Cognitive Therapy and Research, Vol. 29, No.

6, December 2005 (

C 2006), pp. 733–746

DOI: 10.1007/s10608-005-4607-6

Relationship Patterns Associated with Dimensions

of Vulnerability to Psychopathology
Nicholas B. Allen,1,2 Nick Haslam,1 and Assaf Semedar1

Associations between vulnerability to psychopathology and atypical ways of perceiv-

ing and conducting relationships were investigated in a study of 350 undergraduates.
Self-report measures of traits associated with vulnerability to depression, bipolar dis-
order and psychosis were used to predict atypical relationships with family members,
close friends, student peers, and authority figures. Tendencies to implement four basic
relational models in personal relationships of each type were examined. Depression-
proneness was associated with tendencies to apprehend family and close friend rela-
tionships in unusually close and asymmetrical ways. Vulnerability to bipolar disorder
was manifest in relationships with authority figures, which vulnerable individuals con-
ducted in an unusually communal and egalitarian manner. Psychosis-proneness was
associated with two distinct patterns. Socially anhedonic individuals were less apt than
others to view family and close friend relationships as warm and balanced, and viewed
relations with friends in a hierarchical fashion. This perception of asymmetry in friend
and also family relationships was also apparent among individuals with more “pos-
itive” schizotypal features. Atypical relational patterns appear to be associated with
vulnerability for major mental disorders.
KEY WORDS: relational models; social cognition; psychopathology; vulnerability; dysthymia; hypoma-
nia; schizotypy; social anhedonia.

RELATIONSHIP PATTERNS ASSOCIATED WITH VULNERABILITY

TO PSYCHOPATHOLOGY

Disturbances in interpersonal functioning are prominent features of most forms

of psychopathology. Impaired social cognition may be a critical mediator of these
processes, compromising the ability to process interpersonal information (Broks,
1997) and to function adaptively in social contexts (e.g., Blanchard, Mueser, &
Bellack, 1998). For instance, interpersonal processes play an important role in the
1 Universityof Melbourne, ORYGEN Research Centre.
2 Correspondence should be directed to Nicholas B. Allen, ORYGEN Research Centre and Department
of Psychology, University of Melbourne, Victoria 3010, Australia; e-mail: [email protected].

733
0147-5916/05/1200-0733/0

C 2006 Springer Science+Business Media, Inc.
734 Allen, Haslam, and Semedar

development and maintenance of depression (Segrin & Abramson, 1994), where it

has been shown that depressed people tend to elicit rejection from others (Coyne,
1976). Social cognition is also markedly abnormal in depressed individuals, who en-
dorse beliefs that are likely to result in negative interpretations of interpersonal
situations (e.g., “People will think less of me if I make a mistake”; Dohr, Rush,
& Bernstein, 1989; Lewinsohn, Steinmetz, & Larson, & Franklin, 1981), and direct
their attention towards stimuli indicating “social threat” (Mathews, Ridgeway, &
Williamson, 1996). This hypervigilance towards social threat, which may be adap-
tive in circumstances of genuine threat to social well-being, may result in markedly
maladaptive social behavior when benign social cues are misclassified as threaten-
ing (Allen & Badcock, 2003; Badcock & Allen, 2003). Interpersonal dependency
and excessive reassurance seeking are examples of the sorts of maladaptive social
behaviors that might emerge from such a misclassification and are strongly associ-
ated with vulnerability to depression (Coyne & Whiffen, 1995; Joiner & Schmidt,
1998).
Interpersonal processes have also been recognized as important factors in the
development and maintenance of schizophrenia-spectrum disorders (Penn, Corri-
gan, Bentall, Racenstein, & Newman, 1997). In addition to deficits in general cog-
nition and interpersonal behavior, the schizophrenias have been found to involve
specific abnormalities of social cognition (e.g., Corrigan & Toomey, 1995; Frith &
Corcoran, 1996; Penn, Spaulding, Reed, & Sullivan, 1996). A number of such abnor-
malities have been discussed, including deficits in the perception of emotion in other
people (e.g., Edwards, Jackson, & Pattison, 2002) and more general deficits in being
able to infer the mental states of others (i.e., a deficit in “theory of mind”; Frith &
Corcoran, 1996). Bentall (2001) has recently discussed how these biases and deficits
in social cognition might combine to produce symptoms of paranoia, and other so-
cially oriented delusions, where the social world is seen as excessively hostile and
dangerous.
Interpersonal problems have also been associated with symptoms of bipolar
affective disorder, and one of its putative phenotypic precursors, hypomanic per-
sonality. These problems have included high expressed emotion in families and low
levels of social support more generally (Johnson & Meyer, 2003; Klein, Lewinsohn,
& Seeley, 1996). Unfortunately relatively little is understood about the social cog-
nitive deficits that might underlie such processes in those suffering from, or at risk
of, bipolar disorder, but there are some suggestive recent findings. Neale (1988) has
provided a modern interpretation of the classic psychoanalytic concept of manic
defense, suggesting that one of the core symptoms of mania, grandiose ideas, is gen-
erally precipitated when distressing thoughts or psychosocial stress begin to over-
whelm the person. The mood elevation and other manic symptoms then act to keep
such distressing thoughts out of consciousness (Neale, 1988). This hypothesis pre-
dicts that during mania there is dissociation between the explicit presentation of
the self (i.e., as powerful, loved, admired) and the implicit or latent categorization
of the self (i.e., as powerless, unlovable). In support of this model Lyon, Bentall
and Startup (1999) found that both manic and depressive patient groups attributed
more negative than positive events to self, and demonstrated slowed color naming of
depression-related words in a Stroop task. However, like the control group, manic
Social Cognition and Relationships 735

