424

British Journal of Clinical Psychology (2013), 52, 424–437

© 2013 The British Psychological Society
www.wileyonlinelibrary.com

Attachment: Developmental pathways to affective

dysregulation in young people at ultra-high risk of
developing psychosis
Ruchika Gajwani1*, Paul Patterson1,2 and Max Birchwood1,2
1
School of Psychology, University of Birmingham, UK
2
YouthSpace Mental Health Service, Birmingham and Solihull Mental Health
Foundation Trust, Birmingham, UK

Objective. Embedded in attachment theory is its association with affect regulation,

which provides a framework for affective dysregulation in the emerging psychosis.
Method. Fifty-one participants meeting criteria for ultra-high risk (UHR) of developing
psychosis were recruited from a youth mental health service within the United Kingdom.
At intake baseline, prior to starting therapeutic intervention, all clients were assessed on
measures of affective dysregulation and attachment.
Results. A large proportion of our sample (N = 51) reported clinically significant levels
of depression (78%), state anxiety (59%), and social anxiety (65%). Eighty per cent of the
UHR sample was insecurely attached. Insecure attachment was significantly associated
with elevated depression and social anxiety. Attachment styles were associated with
anxiety, depression, and social anxiety. There was no support for a mediating role of social
anxiety between attachment styles and depression.
Conclusion. Clinically significant levels of distress and anxiety experienced by the young
people at high risk of psychosis. Clinical implications for the treatment of affective
dysregulation in young people at UHR in relation to their attachment styles have been
discussed.

Practitioner points
 The treatment of affective dysregulation has implications on social integration, crucial to the recovery
of young people in the emerging psychosis.
 Assessment of adult attachment difficulties may bring to the forefront specific negative internal
working models (and interpersonal schemas) which may indicate vulnerability to social anxiety and
depression.
 Young adults at ultra-high risk of psychosis with emotional difficulties may be better at help seeking and
therefore over-represented in this sample.
 Cross-sectional data analysis makes it difficult to draw aetiological links between adult attachment and
affective dysregulation.

The importance of studying infant–mother relationships was crystallized by Freud in the

psychoanalytic theory, as being the ‘prototype for all later love relations’ (Fonagy, 1999).

*Correspondence should be addressed to Dr Ruchika Gajwani, School of Psychology, University of Birmingham, Edgbaston,
Birmingham B15 2TT, UK (email: [email protected]).

DOI:10.1111/bjc.12027
 Attachment, affective dysregulation and psychosis 425

Bowlby conceptualized attachment as a human tendency to form affective bonds with

particular others, which in the case of separation and loss may lead to emotional
disturbances such as depression, separation anxiety, anger, and emotional withdrawal
(Bowlby, 1979). The nature of this affective bond and its emotional quality are crucial for
the stability of attachment formation (Carlson & Sroufe, 1995). Bowlby’s work with
infants who experienced loss or separation from caregiver reflected a range of observed
emotions and highlighted the role of affective investment in an attachment relationship
(Ainsworth, Blehar, Waters, & Wall, 1978). Attachment exists for the purpose of
protection and is intrinsic to care-giving and care-eliciting, all of which are highly relevant
to later onset of affective disorders.
One of the mechanisms linking early attachment and subsequent psychosocial
functioning are the enduring internal working models, characterized as internalized
personal representations (Bowlby, 1979), providing implicit decision rules for under-
standing self and others (Thompson, 1999). The infant at start has limited capability for
self-management and self-regulation; therefore, the role of the caregiver responsiveness
and availability is primary. Response to the child’s biological activity develops a regularity
pattern, forming a chained interaction sequence. Through development and shared
experiences, a dyadic relationship develops which forms the basis for attachment styles.
The increase in affective expression marks the coordination of emotion and cognition.
Cognitive development and affective regulation helps the child to, gradually, internally
represent past experiences, described here as internal working models (Carlson & Sroufe,
1995). Drawing on Bowlby’s formulation of the internal working model and subsequent
advancements in attachment theory, Bartholomew (1990, 1997) proposed a four-category
model of adult attachment. This model describes four distinctive attachment styles
(secure, preoccupied, dismissive, and fearful)1 based on the interaction of two
dimensions of the internal working model- ‘model-of-self’ and ‘model-of-other’.
The role of attachment strategies for the association between cognitive factors and
emotional regulation (Mikulincer, Shaver, & Pereg, 2003) helps expand the existing
cognitive models of psychosis (Garety, Kuipers, Fowler, Freeman, & Bebbington, 2001).
The unavailability of a secure figure or an overbearing attachment can lead to insecure
attachment patterns, which under stress may prompt defensive emotional reaction of
hyperactivity or emotional distancing. The influence of affect on cognition however is not
straightforward. Mikulincer et al. (2003) suggest that attachment strategies play a
moderating role in the link between emotional reaction and cognitive responses. Secure
attachment strategies would incline a person to constructively cope with negative affect
and to explore ingeniously the potential of one self and the environment under positive
affect. Those individuals who rate high on attachment anxiety are hypervigilant of
attachment figure unavailability, with intensified emotional reaction to pressure. In
contrast, avoidant attachment strategies are used to emotionally detach oneself from
positive or negative conditions by processing information cognitively but not emotionally
(Mikulincer et al., 2003). Adults may avoid attachment either because of fear of intimacy
and rejection or because of lack of interest in social interaction (Bartholomew, 1990).
Specific to psychotic experiences in clinical and non-clinical samples, stronger
attachment avoidance is related to treatment resistance (Dozier, 1990), emotional

1
Secure attachment style represents positive self and other cognitions with low attachment avoidance and anxiety/dependence.
Preoccupied and fearful attachment styles represent elevated dependency and a negative model-model-of-self; but differ in their
model-of-other. Fearful have high avoidance and a negative world view, whereas preoccupied are dependent and have a positive
world view. Dismissive attachment style avoid intimacy, have a positive self-worth and a negative world view.
426 Ruchika Gajwani et al.

