96
3.08

5.41
5.51
5.41
6.28
5.86
5.74

2.33
2.44
2.20
2.14
2.42
2.32

1.45
1.70
0.47
1.92
1.45
0.95

0.15
0.10
0.64
0.06
0.15
0.35

0.5
0.5
0.1
0.6
0.5
0.3

Bold moderate effect size.

with Meyer et al.'s (2004) study showing that BPD features in healthy
students are associated with negative bias in appraising social cues.
If BPD is characterised by an underlying emotional sensitivity
(Linehan, 1993), individuals with BPD features should demonstrate
correct recognition of emotions at lower threshold levels regardless of
emotional valence. However, the present sample of youth with BPD
features showed recognition response latencies similar to the
community group, with somewhat longer latencies for the recognition of fear and disgust (effect sizes of 0.5 to 0.6, respectively).
Although it appears that the participants in the control group are
somewhat older than the clinical sample, it is unlikely that this
inuenced the pattern of results as both groups are in the same age
range (15 to 24 years) and therefore in the similar stage of
development in the key brain areas (e.g., frontal lobe) that are at
least partially involved in recognising facial emotions (Adolphs et al.,
2003; Toga et al., 2006).
While these ndings appear to be inconsistent with the theoretical
prediction that emotion dysregulation in BPD is associated with
emotional sensitivity (Linehan, 1993), the ndings from this sample
of young individuals with BPD features might also be taken to suggest
that emotional sensitivity is present only in severe BPD or develops
later in the course of the disorder, perhaps through continuing
exposure to traumatic life events and recurrent mental state disorders
(Jovev and Jackson, 2006). Recurrent or chronic Axis I disorders
(notably mood and substance use disorders) and cumulative
traumatic life events might lead to changes in cognition, emotion,
behaviour or biology (Lewinsohn et al., 1981; Shea et al., 1996; Ormel
et al., 2004; Beevers et al., 2007) and render individuals vulnerable to
developing cognitive biases, such as increased emotional sensitivity,
later in the course of the disorder.
The BPD sample in this study was specically chosen to minimise
duration of illness factors and is different in several regards to the
late-stage BPD syndrome described in the DSM-IV-TR (APA, 2000)
and typically seen in adult mental health settings. Participants in this
study were youth with both sub-syndromal and full-syndrome BPD
who were, on average, younger than participants in comparable
studies of adult BPD. Nevertheless, diagnosing adolescent personality
pathology remains controversial (Chanen and McCutcheon, 2008) but
has been shown to be as reliable and valid in adolescence as it is in
adulthood (Chanen et al., 2008a; Miller et al., 2008). Moreover, it is
widely acknowledged that personality disorders are best conceptualised as dimensional constructs (Clark, 2007) and that BPD exists on a
continuum of clinical severity, with no distinct boundary between
sub-syndromal and threshold BPD (Johansen et al., 2004).
One limitation of the present study relates to the specicity of our
ndings to BPD. Comorbidity is typical of BPD samples at any age
(Chanen et al., 2007) and future studies should employ a clinical
comparison group. Similar to Domes et al. (2008), there were high
rates of PTSD (48%) in the present sample and although the healthy
comparison group were screened for the absence of PTSD, we did not
assess them for histories of childhood abuse or neglect. An exploratory
three-group analysis (BPD vs. BPD + PTSD vs. community control)