patients explicitly endorsed more positive words as true of self. Such a dissociation
between implicit and explicit self-concepts would likely cause considerable prob-
lems in interpersonal relationships, especially hierarchically oriented relationships,
where the bipolar person’s defensive tendency to see themselves in excessively pow-
erful and attractive terms may prevent them from engaging in adaptively submissive
behavior when dealing with superiors.
Despite the dramatic advances that have been made in studying interpersonal
and social-cognitive aspects of mental disorders, a number of issues remain to be
clarified. First, most investigations of such phenomena have compared a single dis-
order to a healthy comparison group. Second, in order to clarify which processes are
etiologically significant, as opposed to those features that are primarily associated
with the acute or residual phases of the disorder, it is important to study phenotypic
variations in vulnerability to disorder. This also appears to be a lacuna in the litera-
ture. For example, Penn and colleagues (1997) have pointed out there is very little
research on social cognition with populations at risk of schizophrenia, and similar as-
sertions can be made regarding affective disorders (Sacco, 1999). Finally, relatively
little of the research to date has utilized a comprehensive model of human social
cognition, and the social behavior that emerges from it, to organize the examina-
tion of disordered social cognition and behavior in psychopathology. Such a model
is offered by Fiske’s (1992) work on relational models.
Relational models theory (Fiske, 1992) proposes four fundamental cognitive-
affective models in terms of which people interpret, construct, evaluate, and seek
social relationships. The Communal Sharing model structures relationships in terms
of equivalence and shared identity, in which members of a group or dyad perceive
themselves as an undifferentiated unit, and is often exemplified by romantic and
family relationships. The Equality Matching model, which often governs relations
with friends and colleagues, is based on balanced egalitarian reciprocity, manifested
in turn-taking and keeping track of favors and obligations. Authority Ranking rep-
resents relationships in asymmetrical and hierarchically ordered terms, based on
differences in status, knowledge, or power. Market Pricing, finally, construes rela-
tionships in terms of proportionality or equity, using socially meaningful ratios such
as rates of return on time, effort or investment, and often but not always involves
money.
In theory, the four relational models are universal and discrete, and frequently
govern relationships in combination, with research consistently supporting these
claims (Fiske, 1993; Haslam, 1994; Haslam & Fiske, 1999). An extensive empir-
ical literature supports the capacity of the RMs to illuminate an assortment of
social-cognitive and interactional phenomena (see Fiske & Haslam, 1996; Haslam,
2004, for reviews). For example, the models have been shown to organize implicit
and explicit classification of personal relationships, to guide judgments of relation-
ship prototypicality, to predict the patterning of social errors and social activity
planning, to underpin person memory, and to influence judgments of relational
transgressions.
Although relational models theory has garnered extensive support as an
account of the mental representation of social relationships, little research has been
conducted on associations between individual differences in the implementation
736 Allen, Haslam, and Semedar

of the models and normal personality or psychopathology. Haslam, Reichert,

and Fiske (2002) obtained several predicted associations between personality
disorder symptomatology and broad tendencies to implement and be invested in
particular relational models. Relatively strong tendencies to implement Authority
Ranking in personal relationships were associated with a wide range of personality
disorder symptoms, but were independent of the interpersonal tendencies captured
by the interpersonal circumplex (e.g., Kiesler, 1983). Caralis and Haslam (2004)
demonstrated a number of links between individual differences in implementation
of the relational models and the five-factor model of personality, with Neuroticism,
for example, correlating with relatively high implementation of Authority Ranking
and Market Pricing and relatively low implementation of Communal Sharing and
Equality Matching. These two studies point to promising links between personality
and patterns of implementation of the relational models, consistent with the view
that people experiencing psychological disturbance may conduct their relationships
in atypical ways. However, no studies to date have investigated personality diathe-
ses or axis I mental disorders, or examined how particular relationship types are
cognized.
Bearing in mind these considerations, this paper reports the first examination
of how atypical ways of implementing the relational models are associated with trait
indicators of risk for serious psychopathology (i.e., psychosis, bipolar and unipolar
affective disorder), in an undergraduate sample. We examined how the relational
models were implemented within particular relational types, as higher levels of vul-
nerability for psychopathology might be manifest in atypical use of particular mod-
els in particular interpersonal contexts.
Although the study was preliminary, employing an unselected rather than “at
risk” sample, we approached it with three predictions drawn from the research lit-
erature. First, we predicted that vulnerability to unipolar depression would be asso-
ciated with unusually strong implementation of Communal Sharing and Authority
Ranking. Interpersonal dependency and reassurance seeking are associated with de-
pression (e.g., Coyne & Whiffen, 1995; Joiner & Schmidt, 1998), and Haslam et al.
(2002) found that dependent personality was associated with relatively strong in-
vestment in these relationships. People scoring high on vulnerability to depression
should therefore construe and conduct their relationships in an unusually close,
communal, and asymmetrical fashion. This may be especially true of close rela-
tionships, where dependency and reassurance seeking behaviors are most often ex-
pressed (Allen, Gilbert, & Semedar, 2004). Second, we predicted that people high
in psychosis proneness would conceive of relationships as relatively asymmetrical,
cold and potentially exploitative, by over implementing Authority Ranking and un-
der implementing Communal Sharing and Equality Matching. Vulnerability to psy-
chosis is associated with deficits in theory of mind, leading to suspiciousness and
paranoia within relationships, and Haslam et al. (2002) found that schizotypal per-
sonality was associated with relatively strong implementation of Authority Ranking
in relationships. Third, we predicted that people relatively high in vulnerability to
bipolar disorder would demonstrate abnormalities in implementation of relational
models in typically hierarchical relationships where Authority Ranking is the usual
mode of interaction. We predicted that such people’s explicit view of themselves
as unusually powerful and attractive may render them likely to treat relationships
Social Cognition and Relationships 737