regulation difficulties (Owens, Haddock, & Berry, 2012), dissatisfaction with quality of life
(Couture, Lecomte, & Leclerc, 2007) and negative schizotypal feature of social anhedonia
(Berry, Wearden, Barrowclough, & Liversidge, 2006). Also, embedded in attachment
theory is its association with emotional functioning (Bowlby, 1982), which provides a
developmental framework for affective dysregulation in psychosis (Berry, Barrowclough,
& Wearden, 2007). The evidence supporting the association between insecure attach-
ment and affective disorders is well documented (Bowlby, 1973, 1980; Cummings &
Cicchetti, 1990; Roberts, Gotlib, & Kassel, 1996), and therefore crucial for this research as
affective dysregulation in the emerging psychosis is a vital predictive factor for transition
to psychosis (Johnstone, Ebmeier, Miller, Owens, & Lawrie, 2005; Owens, Miller, Lawrie,
& Johnstone, 2005; Yung et al., 2003).
Childhood traumatic experiences and anomalous attachment may increase vulnera-
bility to negative interpersonal schemas, which in turn may partly account for
psychotic-like experiences (Fisher, Appiah-Kusi, & Grant, 2012), emotional response to
psychosis (Birchwood, 2003; Birchwood et al., 2004; Read & Gumley, 2008), and
recovery after psychosis (Gumley & MacBeth, 2006). Predictions from attachment to
psychopathology get more complex as attachment styles are more durable in low-risk
samples and more variable in high-risk samples (Van IJzendoorn & Bakermans-Kranen-
burg, 1997) and in this respect the association between attachment and the emerging
psychosis should be deliberated under the larger framework of developmental risk and
protective factors (Sroufe, Carlson, Levy, & Egeland, 1999).
The study endeavours to describe the composition of attachment styles in the
ultra-high-risk (UHR) sample, and contrast these with the standardized scores from
normative data. We hypothesized that (1) individuals at UHR of psychosis with an insecure
attachment style (preoccupied, dissociated, and fearful) will report significantly higher
levels of depression, state anxiety, and social anxiety, compared to UHR individuals with a
secure attachment style, and (2) social anxiety would mediate the relationship between
attachment styles and depression in the UHR sample, based on previous findings by Eng,
Heimberg, Hart, Schneier, and Liebowitz (2001).

Method
Sample
Fifty-one participants meeting the UHR criteria for developing psychosis (McGlashan
et al., 2002), between 16 and 35 years, were recruited from a youth mental health service
based within an Early Intervention Team for psychosis in the United Kingdom. Any
participant that had made transition to psychosis or did not meet UHR criteria at the time
of screening assessment was excluded from the study. The sample comprised 33 males
and 18 females, with a mean age of 19 (SD = 3.09, minimum = 16, maximum = 30). The
ethnic breakdown of the sample was 57% White British, 31% Asian, 4% Black/Black British
Caribbean, 2% Black/Black British African and 6% were of Other ethnic background. The
majority of the sample (51%) was in education at the time of assessment and 65% of the
sample had a General Certificate of Secondary Education (GCSE) as a minimum
educational qualification.

Design
Clients meeting UHR criteria were contacted individually to arrange appointments at their
GP surgery, community mental health service or at their accommodation, depending on
 Attachment, affective dysregulation and psychosis 427

their suitability and risk assessment. Clients meeting the screening criteria for UHR of
psychosis were offered clinical intervention and voluntary participation in research
conducted by the team. Consented clients completed a baseline assessment pack which
included measures outlined below within 2 weeks of acceptance to the youth mental
health team, prior to any therapeutic intervention.

Assessments
Ultra-high risk of psychosis
To assess intake criteria for prodromal symptoms of psychosis, referrals were assessed
using the Structured Interview for Prodromal Syndromes (SIPS)2 (McGlashan et al., 2002).
The SIPS is a structured diagnostic interview consisting of four measures: (1) Scale of
Prodromal Symptoms (SOPS), a 19-item scale assessing positive, negative, disorganization,
and general prodromal symptoms, (2) DSM-IV schizotypal personality disorder checklist,
(3) questionnaire for family history of mental illness, and (4) Global Assessment of
Functioning scale. The criteria for prodromal syndrome evaluated using the above
measures are given in the following: (1) brief intermittent psychotic symptom syndrome,
(2) attenuated positive symptom syndrome, and (3) genetic risk and deterioration
syndrome. The diagnostic interview has shown excellent inter-rater reliability and
predictive validity (Miller et al., 2002).

Adult attachment
The Revised Adult Attachment Scale (RAAS; Collins, 1996) is a modified version of the
Adult Attachment Scale (Collins & Read, 1990) and is used to establish attachment styles in
adulthood. The RAAS is an 18-item self-report questionnaire with three subscales (each
composed of six items): (1) comfort with closeness, (2) comfort depending on others, and
(3) anxiety in relationships. Each response is rated on a 5-point scale (1 = not at all
characteristic of me to 5 = very characteristic of me). A higher score on the subscales
represents greater comfort with closeness, greater confidence in dependability of others
and higher anxiety associated with relationships. On the basis of the three RAAS subscales,
adult attachment can be assessed using a continuous measure (extent of comfort with
closeness, comfort depending on others, and fear of rejection) or can be categorized into
Bartholomew’s (1990) four attachment styles (secure, preoccupied, fearful, and
dismissing). The RAAS has good reliability for the three attachment dimensions of Close,
Depend and Anxiety (a = .77, .78, and .85, respectively) and its validity is reaffirmed by its
strong correlation (r = .98) with the original Adult Attachment Scale (Collins & Read,
1990). Within psychosis, the scale has demonstrated excellent internal consistency;
Closeness a = .86, Dependence a = .86, and Anxiety a = .97 (Tait, Birchwood, &
Trower, 2004). Tait et al. (2004) also showed moderate correlations (significant)
between the Parental Bonding Instrument (Parker et al., 1997) and the RAAS.