M. Jovev et al. / Psychiatry Research 187 (2011) 234240

indicated that the BPD group with no PTSD diagnosis was slower at
recognising happiness than the other two groups (ES = 0.5). Larger
studies that can include comparison groups of individuals with
comorbid PTSD, depression, other personality disorders, and/or no
other personality pathology would afford the ability to disentangle
the effects of Axis I anxiety and depressive disorders as opposed to
personality pathology on emotion sensitivity.
Another limitation of this study related to the high level of
education in the healthy control group, which might limit generalizability. Although every effort was made to advertise study
participation in the wider community, the geographical area that is
covered by the clinical service incorporates several universities that
led to the high number of tertiary-educated individuals volunteering
to participate in the study. Future research should strive to better
match the participants in this domain.
It is also of note that females perform affect recognition tasks
better than males (McClure, 2000; Thayer and Johnsen, 2000) and the
clinical group in the present study had somewhat more females than
the community group (although not signicant due to the small
sample size). It is therefore possible that the mixing of sexes obscured
group differences; however, repeating the analyses only for the
female participants did not alter the general pattern of the results.
Females with BPD features showed recognition response latencies
similar to the community females, with somewhat longer latencies for
the recognition of fear and disgust and effect sizes similar to the entire
sample (0.4 to 0.6, respectively). Moreover, if BPD is characterised by
an underlying emotional sensitivity (Linehan, 1993) and females are
more sensitive to affect than males, the clinical group in the present
study (with more females and more BPD features) should demonstrate shorter latencies for correctly recognised emotions. The
ndings of the present study suggest similar response latencies
between the two groups, and where the differences might be present
(fear and disgust) these are in the opposite direction to the pattern
expected for emotional sensitivity.
Moreover, there are some general limitations that need to be
acknowledged when using both static and dynamic affect recognition
paradigms. Most of these studies use faces unfamiliar to the
participants, and do not examine social interactions or social context
effects. The exploration of these factors in experimental settings may
inform issues regarding therapeutic alliance and psychosocial functioning in everyday social situations.
Despite these limitations, this is the rst study to explore emotional
sensitivity using dynamic facial expressions in youth early in the
course of BPD. This allowed us to minimise duration of illness effects,
although less severe levels of BPD pathology might have also resulted
in moderate effect sizes observed in this study. Although our sample
size is relatively small to detect the moderate effect sizes, it is comparable to the sample sizes used in the studies by Lynch et al. (2006)
and Domes et al. (2008). Further research is needed to replicate these
ndings in a larger sample. In addition, we used validated diagnostic
measures to determining study eligibility, thus representing more
stringent inclusion criteria than previous studies of BPD.
In summary, outpatient youth with BPD features did not show
evidence of heightened sensitivity to emotional facial expressions. They
were accurate in recognising emotional expressions. These ndings
suggest that emotional sensitivity might be a core feature of severe BPD
only, or it might develop later in the course of the disorder. Further
research is needed to examine emotional reactivity and sensitivity
across various phases of the disorder in larger clinical samples.

Acknowledgements
The face morph task was modied for use by Dr Tim Bates at Macquarie Centre for
Cognitive Science, Macquarie University, Sydney, Australia.
This project was supported by a 2007 Early Career Grant awarded by the University
of Melbourne to the rst author.