with superiors in ways that are overly close and egalitarian (i.e., relatively high in
Communal Sharing and Equality Matching).

METHOD

Participants
Participants were 350 undergraduate volunteers, 295 women and 55 men, who
received course credit for their participation. They ranged in age from 16 to 49 (M =
19.8, SD = 4.4) and were ethnically diverse (50.0% Anglo-Australian, 28.7% Asian,
6.2% European, 3.1% Asian-Australian, 2.1% Indian, 6.2% other).

Materials
Modes of Relationship Questionnaire (MORQ)
The MORQ is a 32-item self-report measure (Haslam & Fiske, 1999) for as-
sessing personal relationships in terms of RM theory. Items were constructed to
sample multiple social domains (decision making, distribution and use of resources,
exchange, identity, moral evaluations, social influence, work, and other) and are
rated from 0 (not at all true of this relationship) to 6 (very true of this relationship).
Confirmatory factor analysis (Haslam & Fiske, 1999) supports the coherence of the
four 8-item scales, and previous studies have obtained scale reliabilities between 0.8
and 0.9.
Following standard practice (e.g., Haslam, 1994), participants freely listed 40
personal acquaintances and rated the relationship they had with every fourth ac-
quaintance on the 32 items. By this means a broad and representative sample of 10
personal relationships was rated. Each relationship was scored on each relational
model scale, with scores equaling the mean item rating on each scale. In addition,
participants were instructed to identify the role that each of the 10 selected acquain-
tances had towards them, using ten listed categories (family member, close friend,
fellow student, business associate, acquaintance, fellow colleague, tutor, boss, eco-
nomic relationship, and other).

General Behavior Inventory (GBI; Depue & Klein, 1988; Depue, Krauss, Spoont,
& Arbisi, 1989)
The GBI is a 73-item self-report inventory designed to identify individuals at
risk of mood disorders, assessing enduring affective symptom patterns. It contains
three scales—dysphoria, hypomania, and biphasic—which are rated on four-point
frequency-based scales. Research has validated the GBI’s concordance with blind
diagnoses based on informant reports (Depue et al., 1981) and with clinician ratings
of subaffective symptoms (Goodnick et al., 1986). Studies have demonstrated asso-
ciations between GBI scores and mood disorders in first-degree relatives (Depue
et al., 1981; Klein, Depue, & Slater, 1986), and shown that the GBI predicts endur-
ing impairment (Klein & Depue, 1984). In the current sample the mean, standard
deviation, minimum and maximum scores (respectively) on the three GBI scales
738 Allen, Haslam, and Semedar

were as follows; dysphoria (6.94, 8.41, 0, 41), hypomania (2.45, 2.99, 0, 18), biphasic
(1.51, 1.60, 0, 6).

Chapman Scales
Three scales developed by Chapman, Chapman and colleagues were employed
to detect psychosis-prone participants. The Magical Ideation scale (Eckblad &
Chapman, 1983), the Perceptual Aberration scale (Chapman, Chapman, & Raulin,
1978) and the Revised Social Anhedonia scale (Mishlove & Chapman, 1985), con-
taining 30, 35, and 40 items, respectively, assess psychosis-related symptomatol-
ogy and were designed for use with normal populations. These self-report inven-
tories are among the most widely used self-report measures of psychosis-proneness,
and have well-establish psychometric credentials (Chapman, Chapman, & Kwapil,
1995). Ten-year follow-up and retrospective studies (Chapman, Chapman, Kwapil,
Eckblad, & Zinser, 1994; Kwapil, Miller, Zinser, Chapman, & Chapman, 1997;
Kwapil, 1998) have found that high scorers have an increased risk of experienc-
ing clinical psychosis or psychotic symptoms. A combination of high scores on the
Magical Ideation and Social Anhedonia scales was particularly predictive of future
psychosis. In the current sample the mean, standard deviation, minimum and max-
imum scores (respectively) on the three Chapman scales were as follows; magical
ideation (6.15, 4.50, 0, 24), perceptual aberration (4.10, 4.07, 0, 27), social anhedonia
(7.25, 5.53, 0, 35).