Anxiety
The Beck Anxiety Inventory (BAI; Beck, Epstein, Brown, & Steer, 1988) is a 21-item
self-report inventory, with each response rated on a 4-point scale ranging from 0 (not at

2
Prior to screening and recruitment of UHR participants, the primary researcher (RG) underwent training to adequate levels of
reliability on the Structured Interview for Prodromal Syndrome.
428 Ruchika Gajwani et al.

all) to 3 (severe-it bothered me a lot), based on symptoms of anxiety in the previous week.
It is a widely used measure of anxiety among psychiatric samples, with a high internal
consistency (a = .92) and good test–retest reliability (r = .75). BAI also has good
concurrent validity as it correlates well with other anxiety measures (Beck et al., 1988).

Social anxiety
The Social Interaction Anxiety Scale (SIAS) and Social Phobia Scale (SPS) were both
developed by Mattick and Clarke (1998) as companion measures to assess social anxiety.

Social Interaction Anxiety Scale. The SIAS (Mattick & Clarke, 1998) is a 20-item
self-report measure, assessing general social interaction anxiety. The SIAS assesses fear of
general social anxiety through the items describing apprehension in different group
interactions, such as difficulty talking to someone of the opposite sex, worries about
expressing one’s feelings, or nervousness mixing with people one doesn’t know well.
Total scores range from 0 to 80 with a higher score indicative of greater social interaction
anxiety. With a cut-off score of 36, SIAS has demonstrated discriminate validity is assessing
social anxiety compared to other anxiety disorders, with a sensitivity of 0.93, specificity of
0.66 and a positive predictive value of 0.84 (Peters, 2000).

Social Phobia Scale. The SPS (Mattick & Clarke, 1998) is a 20-item self-report measure,
assessing performance related anxiety and a fear of scrutiny. The SPS measures fear of
scrutiny with the items describing different performance-related anxieties, such as
uneasiness eating, drinking, writing, using public toilets, travelling on public transport, or
being looked at. With a cut off score of 26, SPS has shown high levels of internal
consistency (a = .89–.94) and test-retest reliability (r = .93) (Mattick & Clarke, 1998;
Peters, 2000).

Depression
The Beck Depression Inventory (BDI; Beck, Rush, Shaw, & Emery, 1979) is a 21-item
self-report measure used to assess depression severity in the previous week including the
day of assessment. It is a widely used measure for depression prevalence and intensity in
different psychiatric and non-psychiatric samples. The BDI has demonstrated good
concurrent validity, shown in high mean correlations with other measures of depression
and has good discriminant validity, reliably differentiating depressed from the
non-depressed individuals (Richter, Werner, Heerlein, Kraus, & Sauer, 1998).

Statistical analysis
Analyses for this study were performed using SPSS v.18 (SPSS Inc., Chicago, IL, USA) for
Windows. The normality of distribution of all measures was assessed using the
Kolmogorov–Smirnov test. Descriptive analysis and ANOVA were used for preliminary
analysis of the data and to investigate the association between attachment styles with
depression, anxiety, and social anxiety. To test the mediating relationship between
attachment styles (i.e., predictor variable) and depression with social anxiety as the
mediating variable, multiple regression analysis was conducted.
 Attachment, affective dysregulation and psychosis 429

Results
Forty-five participants met UHR criteria for attenuated positive symptom syndrome, four
for genetic risk and deterioration syndrome, and two for brief intermittent psychotic
symptom syndrome. Seventy-eight per cent of the sample reported moderate to severe
levels of depression (mean = 28.50, SD = 12.34) at intake baseline, 59% reported
moderate to severe levels of anxiety state (mean = 22.65, SD = 12.83), 65% met the cut
off for social interaction anxiety (mean = 39.29, SD = 17.17), and 74% reported social
performance anxiety (mean = 33.87, SD = 15.44). Mean scores (SD) on RAAS for the
Close, Depend, and Anxiety subscales were 17.93 (5.54), 16.05 (4.87), and 19.58 (6.42)
respectively. The distribution of the four attachment styles in the UHR sample were
compared to published findings in a community sample of young adults. Eighty per cent of
the UHR sample were insecurely attached [i.e., fearful (43%), preoccupied (20%),
dismissive (17%)] in comparison to 37% reported in the community sample by Stein et al.
(2002). There were no age and gender differences in attachment styles.

Hypothesis 1: Adult attachment and relationship with depression and anxiety

Overall difference between secure and insecure attachment was significant for depression
(F(1,50) = 12.32, p < .001), social anxiety (F(1,50) = 9.43, p < .01), social phobia
(F(1,50) = 8.10, p < .01), and non-significant for state anxiety (F(1,50) = 0.78, p = .38).
Results of the ANOVA show that there were significant differences between attachment
styles and level of reported depression (F(3,48) = 9.70, p < .001), social phobia
(F(3,48) = 6.03, p < .01), social anxiety (F(3,48) = 8.47, p < .001), and state anxiety
(F(3,48) = 4.47, p < .01). Means (SD) and Bonferroni post-hoc comparisons for the four
attachment styles are presented in Table 1. Depression and social interaction anxiety
were significantly lower in UHR individuals with secure attachment compared to those
with fearful or preoccupied attachment styles. Participants with secure or dismissive

Table 1. ANOVAs and post-hoc comparisons with attachment styles as the grouping variable and
depression and social anxiety symptoms as dependent variables