239

References
Adolphs, R., 2002. Neural systems for recognizing emotion. Current Opinion in
Neurobiology 12, 169177.
Adolphs, R., Tranel, D., Damasio, A.R., 2003. Dissociable neural systems for recognizing
emotions. Brain and Cognition 52, 6169.
APA, 2000. Diagnostic and statistical manual of mental disorders, 4th TR (text revision)
ed. American Psychiatric Association, Washington, DC.
Beevers, C.G., Rohde, P., Stice, E., Nolen-Hoeksema, S., 2007. Recovery from major
depressive disorder among female adolescents: a prospective test of the scar
hypothesis. Journal of Consulting and Clinical Psychology 75, 888900.
Blair, R.J., Colledge, E., Murray, L., Mitchell, D.G., 2001. A selective impairment in the
processing of sad and fearful expressions in children with psychopathic tendencies.
Journal of Abnormal Child Psychology 29, 491498.
Bland, A.R., Williams, C.A., Scharer, K., Manning, S., 2004. Emotion processing in
borderline personality disorder. Issues in Mental Health Nursing 25, 655672.
Chanen, A.M., McCutcheon, L.K., 2008. Personality disorder in adolescence: the
diagnosis that dare not speak its name. Personal Mental Health 2, 3541.
Chanen, A.M., Jovev, M., Jackson, H., 2007. Adaptive functioning and psychiatric
symptoms in adolescents with borderline personality disorder. The Journal of
Clinical Psychiatry 68, 297306.
Chanen, A.M., Jovev, M., McCutcheon, L.K., Jackson, H.J., McGorry, P.D., 2008a.
Borderline personality disorder in young people and the prospects for prevention
and early intervention. Current Psychiatry Reviews 4, 4857.
Chanen, A.M., Velakoulis, D., Carison, K., Gaunson, K., Wood, S.J., Ycel, M., Yuen, H.P.,
Jackson, H.J., McGorry, P.D., Pantelis, C., 2008b. Orbitofrontal, amygdala and
hippocampal volumes in teenagers with rst-presentation borderline personality
disorder. Psychiatry Research Neuroimaging 163 (2), 116125.
Chanen, A.M., McCutcheon, L.K., Germano, D., Nistico, H., Jackson, H., McGorry, P.D.,
2009. The HYPE clinic: an early intervention service for borderline personality
disorder. Journal of Psychiatric Practice 15 (3), 163172.
Clark, L.A., 2007. Assessment and diagnosis of personality disorder: Perennial issues
and an emerging reconceptualization. Annual Review of Psychology 58, 227257.
Cohen, J., 1992. A power primer. Psychological Bulletin 112 (1), 155159.
Crowell, S.E., Beauchaine, T.P., Linehan, M.M., 2009. A biosocial developmental model of
borderline personality: elaborating and extending Linehan's theory. Psychological
Bulletin 135 (3), 495510.
Domes, G., Czieschnek, D., Weidler, F., Berger, C., Fast, K., Herpertz, S.C., 2008.
Recognition of facial affect in borderline personality disorder. Journal of Personality
Disorders 22 (2), 135147.
Domes, G., Schulze, L., Herpertz, S.C., 2009. Emotion recognition in borderline personality
disorder a review of the literature. Journal of Personality Disorders 23 (1), 619.
Ekman, P., Friesen, W., 1976. The Pictures of Facial Affect. Consulting Psychologists
Press, Palo Alto, CA.
Fertuck, E.A., Jekal, A., Song, I., Wyman, B., Morris, M.C., Wilson, S.T., Brodsky, B.S.,
Stanley, B., 2009. Enhanced reading the mind in the eyes in borderline personality
disorder compared to healthy controls. Psychological Medicine 39 (12), 19791988.
First, M.B., Spitzer, R.L., Gibbon, M., Williams, J.B., 1996. Structured Clinical Interview for
DSM-IV Axis I Disorders, Research Version, Non-patient Edition. (SCID-I/NP).
Biometrics Research, New York State Psychiatric Institute, New York.
First, M.B., Gibbon, M., Spitzer, R.L., Williams, J.B.W., Benjamin, L.S., 1997a. User's Guide
for the Structured Clinical Interview for DSM-IV Axis II Personality Disorders.
American Psychiatric Press, Washington, DC.
First, M.B., Spitzer, R.L., Gibbon, M., Williams, J.B., 1997b. Structured Clinical Interview
for DSM-IV Axis I Disorders, Research Version, Patient Edition, (SCID-I/P). American
Psychiatric Press, Inc., Washington, D.C.
Fonagy, P., Target, M., Gergely, G., 2000. Attachment and borderline personality
disorder: a theory and some evidence. Psychiatric Clinics of North America. Special
Issue: Borderline Personality Disorder 23 (1), 103122.
Frank, H., Hoffman, N., 1986. Borderline empathy: an empirical investigation.
Comprehensive Psychiatry 27 (4), 387395.
Geday, J., Kupers, R., Gjedde, A., 2007. As time goes by: temporal constraints on emotion
activation of inferior medial prefrontal cortex. Cerebral Cortex 17, 27532759.
Herpertz, S.C., Kunert, H.J., Schwenger, U.B., Sass, H., 1999. Affective responsiveness in
borderline personality disorder: a psychophysiological approach. The American
Journal of Psychiatry 156 (10), 15501556.
Herpertz, S.C., Schwenger, U.B., Kunert, H.J., Lukas, G., Gretzer, U., Nutzmann, J.,
Schuerkens, A., Sass, H., 2000. Emotional responses in patients with borderline as
compared with avoidant personality disorder. Journal of Personality Disorders 14
(4), 339351.
Herpertz, S.C., Wenning, B., Mueller, B., Qunaibi, M., Sass, H., Herpertz-Dahlmann, B.,
2001. Psychophysiological responses in ADHD boys with and without conduct
disorder: implications for adult antisocial behavior. Journal of the American
Academy of Child and Adolescent Psychiatry 40 (10), 12221230.
Johansen, M., Karterud, S., Pedersen, G., Gude, T., Falkum, E., 2004. An investigation of
the prototype validity of the borderline DSM-IV construct. Acta Psychiatrica
Scandinavica 109 (4), 289298.
Jovev, M., Jackson, H.J., 2006. The relationship of borderline personality disorder, life
events and functioning in an Australian psychiatric sample. Journal of Personality
Disorders 20 (3), 205217.
Judd, P.H., McGlashan, T.H., 2003. A developmental model of BPD: understanding
variations in course and outcome. American Psychiatric Press, Arlington, VA.
Keltner, D., Kring, A.M., 1998. Emotion, social function, and psychopathology. Review of
General Psychology 2, 320342.
Kernberg, O.F., 1967. Borderline personality organization. Journal of the American
Psychoanalytic Association 15, 641685.