Procedure
Groups of 5 to 20 undergraduate participants completed the questionnaire
packet in a classroom setting in the presence of one investigator.

RESULTS

Analyses addressed how dimensions of psychological vulnerability are associ-

ated with distinctive relational patterns. These analyses focused on particular rela-
tionship types to determine how, if at all, more vulnerable individuals construe and
implement certain kinds of relationships. Four relationship types were selected in an
effort to sample relationships that were ecologically salient and prototypic of each
of the relational models, so that possible deviations from normative relational pat-
terns could be examined. “Family member” was common, occurring in the relation-
ship samples of 327 of the 350 participants, and also had the highest mean ratings
of all the relationship types on the Communal Sharing scale. “Close friend” was
the most common relationship type, occurring in 338 of the participants’ samples,
and combined high mean ratings on the Communal Sharing and Equality Match-
ing scales. “Fellow student” was another salient relationship type, with its highest
rating on Equality Matching, and was listed by 238 participants. “Boss” and “tu-
tor” (akin to teaching assistant in the North American system) had the two highest
mean scores on the Authority Ranking scale, and occurred in 62 and 76 relationship
samples, respectively. In view of their similar relational profiles, these relationship
Social Cognition and Relationships 739

Table I. Mean (SD) MORQ Scale Ratings for Each Relationship Type
Communal sharing Equality matching Authority ranking Market pricing
Family member 3.59 (1.19) 3.20 (1.12) 3.02 (1.29) 2.26 (.87)
Friend 3.29 (.95) 4.02 (.91) 1.59 (.96) 2.66 (.73)
Fellow student 2.18 (.95) 3.75 (1.03) 1.28 (.86) 2.81 (.80)
Boss/tutor 1.73 (1.04) 2.38 (1.05) 3.75 (1.29) 2.96 (.94)

types were combined to yield a satisfactory sample for statistical analyses (126).
“Economic relationship” was most prototypic for MP. However it occurred in only
19 of the relationship samples, allowing insufficient statistical power, and was not
investigated further. Thus the four selected relationship types consist of the most
prototypic types for Communal Sharing, Equality Matching and Authority Ranking
in addition to the most commonly reported type.
Participants’ scores on each MORQ scale for each relationship type were com-
puted as follows. If participants listed no relationship of a particular type in their
personal relationship sample they were excluded from any analysis concerning that
type. If they listed one relationship of a particular type its score on the relevant
MORQ scale (possible range 0–6) was used. If they listed more than one such rela-
tionship, their mean scale score across these relationships was used. Thus analyses
of relational tendencies associated with psychological vulnerability were based on
at least one personal relationship for each included participant. Mean scale scores
for each relationship type are presented in Table I.
To assess the associations between dimensions of psychological vulnerability
and relational patterns, tendencies to construe relationships of each type according
to each relational model were simultaneously regressed onto the four vulnerability
dimensions. Prior to doing so, the six vulnerability scales were subjected to prin-
cipal components analysis to extract independent predictor variables and thus to
prevent serious multicollinearity. The scree test supported a readily interpretable
four-component solution that accounted for 91.0% of the variance in the intercor-
relation matrix. Varimax rotated loadings are presented in Table II. The first com-
ponent combined the Depression and Biphasic scales of the GBI, which correlated
very highly (r = 0.76). The second component joined the Magical Ideation and Per-
ceptual Aberration scales, which also correlated very highly (r = .68), leaving the re-
maining Chapman scale—the Revised Social Anhedonia scale—as the sole loading
on the third component. The GBI’s Hypomania scale composed the fourth factor.
These four factors served as uncorrelated predictors in a series of 16 simultaneous
multiple regression analyses—four relationship types by four MORQ scales—that
are summarized in Table III. As noted above, these analyses were restricted to those
participants who reported at least one personal relationship of the pertinent rela-
tionship type, so sample sizes vary from 126 to 338.
Consistent with our first prediction, depression-proneness was associated with
relatively high implementation of the Communal Sharing and Authority Rank-
ing models, although these associations were confined to relationships with family
members and also with close friends in the latter case. Consistent with our third
prediction, hypomania was associated with relatively high implementation of the
740 Allen, Haslam, and Semedar

Table II. Varimax Rotated Loadings of Vulnerability Scales

(Decimal Omitted)
Component
I II III IV
Depression (GBI) 91 14 09 20
Biphasic (GBI) 89 16 07 23
Hypomania (GBI) 38 19 −02 90
Social anhedonia (RSA) 11 08 99 −02
Magical ideation (MIS) 08 90 06 18
Perceptual aberration (PAS) 20 90 06 05
Percentage variance 30.2 28.4 16.7 15.7