Secure Insecure attachment style

attachment Post-hoc
Symptoms style Preoccupied Dismissive Fearful (Bonferroni)

Social anxiety
SIAS 23.00 (16.24) 42.10 (12.88) 31.56 (17.12) 48.59 (12.65) Secure < preoccupied*,
fearful***
SPS 20.75 (11.85) 35.89 (14.16) 25.00 (16.58) 41.38 (12.21) Fearful > secure**,
dismissive*
Depression
BDI 15.70 (9.20) 38.56 (8.71) 24.89 (9.60) 31.68 (10.83) Secure < preoccupied***,
fearful***
Anxiety
BAI 17.50 (10.31) 33.00 (12.31) 15.33 (10.36) 23.27 (12.40) Preoccupied > secure*,
dismissive*

Note. SIAS = Social Interaction Anxiety Scale; SPS = Social Phobia Scale; BDI = Beck Depression
Inventory; BAI = Beck Anxiety Inventory. *p < .05; **p < .01; ***p < .001.
430 Ruchika Gajwani et al.

attachment style rated themselves with significantly lower social performance anxiety
than those with fearful attachment style. State anxiety was significantly lower in secure
and dismissive attachment styles than those with preoccupied attachment style.

Hypothesis 2: Mediational analysis

For the purpose of this analysis, three categories of attachment styles (fearful,
preoccupied, and dismissive) were combined into one category of insecure attachment.
Three steps of regression analysis were carried out to test the mediating relationship, as
recommended by Baron and Kenny (1986), shown in Table 2. In step 1, depression was
regressed onto the attachment styles (secure = 1, insecure = 2), which showed a
significant relationship between the two variables (F(1,50) = 18.17, p < .001, R2 = .28).
Step 2 found that attachment style (predictor variable) was significantly associated with
social anxiety (F(1,50) = 14.14, p < .001, R2 = .22) and social phobia (F(1,50) = 24.08,
p < .01, R2 = .15). In Step 3, depression was regressed onto social anxiety, social phobia,
and attachment styles. Although a significant model was produced (F(3,48) = 8.54,
p < .001, R2 = .39), social phobia was a stronger predictor of depression than attachment
and social anxiety (Table 3). These results suggest that social phobia mediates the
relationship between adult attachment and depression in the UHR sample.
For the purpose of a cross-sectional analysis, a model with depression mediating the
relationship between adult attachment and social anxiety was tested. Step 1 showed that
adult attachment significantly predicted depression, and in Step 2 adult attachment
significantly predicted social anxiety and social phobia (Table 3). In Step 3, social anxiety
was regressed onto depression and adult attachment which produced a significant model
(F(2,49) = 14.66, p < .001, R2 = .38). As shown in Table 4, the relationship between adult
attachment and social anxiety was no longer significant after depression was put into the
multiple regression equation. This suggests that the significant relationship between adult
attachment and social anxiety is mediated by depression in the UHR sample.

Table 2. Regression analysis with attachment style as a predictor variable and social anxiety or
depression as the outcome variable

Outcome variable R2 B SEB b

Beck Depression Inventory .28 16 3.75 .52**

Social Anxiety Inventory .22 20.27 5.38 .47**
Social Phobia Scale .15 15.88 5.58 .39*

Note. *p < .01; **p < .001.

Table 3. Mediational analysis with depression as the outcome variable

Predictor variable B SEB b

Revised Adult Attachment Scale 8.27 4.13 .28†

Social Anxiety Inventory 0.10 0.24 .14
Social Phobia Scale 0.26 0.12 .34*

Note. *p < .01; †p = .052.

 Attachment, affective dysregulation and psychosis 431

Table 4. Mediational analysis with social anxiety as the outcome variable

Predictor variable B SEB b

Revised Adult Attachment Scale 9.88 5.77 .23†

Beck Depression Inventory 0.65 0.19 .47*

Note. *p < .01; †p = .09.

Discussion
Results on adult attachment dysfunction showed that 80% of young adults at UHR of
psychosis in this study (N = 51) were insecurely attached, of which 43% (n = 22)
presented with fearful attachment, 20% (n = 10) preoccupied attachment and 17% (n = 9)
dismissive attachment. The findings suggest that the UHR sample predominantly expressed
anxiety about relationships, such as fears of rejection and being unloved. Young people at
UHR of psychosis reported greater discomfort with intimacy and dependence on others, as
well as fears of abandonment in comparison with the undergraduate samples on which the
Adult Attachment Scale was developed (Collins, 1996). In another study of a community
sample of young adults (Stein et al., 2002), 63% of their sample was securely attached,
which reflects a striking contrast to young people presenting to an early intervention
service for high risk of psychosis. The attachment profiles of the UHR sample in this study
are comparable to a depressed group (BDI ≥ 16) of respondents (Murphy & Bates, 1997;
Reinecke & Rogers, 2001). This is reflected in predominantly fearful and preoccupied
attachment styles in this study’s UHR group. While both these styles are characterized by an
elevated desire for dependency (Bartholomew, 1990), individuals with a preoccupied
attachment style rely on others’ opinion to maintain self-worth (i.e., positive internal
working model of other and a negative internal working of self), whereas individuals with
fearful attachment avoid intimacy with others and have a negative internal working model
of self and other. Overall, these findings suggest that the young people at UHR of psychosis
in this sample may resemble an affectively disturbed group (Morrison et al., 2012; Wigman
et al., 2012). These findings have clinical implications for screening, therapeutic alliance
and intervention, as the focus may as much have to be on the affective dysregulation, as on
positive symptoms of psychosis in the high-risk group.
The findings showed that adult attachment was significantly associated with
depression, anxiety, and social anxiety, with insecurely attached participants reporting
elevated scores on all three affective dimensions. More specifically, participants with
fearful and preoccupied attachment styles had greater social interaction anxiety and
depression than those with secure attachment styles, and those with fearful attachment
also had higher scores on social phobia. Bartholomew’s (1990) four-category model of
adult attachment demonstrates that individuals with a preoccupied attachment style have
a negative view-of-self but a positive view-of-other and those with fearful attachment have
a negative view of self and other. The results of the association between adult attachment
dysfunction and affective dysregulation in the UHR sample suggest that a negative
self-schema, regardless of a view of others is associated with a significant apprehension of
group interaction and vulnerability to depression. However, those who have a negative
view of self and other (i.e., fearful attachment) are likely to experience greater depression,
social anxiety, and social phobia than someone with a secure attachment style.
Our findings are also in line with those of previous studies (Lee & Hankin, 2009) in
which an association between depression and anxious attachment has consistently been
432 Ruchika Gajwani et al.