240

M. Jovev et al. / Psychiatry Research 187 (2011) 234240

Kernberg, O.F., 1975. Borderline Conditions and Pathological Narcissism. Aronson, New
York.
Kernberg, O.F., 1976. Object-relations theory and clinical psychoanalysis. Aronson, New York.
Korne, L., Hooley, J.M., 2000. Directed forgetting of emotional stimuli in borderline
personality disorder. Journal of Abnormal Psychology 109 (2), 214221.
LaBar, K.S., Crupain, M.J., Voyvodic, J.T., McCarthy, G., 2003. Dynamic perception of facial
affect and identity in the human brain. Cerebral Cortex 13, 10231033.
Ladisch, W., Feil, W.B., 1988. Empathy in psychiatric patients. The British Journal of
Medical Psychology 61, 155162.
Levine, D., Marziali, E., Hood, J., 1997. Emotion processing in borderline personality
disorders. The Journal of Nervous and Mental Disease 185 (4), 240246.
Lewinsohn, P.M., Steinmetz, J.L., Larson, D.W., Franklin, J., 1981. Depression-related
cognitions: antecedent or consequence? Journal of Abnormal Psychology 90,
213219.
Linehan, M.M., 1993. CognitiveBehavioral Treatment of Borderline Personality
Disorder. Guliford Press, New York.
Lynch, T.R., Rosenthal, Z.M., Kosson, D.S., Cheavens, J.S., Lejuez, C.W., Blair, R.J.R., 2006.
Heightened sensitivity to facial expressions of emotion in borderline personality
disorder. Emotion 6 (4), 647655.
McClure, E.B., 2000. A meta-analytic review of sex differences in facial expression
processing and their development in infants, children, and adolescents. Psychological Bulletin 126, 424453.
Meyer, B.R., Pilkonis, P.A., Beevers, C.G., 2004. What's in a (neutral) face? Personality
disorders, attachment styles, and the appraisal of ambiguous social cues. Journal of
Personality Disorders 18 (4), 320336.
Miller, A.L., Muehlenkamp, J.J., Jacobson, C.M., 2008. Fact or ction: diagnosing
borderline personality disorder in adolescents. Clinical Psychology Review 28
(6), 969981.
Minzenberg, M., Poole, J.H., Vinogradov, S., 2006. Social-emotion recognition in
borderline personality disorder. Comprehensive Psychiatry 47, 468474.
Ormel, J., Oldehinkel, A.J., Nolen, W.A., Vollebergh, W., 2004. Psychosocial disability
before, during, and after a major depressive episode: a 3-wave population-based
study of state, scar, and trait effects. Archives of General Psychiatry 61, 387392.
Pilkonis, P.A., Heape, C.L., Ruddy, J., Serrao, P., 1991. Validity in the diagnosis of
personality disorders: The use of the LEAD standard. Psychological Assessment 3
(1), 4654.

Renneberg, B., Heyn, K., Gebhard, R., Bachmann, S., 2005. Facial expression of emotions
in borderline personality disorder and depression. Journal of Behavior Therapy and
Experimental Psychiatry 36 (3), 183196.
Sato, W., Kochiyama, T., Yoshikawa, S., Naito, E., Matsumura, M., 2004. Enhanced neural
activity in response to dynamic facial expressions of emotion: an fMRI study.
Cognitive Brain Research 20, 8191.
Shea, M.T., Leon, A.C., Mueller, T.I., Solomon, D.A., Warshaw, M.G., Keller, M.B., 1996.
Does major depression result in lasting personality change? The American Journal
of Psychiatry 153, 14041410.
Thayer, J.F., Johnsen, B.H., 2000. Sex differences in judgement of facial affect: a
multivariate analysis of recognition errors. Scandinavian Journal of Psychology 41,
243246.
Toga, A.W., Thompson, P.M., Sowell, E.R., 2006. Mapping brain maturation. Trends in
Neuroscience 29, 148159.
Veen, G., Arntz, A., 2000. Multidimensional dichotomous thinking characterizes
borderline personality disorder. Cognitive Therapy and Research 24 (1), 2345.
Wagner, A.W., Linehan, M.M., 1999. Facial expression recognition ability among women
with borderline personality disorder: implications for emotion regulation. Journal
of Personality Disorders 13 (4), 329344.
Zanarini, M.C., Frankenburg, F.R., Sickel, A.E., Yong, L., 1996. The Diagnostic Interview for
DSM-IV Personality Disorders (DIPD-IV). McLean Hospital, Belmont, Massachusetts.
Zanarini, M.C., Vujanovic, A.A., Parachini, E.A., Boulanger, J.L., Frankenburg, F.R., Hennen, J.,
2003. A screening measure for BPD: the McLean Screening Instrument for Borderline
Personality Disorder (MSI-BPD). Journal of Personality Disorders 17 (6), 568573.
Zanarini, M.C., Frankenburg, F.R., Hennen, J., Reich, D.B., Silk, K.R., 2004a. Axis I
comorbidity in patients with borderline personality disorder: 6-year follow-up and
prediction of time to remission. The American Journal of Psychiatry 161 (11),
21082114.
Zanarini, M.C., Frankenburg, F.R., Hennen, J., Silk, K.R., 2004b. Mental health service
utilization by borderline personality disorder patients and Axis II comparison
subjects followed prospectively for 6 years. The Journal of Clinical Psychiatry 65
(1), 2836.
Zanarini, M.C., Frankenburg, F.R., Reich, D.B., Hennen, J., Silk, K.R., 2005. Adult
experiences of abuse reported by borderline patients and Axis II comparison
subjects over six years of prospective follow-up. The Journal of Nervous and Mental
Disease 193 (6), 412416.