Communal Sharing and Equality Matching models in authority relationships, with

the former association also obtained for relationships with close friends.
Associations between psychosis-proneness and implementation of the rela-
tional models were somewhat more complex. Consistent with our second predic-
tion, participants scoring relatively high on psychosis-proneness had relatively cold
(low Communal Sharing and Equality Matching) relationships. For social anhedo-
nia, this association only reached significance for family and boss/tutor relationships,
and for positive psychosis-proneness it held only for Equality Matching relation-
ships with fellow students. Also as predicted, psychosis-proneness was associated
with higher levels of Authority Ranking, although only in family and close friend
relationships. In sum, of the 34 associations relevant to the three study predictions,

Table III. Summary of Regression Analyses (Standardized Beta Coefficients and Overall
Model F Values) Predicting Relational Patterns from Vulnerability Components
Vulnerability component
Dysthymia Hypomania Social anhedonia Psychosis F
Communal sharing
Family member .14∗ .08 −.12∗ .11 4.10∗∗
Close friend .03 .15∗∗ −.10 .06 2.87∗
Fellow student .07 .01 −.10 −.11 1.42
Boss/tutor .09 .31∗∗ −.08 −.06 2.62∗
Equality matching
Family member .12∗ .09 −.17∗∗ .02 3.87∗∗
Close friend .02 .08 −.05 −.02 0.75
Fellow student .11 .04 −.06 −.19∗∗ 2.86∗
Boss/tutor .06 .35∗∗∗ −.19∗ −.14 3.97∗∗
Authority ranking
Family member .11∗ .04 .06 .20∗∗∗ 4.45∗∗
Close friend .13∗ .07 .16∗∗ .12∗ 5.04∗∗∗
Fellow student .03 .06 .03 .10 0.91
Boss/tutor .09 .04 −.11 −.02 0.57
Market pricing
Family member .07 .09 −.03 .04 1.20
Close friend .07 .16∗∗ .03 .05 2.77∗
Fellow student .13 .09 −.02 −.05 1.30
Boss/tutor .04 .15 −.14 .00 1.00
∗p < .05. ∗∗ p < .01. ∗∗∗ p < .001.
Social Cognition and Relationships 741

12 were statistically significant and 29 were in the predicted direction. With one
exception (positive psychosis-proneness and Communal Sharing), every significant
predicted association was observed for at least one of the relationship types.
In addition to the predicted associations, it is interesting to note that imple-
mentation of the Market Pricing model was not associated with any of the vul-
nerability dimensions. A similar observation can be made about relationships with
fellow students. Closer, more intimate and involving relationships, and relation-
ships with authority figures, seem more apt to reveal associations with psychological
vulnerability.

DISCUSSION

Although they are preliminary and largely exploratory, and do not offer a rig-
orous test of the specificity of the associations between relational models and vul-
nerability dimensions, our findings offer promising support for the proposition that
vulnerability to major forms of psychopathology is associated with distinctive ways
of understanding and implementing social relationships. Each vulnerability dimen-
sion was associated with particular ways of construing personal relationships with
family members, close friends, student peers, or authority figures. These associa-
tions were generally modest in magnitude and tended not to generalize across rela-
tionship types.
The picture that emerges of the relational tendencies associated with dysthymia
is quite plausible. Individuals high in depression-proneness may have unusually
complicated family relationships. These relationships, which are normatively close,
appear to be apprehended by more dysthymic participants in an especially com-
munal fashion. This pattern might be reflected in heightened desires and demands
for closeness, care and support, and in failure to respect and acknowledge family
members’ autonomy; a pattern that is very consistent with excessive reassurance
seeking behavior that has been previously associated with vulnerability to depres-
sion (Coyne, 1976; Joiner & Schmidt, 1998). The relatively high levels of Authority
Ranking construal of family relationships among people scoring high on dysthymia
may similarly indicate overheated or over-invested family relationships, consistent
with proposals that depressed individuals subordinate themselves in the hope of
eliciting care (Sloman & Gilbert, 2000). This last pattern was also obtained with
close friendships, consistent with the view that depression-prone people may act in
a dependent fashion in search of reassurance and support, particularly in close re-
lationships where such demands of support are more likely to be tolerated (Allen,
Gilbert, & Semedar, 2004). As they pertain to the Communal Sharing and Author-
ity Ranking models, these findings are consistent with our expectations, and with
past work linking dependency with relatively strong use of these models (Haslam
et al., 2002).
These relational patterns may be associated with significant problems for
depression-prone individuals. Communal Sharing and Authority Ranking elements
are common in family relationships, but implementing them to usually strong
degrees may be linked to intense and ambiguous relationships that are marked
by conflict and high levels of emotional expression (Hooley, 1986). The tendency
742 Allen, Haslam, and Semedar