reported. In two separate studies (Carnelley, Pietromonaco, & Jaffe, 1994; Murphy & Bates,
1997), depression was significantly associated with preoccupied and fearful attachment,
which suggests that depressed mood is linked to a negative view of self, irrespective of a
positive or negative view of others. Support for the association between social anxiety and
insecure attachment has been reported in a longitudinal study of young children who were
followed from the age of 15 months to 9 years (Bohlin, Hagekull, & Rydell, 2000). Their
study revealed that securely attached infants had better social functioning at school and less
social anxiety than children who had been insecurely attached.
The results of this study are consistent with evidence for insecure attachment
relationships as predictors of anxiety symptoms and related affective disorders. In a
longitudinal study, Warren, Huston, Egeland, and Sroufe (1997) found that young children
with anxious/resistant attachment were more likely to develop anxiety disorders in
adolescence than those who were securely attached. These findings were significant after
accounting for the effect of maternal anxiety and temperament assessed when the
children were newborns. There is also evidence from a cross-sectional analysis in a sample
of young adolescents (N = 155), for a significant association between insecure attach-
ment and elevated anxiety symptoms (Muris, Meesters, van Melick, & Zwambag, 2001).
Overall, the results are consistent with previous work which suggests that difficulties in
interpersonal relationships are associated with less favourable emotional regulation
(Mikulincer et al., 2003).
The link between attachment difficulties and affective disturbances of depression and
anxiety is fairly well established (Dozier, Stovall, & Albus, 1999; Rosenstein & Horowitz,
1996), the focus on adult attachment and social anxiety has been limited, despite its clinical
implications as a disorder and a psychiatric comorbidity. With the aim of examining the
association between adult attachment, social anxiety and depression, Eng et al. (2001)
found, among patients with social anxiety disorder, two clusters of attachment styles:
patients with anxious attachment style experienced greater social anxiety and avoidance,
depression and fear of scrutiny than the securely attached patients, as well as a non-clinical
control group. Of further interest is the reported mediational role of social anxiety
symptoms in the association between attachment styles and depression. To explain these
results, the authors draw on the cognitive and affective influences of attachment
relationships (Collins & Read, 1990) which may increase the severity of social anxiety, and
this in turn may affect the experience of depression. Based on the findings of Eng et al.
(2001) and results of the ANOVA, a similar mediational analysis was performed. Due to the
cross-sectional nature of the data, it is difficult to establish the predictive capacity of the
variables; however, social anxiety and social phobia were kept as a mediating factors for
depression in the analyses based on the assumption that social anxiety is a risk factor for
subsequent depression (Stein et al., 2001). Results of the analysis in the UHR sample failed
to support the mediating role of social anxiety between attachment styles and depression.
However, of interest was the mediating role of social phobia between attachment and
depression. With regard to performance related anxiety, it is known that self-criticism and
fear of scrutiny are not unique to social phobia (Birchwood et al., 2007). Social functioning
impairments are predictors of transition to psychosis but may also determine vulnerability
to other affective and non-affective disorders in young people at risk of psychosis.
While it may be difficult to clinically differentiate depression and anxiety (Shorter &
Tyrer, 2003), and to study the causal links due to cross-sectional analysis, results suggest
that depression mediated the relationship between adult attachment and social anxiety in
the UHR sample. The negative schemas emerging from an adult attachment dysfunction
may increase the vulnerability to depression (Murphy & Bates, 1997), in turn reducing the
 Attachment, affective dysregulation and psychosis 433

chances of rewarding social interactions, feeding back to feelings of hopelessness.

Alternatively, we speculate that perceptions of affective bonds at the time of assessment
may be influenced by symptoms of depression and social anxiety. Pathways to social
anxiety and depression in the UHR sample may be wide-ranging. Exposure to cumulative
risk factors such as trauma and cannabis use is strongly associated with comorbid social
anxiety and paranoia in the general population (Schutters et al., 2012), as well as
increasing the risk of a poor prognosis for those vulnerable to psychosis (Cougnard et al.,
2007; van Os et al., 2002). The links between trauma and affective dysregulation are
investigated in a forthcoming paper. Although predictions from attachment to psycho-
pathology get more complex as attachment styles are more durable in ‘low-risk samples’
and more variable in ‘high-risk samples’ (Van IJzendoorn & Bakermans-Kranen-
burg,1997), the relevance of attachment relationships is nevertheless pertinent because
they reflect the dominant working models.

Methodological considerations
The high prevalence of affective dysregulation in the UHR sample may be due to a
self-selection/referral bias, with better detection of young adults expressing distress. It
may also be that high-risk young adults with emotional difficulties may be better at help
seeking and therefore over-represented in this sample. The use of a single index measure
of adult attachment may bias the findings as affective dysregulation could contribute to the
current working models of attachment. Also, RAAS has been validated in adult samples
and while it has shown significant correlations with a childhood attachment questionnaire
in an adult sample with psychosis (Tait et al., 2004), the study considers the use of an adult
attachment measure on a proportion of clients under 18 years of age (33%) in this study, a
methodological limitation. Although the cross-sectional nature of the data makes it
difficult to provide sufficient evidence for the causal links between adult attachment styles
and affective dysregulation, our findings are closely in line with longitudinal findings on
developmental psychopathology.

Clinical implications
Taking into account the methodological strengths and weaknesses, findings of this study
have clinical and theoretical implications for understanding developmental pathways to
affective dysregulation, specifically social anxiety and depression in young people at UHR
of developing psychosis. Our findings are vital in bringing to the forefront clinically
significant levels of distress and anxiety experienced by the young people at UHR of
psychosis. The influence of adult attachment difficulties on vulnerability to experiences of
depression and social anxiety may indicate overriding interpersonal negative schemas,
specific to attachment styles. The treatment of affective dysregulation and implications on
social integration is crucial for recovery for young people in the emerging psychosis, as the
established affective bonds reflect the internal models of self and other, and their specific
association with emotional regulation.

Acknowledgements
The authors would like to thank the Early Detection and Intervention Team at Birmingham and
Solihull Mental Health Foundation Trust for their support and help with recruitment. Professor
434 Ruchika Gajwani et al.

Birchwood and Dr Patterson are partly funded by the National Institute for Health Research
(NIHR) through the Collaborations for Leadership in Applied Health Research and Care for
Birmingham and Black Country (CLAHRC-BBC) programme.

References
Ainsworth, M. S., Blehar, M. C., Waters, E., & Wall, S. (1978). Patterns of attachment: A
psychological study of the strange situation. Hillsdale, NJ: Erlbaum.
Baron, R. M., & Kenny, D. A. (1986). The moderator mediator variable distinction in social
psychological-research – Conceptual, strategic, and statistical considerations. Journal of
Personality and Social Psychology, 51, 1173–1182. doi:10.1037/0022-3514.51.6.1173
Bartholomew, K. (1990). Avoidance of intimacy: An attachment perspective. Journal of Social and
Personal Relationships, 7, 147–178. doi:10.1177/0265407590072001
Bartholomew, K. (1997). Adult attachment processes: Individual and couple perspectives. British
Journal of Medical Psychology, 70, 249–263. doi:10.1111/j.2044-8341.1997.tb01903.x
Beck, A. T., Epstein, N., Brown, G., & Steer, R. A. (1988). An inventory for measuring clinical anxiety:
Psychometric properties. Journal of Consulting and Clinical Psychology, 56, 893–897. doi:10.
1037/0022-006X.56.6.893
Beck, A. T., Rush, A. J., Shaw, B. F., & Emery, G. (1979). Cognitive therapy of depression. New York,
NY: Guilford Press.
Berry, K., Barrowclough, C., & Wearden, A. (2007). A review of the role of adult attachment style in
psychosis: Unexplored issues and questions for further research. Clinical Psychology Review,
27, 458–475. doi:10.1016/j.cpr.2006.09.006
Berry, K., Wearden, A., Barrowclough, C., & Liversidge, T. (2006). Attachment styles, interpersonal
relationships and psychotic phenomena in a non-clinical student sample. Personality and
Individual Differences, 41, 707–718. doi:10.1016/j.paid.2006.03.009
Birchwood, M. (2003). Pathways to emotional dysfunction in first-episode psychosis. British
Journal of Psychiatry, 182, 373–375. doi:10.1192/bjp.182.5.373
Birchwood, M., Gilbert, P., Gilbert, J., Trower, P., Meaden, A., Hay, J., … Miles, J. N. (2004).
Interpersonal and role-related schema influence the relationship with the dominant ‘voice’ in
schizophrenia: A comparison of three models. Psychological Medicine, 34, 1571–1580. doi:10.
1017/S0033291704002636
Birchwood, M., Trower, P., Brunet, K., Gilbert, P., Iqbal, Z., & Jackson, C. (2007). Social anxiety and
the shame of psychosis: A study in first episode psychosis. Behaviour Research and Therapy,
45, 1025–1037. doi:10.1016/j.brat.2006.07.011
Bohlin, G., Hagekull, B., & Rydell, A. M. (2000). Attachment and social functioning: A longitudinal
study from infancy to middle childhood. Social Development, 9, 24–39. doi:10.1111/1467-9507.
00109
Bowlby, J. (1973). Attachment and loss: Anxiety and anger. In Attachment and loss: Vol. 2:
Separation, anxiety and anger. London, UK: The Hogarth Press and the Institute of
Psycho-Analysis.
Bowlby, J. (1979). The making & breaking of affectional bonds. London, UK: Tavistock
Publications.
Bowlby, J. (1980). Attachment and loss. Vol. 3: Sadness and depression. New York, NY: Basic
Books.
Bowlby, J. (1982). Attachment and loss. Vol. 1: Attachment. New York, NY: Basic Books.
Carlson, E. A., & Sroufe, L. A. (1995). Contribution of attachment theory to developmental
psychology. In D. Cicchetti & D. J. Cohen (Eds.), Developmental psychopathology, Vol. 1:
Theory and methods (pp. 581–617). New York, NY: Cambridge University Press.
Carnelley, K. B., Pietromonaco, P. R., & Jaffe, K. (1994). Depression, working models of others, and
relationship functioning. Journal of Personality and Social Psychology, 66, 127–140. doi:10.
1037/0022-3514.66.1.127
 Attachment, affective dysregulation and psychosis 435