to apprehend close friendships in asymmetrical (Authority Ranking) terms may

also be associated with interpersonal problems. Friends are likely to feel that
people who interact on this basis are violating the expectation of egalitarian
balance and making unreasonable demands for direction and care. This pattern is
consistent with Coyne’s(1976) account, which proposes that the depressed person’s
reassurance-seeking (i.e., positioning of self as subordinate) is associated with
interpersonal rejection. Both family conflict and rejection by friends are stressors
that might help to precipitate depression among the vulnerable (e.g., Monroe,
Rohde, Seeley, & Lewinsohn, 1999; Rudolf, Hammen, Burge, Lindberg, Herzberg,
& Daley, 2000), so relational tendencies associated with elevated depression-
proneness may contribute to the triggering of this vulnerability. Consistent with
this, Joiner, Alfano, and Metalsky (1992) reported that the degree of reassurance
seeking engaged in by depressed persons was crucial to the likelihood of rejection
by their college roommates, with those seeking greater levels of reassurance being
more likely to experience rejection.
The relational patterns associated with vulnerability to bipolar disorder imply
different kinds of interpersonal conflict. As predicted, higher levels of hypomania
were associated with tendencies to apprehend relationships with authority figures
in unusually communal and egalitarian ways. This construal represents a violation
of expectations for such relationships, as it frames them as relatively warm and con-
ducted by equals. The “superior” in such relationships may perceive this framing as
a sign of excessive familiarity, disrespect, and insubordination and may attempt to
correct or punish this behavior. Relationships with employers and other supervisors
might therefore be particular risky for people prone to bipolar disorder. This atyp-
ical pattern of perceiving superiors may provide a partial explanation for previous
findings indicating high correlations between hypomania and poor job-performance
(e.g., Bartol, 1991).
The vulnerability dimensions normally understood to reflect psychosis-
proneness are associated with less clear-cut relational tendencies. The more “neg-
ative” features captured by social anhedonia are plausibly associated with an un-
usually weak tendency to apprehend some relationship types in terms of the typi-
cally warmer Communal Sharing and Equality Matching models. In addition, social
anhedonia is associated with viewing relationships with close friends as unusually
asymmetrical. This association is also evident among individuals scoring relatively
high on the positive features of psychosis-proneness. Whether such findings are me-
diated by social cognitive deficits, such as impaired theory of mind abilities (Frith
& Corcoran, 1996), will need to be addressed by future research. Nevertheless, the
high Authority Ranking pattern obtained was consistent with prediction, and with
the schizotypy-related findings of Haslam et al. (2002).
The findings offer preliminarysupport the ability of some relational mod-
els to capture aspects of the interpersonal style associated with psychological
vulnerability. The Communal Sharing, Authority Ranking and Equality Match-
ing models, in particular, showed preliminary evidence of being implicated in
vulnerability-related interpersonal profiles. This broad finding is consistent with
Haslam et al.’s (2002) finding that many personality disorders were associated with
atypical implementation of these models. It is theoretically important because some
Social Cognition and Relationships 743

of these models are not captured by most prominent models of interpersonal be-
havior employed in the study of abnormal psychology. Tendencies to construe rela-
tionships in ranked (Authority Ranking) or egalitarian (Equality Matching) terms
are conceptually and empirically distinct from the dimensions of the interpersonal
circle (e.g., Kiesler, 1983) and of Benjamin’s (1996) structural analysis of social be-
havior (SASB) model. The circle describes individuals’ behaviors along an axis of
power or dominance, but cannot capture tendencies to apprehend relationships as
involving differentials on this axis. For example, it can describe a person’s behavior
as unusually dominating, but not his or her tendency to apprehend relationships in
terms of dominance and submission. The SASB model’s axes involve independence
versus interdependence and warmth versus coldness, neither of which reflects Au-
thority Ranking or Equality Matching. The RMs therefore appear to cover distinct
and potentially fruitful interpersonal ground from the more familiar schemes.
Many of the associations between relational patterns and vulnerability dimen-
sions are weak, and their practical importance could be questioned. However, even
when whole-sample correlations are modest, individuals who score in the top few
percentiles on the vulnerability dimensions will tend to have substantially atypical
relational patterns. That is, the individuals who are at most elevated risk of devel-
oping fully-fledged disorders may be relationally atypical despite the small sample-
wide associations. Nevertheless, our exploratory analysis does not allow a strong
or rigorous test of this possibility, and until the associations between vulnerabilities
and relational models are found to be robust in a less exploratory investigation they
must remain provisional. Moreover, the large number of associations tested raises
the likelihood of Type 1 errors, making replication of our findings particularly im-
portant. The import of our findings may extend to etiological and prevention issues.
If people at increased risk for major mental disorders tend to construe and con-
duct their personal relationships in atypical ways, then they are likely to experience
above average levels of interpersonal discord and rejection in their interactions.
Relationships about which participants hold discrepant understandings tend to be
unsatisfying, poorly coordinated, and prone to failure, and research shows that dis-
crepancies also commonly produce moral indignation and upset in the interactants
(Fiske & Tetlock, 1997). Such interpersonal failure, rejection, and strife have been
repeatedly implicated as stressors capable of triggering mental disorders in vulnera-
ble individuals (e.g., Hooley, Orley, & Teasdale, 1986; Monroe et al., 1999; Rudolph
et al., 2000). In a cruel irony, the relational tendencies associated with vulnerabili-
ties to psychopathology may tend to elicit just the sorts of stressors that exacerbate
these vulnerabilities.
The findings of the present study are preliminary and in the most part ex-
ploratory, and further research is needed to develop and test predictions of its re-
lational model-based account of vulnerability. Future research should identify truly
“at risk” samples who score high on the vulnerability dimensions, study their re-
lational cognition and actual interpersonal behaviors, develop more specific hy-
potheses, test more rigorously the specificity of associations between vulnerability
dimensions and particular relational models, and employ more multi-method assess-
ment of both vulnerability and relational patterns. Nevertheless, the study offers
encouraging support for a new account of the interpersonal markers of psycholog-
744 Allen, Haslam, and Semedar

ical risk. Investigating the distinct and atypical ways in which people perceive and
construct their relationships is a neglected approach to the study of vulnerability.
However this relational approach is one that complements more established lines of
interpersonal vulnerability research that focuses on personality traits, information
processing biases, dysfunctional beliefs, and social skill deficits. Understanding the
role that cognition about relationships plays in vulnerability to mental disorders is
therefore a promising topic for further investigation.

REFERENCES

Allen, N. B., & Badcock, P. B. T. (2003). Social risk and depressed mood: Evolutionary, psychosocial,
and neurobiological perspectives. Psychological Bulletin, 129, 887–913.
Allen, N. B., Gilbert, P., & Semedar, A. (2004). Depression as an interpersonal strategy: The importance
of relational models. In N. Haslam (Ed.), Relational models theory: A contemporary overview (pp.
309–334). Mahwah, NJ: Lawrence Erlbaum and Associates.
Badcock, P. B. T., & Allen, N. B. (2003). Adaptive social reasoning in depressed mood and depressive
vulnerability. Cognition and Emotion, 17, 647–670.
Bartol, C. R. (1991). Predictive validation of the MMPI for small town police officers who fail. Profes-
sional Psychology; Research and Practice, 22, 127–132.
Benjamin, L. S. (1996). Interpersonal diagnosis and treatment of personality disorders (2nd ed.). New
York: Guilford.
Bentall, R. P. (2001). Social cognition and delusional beliefs. In P. W. Corrigan & D. L. Penn (Eds.),
Social cognition and schizophrenia (pp. 123–148). Washington, DC: American Psychological Asso-
ciation.
Blanchard, J. J., Mueser, K., & Bellack, A. S. (1998). Anhedonia, positive and negative affect, and social
functioning in schizophrenia. Schizophrenia Bulletin, 24, 413–424.
Broks, P. (1997). Brain, self, and others: The neuropsychology of social cognition. In G. Claridge (Ed.),
Schizotypy: Implications for illness and health (pp. 98–123) Oxford: Oxford University Press.
Caralis, D., & Haslam, N. (2004). Personality correlates of the relational models. Psychology and Psy-
chotherapy: Theory, Research and Practice, 77, 397–402.
Chapman, J. P., Chapman, L. J., & Kwapil, T. R. (1995). Scales for the measurement of schizotypy. In
A. Raine & T. Lencz (Eds.), Schizotypal personality (pp. 79–106). New York: Cambridge University
Press.
Chapman, L. J., Chapman, J. P., Kwapil, T. R., Eckblad, M., & Zinser, M. C. (1994). Putatively psychosis-
prone subjects 10 years later. Journal of Abnormal Psychology, 103, 171–183.
Chapman, L. J., Chapman, J. P., & Raulin, M. L. (1978). Body-image aberration in schizophrenia. Journal
of Abnormal Psychology, 87, 399–407.
Corrigan, P. W., & Toomey, R. (1995). Interpersonal problem solving and information processing in
schizophrenia. Schizophrenia Bulletin, 21, 395–404.
Coyne, J. C. (1976). Toward and interactional description of depression. Psychiatry, 39, 28–40.
Coyne, J. C., & Whiffen, V. E. (1995). Issues in personality as diathesis for depression: The case of
sociotropy-dependency and autonomy-self-criticism. Psychological Bulletin, 118, 358–78.
Depue, R. A., & Klein, D. N. (1988). Identification of unipolar and bipolar affective conditions in non-
clinical and clinical populations by the General Behavior Inventory. In D. L. Dunner, E. S. Gershon,
& J. E. Barrett (Eds.), Relatives at risk for mental disorders (pp. 179–202). New York: Raven Press.
Depue, R. A., Krauss, S., Spoont, M. R., & Arbisi, P. (1989). General Behavior Inventory identifica-
tion of unipolar and bipolar affective conditions in a non-clinical university population. Journal of
Abnormal Psychology, 98, 117–126.
Depue, R. A., Slater, J. F., Wolfstetter-Kausch, H., Klein, D. N., Goplerud, E., & Farr, D. (1981). A
behavioral paradigm for identifying persons at risk for bipolar depressive disorder: A conceptual
framework and five validation studies. Journal of Abnormal Psychology, 90, 381–437.
Dohr, K. B., Rush, A. J., & Bernstein, I. H. (1989). Cognitive biases and depression. Journal of Abnormal
Psychology, 98, 263–267.
Eckblad, M., & Chapman, L. J. (1983). Magical ideation as an indicator of schizotypy. Journal of Con-
sulting and Clinical Psychology, 51, 215–225.
Social Cognition and Relationships 745