Collins, N. L. (1996). Working models of attachment: Implications for explanation, emotion, and
behavior. Journal of Personality and Social Psychology, 71, 810–832. doi:10.1037/0022-3514.
71.4.810
Collins, N. L., & Read, S. J. (1990). Adult attachment, working models, and relationship quality in
dating couples. Journal of Personality and Social Psychology, 58, 644–663. doi:10.1037/
0022-3514.58.4.644
Cougnard, A., Marcelis, M., Myin-Germeys, I., De Graaf, R., Vollebergh, W., Krabbendam, L., …Van
Os, J. (2007). Does normal developmental expression of psychosis combine with environmental
risk to cause persistence of psychosis? A psychosis proneness-persistence model. Psychological
Medicine, 37, 513–528. doi:10.1017/S0033291706009731
Couture, S., Lecomte, T., & Leclerc, C. (2007). Personality characteristics and attachment in first
episode psychosis: Impact on social functioning. The Journal of Nervous and Mental Disease,
195, 631–639. doi:10.1097/NMD.0b013e31811f4021
Cummings, E. M., & Cicchetti, D. (1990). Toward a transactional model of relations between
attachment and depression. In M. T. Greenberg, D. Cicchetti & E. M. Cummings (Eds.),
Attachment in the preschool years: Theory, research, and intervention (pp. 339–372).
Chicago, IL: University of Chicago Press.
Dozier, M. (1990). Attachment organization and treatment use for adults with serious
psychopathological disorders. Development and Psychopathology, 2, 47–60. doi:10.1017/
S0954579400000584
Dozier, M., Stovall, K. C., & Albus, K. (1999). Attachment and psychopathology in adulthood. In J.
Cassidy & P. R. Shaver (Eds.), Handbook of attachment: Theory, research, and clinical
applications (pp. 497–519). New York, NY: Guilford Press.
Eng, W., Heimberg, R. G., Hart, T. A., Schneier, F. R., & Liebowitz, M. R. (2001). Attachment in
individuals with social anxiety disorder: The relationship among adult attachment styles, social
anxiety, and depression. Emotion, 1, 365–380. doi:10.1037/1528-3542.1.4.365
Fisher, H. L., Appiah-Kusi, E., & Grant, C. (2012). Anxiety and negative self-schemas mediate the
association between childhood maltreatment and paranoia. Psychiatry Research, 196, 323–
324. doi:10.1016/j.psychres.2011.12.004
Fonagy, P. (1999). Psychoanalytic theory from the viewpoint of attachment theory and research. In J.
Cassidy & P. R. Shaver (Eds.), Handbook of attachment: Theory, research, and clinical
applications (pp. 595–624). New York, NY: Guilford Press.
Garety, P. A., Kuipers, E., Fowler, D., Freeman, D., & Bebbington, P. E. (2001). A cognitive model of
the positive symptoms of psychosis. Psychological Medicine, 31, 189–195. doi:10.1017/
S0033291701003312
Gumley, A. I., & MacBeth, A. (2006). A trauma-based model of relapse in psychosis. In W. Larkin &
A. P. Morrison (Eds.), Trauma and psychosis: New directions for theory and therapy (pp. 283–
304). Hove: Routledge.
Johnstone, E. C., Ebmeier, K. P., Miller, P., Owens, D. G. C., & Lawrie, S. M. (2005). Predicting
schizophrenia: Findings from the Edinburgh high-risk study. British Journal of Psychiatry, 186,
18–25. doi:10.1192/bjp.186.1.18
Lee, A., & Hankin, B. L. (2009). Insecure attachment, dysfunctional attitudes, and low self-esteem
predicting prospective symptoms of depression and anxiety during adolescence. Journal of
Clinical Child and Adolescent Psychology, 38, 219–231. doi:10.1080/15374410802698396
Mattick, R. P., & Clarke, J. C. (1998). Development and validation of measures of social phobia
scrutiny fear and social interaction anxiety. Behaviour Research and Therapy, 36, 455–470.
doi:10.1016/S0005-7967(97)10031-6
McGlashan, T. H., Miller, T. J., Woods, S. C., Rosen, J. L., Hoffman, R. E., & Davidson, L. (2002).
Structured Interview for Prodromal Syndromes (version 3.1). New Haven, CT: PRIME
Research Clinic, Yale School of Medicine.
Mikulincer, M., Shaver, P. R., & Pereg, D. (2003). Attachment theory and affect regulation: The
dynamics, development, and cognitive consequences of attachment-related strategies.
Motivation and Emotion, 27, 77–102.
436 Ruchika Gajwani et al.

Miller, T. J., McGlashan, T. H., Rosen, J. L., Somjee, L., Markovich, P. J., Stein, K., & Woods, S. W.
(2002). Prospective diagnosis of the initial prodrome for schizophrenia based on the structured
interview for prodromal syndromes: Preliminary evidence of inter-rater reliability and predictive
validity. American Journal of Psychiatry, 159(5), 863–865. doi:10.1176/appi.ajp.159.5.863
Morrison, A. P., French, P., Stewart, S. L., Birchwood, M., Fowler, D., Gumley, A. I., … Dunn, G.
(2012). Early detection and intervention evaluation for people at risk of psychosis: Multisite
randomised controlled trial. British Medical Journal, 344, e2233. doi:10.1136/bmj.e2233
Muris, P., Meesters, C., van Melick, M., & Zwambag, L. (2001). Self-reported attachment style,
attachment quality, and symptoms of anxiety and depression in young adolescents. Personality
and Individual Differences, 30, 809–818. doi:10.1016/S0191-8869(00)00074-X
Murphy, B., & Bates, G. W. (1997). Adult attachment style and vulnerability to depression.
Personality and Individual Differences, 22, 835–844. doi:10.1016/S0191-8869(96)00277-2
Owens, K. A., Haddock, G., & Berry, K. (2012). The role of the therapeutic alliance in the regulation
of emotion in psychosis: An attachment perspective. Clinical Psychology & Psychotherapy.
Advance online publication. doi:10.1002/cpp.1793
Owens, D. G. C., Miller, P., Lawrie, S. M., & Johnstone, E. C. (2005). Pathogenesis of schizophrenia: A
psychopathological perspective. British Journal of Psychiatry, 186, 386–393. doi:10.1192/bjp.
186.5.386
Parker, G., Roussos, J., Hadzi-Pavlovic, D., Mitchell, P., Wilhelm, K., & Austin, M. P. (1997). The
development of a refined measure of dysfunctional parenting and assessment of its relevance in
patients with affective disorders. Psychological Medicine, 27, 1193–1203. doi:10.1017/
S003329179700545X
Peters, L. (2000). Discriminant validity of the Social Phobia and Anxiety Inventory (SPAI), the Social
Phobia Scale (SPS) and the Social Interaction Anxiety Scale (SIAS). Behaviour Research and
Therapy, 38, 943–950. doi:10.1016/S0005-7967(99)00131-X
Read, J., & Gumley, A. (2008). Can attachment theory help explain the relationship between
childhood adversity and psychosis? Attachment: New Directions in Psychotherapy and
Relational Psychoanalysis, 2, 1–35.
Reinecke, M. A., & Rogers, G. M. (2001). Dysfunctional attitudes and attachment style among
clinically depressed adults. Behavioural and Cognitive Psychotherapy, 29, 129–141. doi:10.
1017/S1352465801002016
Richter, P., Werner, J., Heerlein, A., Kraus, A., & Sauer, H. (1998). On the validity of the Beck
Depression Inventory – A review. Psychopathology, 31, 160–168. doi:10.1159/000066239
Roberts, J. E., Gotlib, I. H., & Kassel, J. D. (1996). Adult attachment security and symptoms of
depression: The mediating roles of dysfunctional attitudes and low self-esteem. Journal of
Personality and Social Psychology, 70, 310–320. doi:10.1037/0022-3514.70.2.310
Rosenstein, D. S., & Horowitz, H. A. (1996). Adolescent attachment and psychopathology. Journal
of Consulting and Clinical Psychology, 64, 244–253. doi:10.1037/0022-006X.64.2.244
Schutters, S. I., Dominguez, M., Knappe, S., Lieb, R., van Os, J., Schruers, K. R. J., & Wittchen, H. U.
(2012). The association between social phobia, social anxiety cognitions and paranoid
symptoms. Acta Psychiatrica Scandinavica, 125, 213–227. doi:10.1111/j.1600-0447.2011.
01787.x
Shorter, E., & Tyrer, P. (2003). Separation of anxiety and depressive disorders: Blind alley in
psychopharmacology and classification of disease. British Medical Journal, 327, 158–160.
doi:10.1136/bmj.327.7407.158
Sroufe, L. A., Carlson, E. A., Levy, A. K., & Egeland, B. (1999). Implications of attachment theory for
developmental psychopathology. Development and Psychopathology, 11, 1–13. doi:10.1017/
S0954579499001923
Stein, M. B., Fuetsch, M., Muller, N., Hofler, M., Lieb, R., & Wittchen, H. U. (2001). Social anxiety
disorder and the risk of depression – A prospective community study of adolescents and young
adults. Archives of General Psychiatry, 58, 251–256. doi:10.1001/archpsyc.58.3.251
 Attachment, affective dysregulation and psychosis 437

Stein, H., Koontz, A. D., Fonagy, P., Allen, J. G., Fultz, J., Brethour, J. R., … Evans, R. B. (2002). Adult
attachment: What are the underlying dimensions? Psychology and Psychotherapy: Theory,
Research and Practice, 75, 77–91. doi:10.1348/147608302169562
Tait, L., Birchwood, M., & Trower, P. (2004). Adapting to the challenge of psychosis: Personal
resilience and the use of sealing-over (avoidant) coping strategies. The British Journal of
Psychiatry, 185, 410–415. doi:10.1192/bjp.185.5.410
Thompson, R. A. (1999). Early attachment and later development. In J. Cassidy & P. R. Shaver (Eds.),
Handbook of attachment (pp. 265–286). New York, NY: Guilford.
van IJzendoorn, M. H., & Bakermans-Kranenburg, M. J. (1997). Intergenerational transmission of
attachment: A move to the contextual level. In L. D. Atkinson & K. J. Zucker (Eds.), Attachment
and psychopathology (pp. 135–168). New York, NY: Guilford Press.
van Os, J., Bak, M., Hanssen, M., Bijl, R. V., De Graaf, R., & Verdoux, H. (2002). Cannabis use and
psychosis: A longitudinal population-based study. American Journal of Epidemiology, 156,
319–327. doi:10.1093/aje/kwf043
Warren, S. L., Huston, L., Egeland, B., & Sroufe, L. A. (1997). Child and adolescent anxiety disorders
and early attachment. Journal of the American Academy of Child and Adolescent Psychiatry,
36, 637–644. doi:10.1097/00004583-199705000-00014
Wigman, J. T., van Nierop, M., Vollebergh, W. A., Lieb, R., Beesdo-Baum, K., Wittchen, H. U., & van
Os, J. (2012). Evidence that psychotic symptoms are prevalent in disorders of anxiety and
depression, impacting on illness onset, risk, and severity — Implications for diagnosis and
ultra-high risk research. Schizophrenia Bulletin, 38, 247–257. doi:10.1093/schbul/sbr196
Yung, A. R., Phillips, L. J., Yuen, H. P., Francey, S. M., McFarlane, C. A., Hallgren, M., & McGorry, P. D.
(2003). Psychosis prediction: 12-month follow up of a high-risk (“prodromal”) group.
Schizophrenia Research, 60, 21–32. doi:10.1016/S0920-9964(02)00167-6

Received 22 January 2013; revised version received 7 July 2013