Edwards, J., Jackson, H. J., & Pattison, P. E. (2002). Emotion recognition via facial expression and affec-
tive prosody in schizophrenia: A methodological review. Clinical Psychology Review, 22, 789–832.
Fiske, A. P. (1992). The four elementary forms of sociality: Framework for a unified theory of social
relations. Psychological Review, 99, 689–723.
Fiske, A. P. (1993). Social errors in four cultures: Evidence about the elementary forms of social relations.
Journal of Cross-Cultural Psychology, 24, 67–94.
Fiske, A. P., & Haslam, N. (1996). Social cognition is thinking about relationships. Current Directions in
Psychological Science, 5, 143–148.
Fiske, A. P., & Tetlock, P. E. (1997). Taboo tradeoffs: Reactions to transactions that transgress spheres
of exchange. Political Psychology, 18, 255–297.
Frith, C., & Corcoran, R. (1996). Exploring “theory of mind” in people with schizophrenia. Psychological
Medicine, 26, 521–530.
Goodnick, P. J., Fieve, R. R., Peselow, E., Schlegel, A., & Filippi, A. (1986). General Behavior Inventory:
Measurement of subclinical changes during depression and lithium prophylaxis. Acta Psychiatrica
Scandinavica, 73, 529–532.
Haslam, N. (1994). Categories of social relationship. Cognition, 53, 59–90.
Haslam, N. (2004). Research on the relational models: An overview. In N. Haslam (Ed.), Relational
models theory: A contemporary overview (pp. 27–57). Mahwah, NJ: Erlbaum.
Haslam, N., & Fiske, A. P. (1999). Relational models theory: A confirmatory factor analysis. Personal
Relationships, 6, 241–250.
Haslam, N., Reichert, T., & Fiske, A. P. (2002). Aberrant social relations in the personality disorders.
Psychology and Psychotherapy: Theory, Research and Practice, 75, 19–31.
Hooley, J. M. (1986). Expressed emotion and depression: Interactions between patients and high- versus
low-expressed-emotion spouses. Journal of Abnormal Psychology, 95, 237–46.
Hooley, J. M., Orley, J., & Teasdale, J. D. (1986). Levels of expressed emotion and relapse in depressed
patients. British Journal of Psychiatry, 148, 642–647.
Joiner, T. E. Jr., & Schmidt, N. B. (1998). Excessive reassurance seeking predicts depressive but not
anxious reactions to acute stress. Journal of Abnormal Psychology, 107, 533–537.
Joiner, T. E., Alfano, M. S., & Metalsky, G. I. (1992). When depression breeds contempt: Reassurance
seeking, self-esteem, and rejection of depressed college students by their roommates. Journal of
Abnormal Psychology, 101, 165–173.
Johnson, S. L., & Meyer, B. (2003). Psychosocial predictors of symptoms. In S. L. Johnson & R. Leahy
(Eds.), Psychological treatment of bipolar disorder. New York, NY: Guilford Press.
Kiesler, D. J. (1983). The 1982 interpersonal circle: A taxonomy for complementarity in human transac-
tions. Psychological Review, 90, 185–214.
Klein, D. N., & Depue, R. A. (1984). Continued impairment in persons at risk for bipolar affec-
tive disorder: Results of a 19-month follow-up study. Journal of Abnormal Psychology, 94, 115–
127.
Klein, D. N., Depue, R. A., & Slater, J. F. (1986). Inventory identification of cyclothymia: IX. Validation
in offspring of bipolar 1 patients. Archives of General Psychiatry, 43, 441–445.
Klein, D. N., Lewinsohn, P. M., & Seeley, J. (1996). Hypomanic personality traits in a community sample
of adolescents. Journal of Affective Disorders, 38, 135–143.
Kwapil, T. R. (1998). Social anhedonia as a predictor of the development of schizophrenia-spectrum
disorders. Journal of Abnormal Psychology, 107, 558–565.
Kwapil, T. R., Miller, M. B., Zinser, M. C., Chapman, J., & Chapman, L. J. (1997). Magical ideation and
social anhedonia as predictors of psychosis proneness: A partial replication. Journal of Abnormal
Psychology, 106, 491–495.
Lewinsohn, P. M., Steinmetz, J. L., Larson, D. W., & Franklin, J. (1981). Depression related cognitions:
Antecedent or consequence? Journal of Abnormal Psychology, 91, 213–219.
Lyon, H. M., Bentall, R. P., & Startup, M. (1999). Social cognition and the manic defense: Attributions,
selective attention, and self-schema in bipolar affective disorder. Journal of Abnormal Psychology,
108, 273–282.
Mathews, A., Ridgeway, V., & Williamson, D. A. (1996). Evidence for attention to threatening stimuli in
depression. Behaviour Research and Therapy, 34, 695–705.
Mishlove, M., & Chapman, L. J. (1985). Social anhedonia in the prediction of psychosis proneness. Jour-
nal of Abnormal Psychology, 94, 384–396.
Monroe, S. M., Rohde, P., Seeley, J. R., & Lewinsohn, P. M. (1999). Life events and depression in ado-
lescence: Relationship loss as a prospective risk factor for first-onset of major depressive disorder.
Journal of Abnormal Psychology, 108, 606–614.
Neale, J. M. (1988). Defensive functions of manic episodes. In T. F. Oltmanns & B. A. Maher (Eds.),
Delusional beliefs (pp. 138–156). New York: Wiley.
746 Allen, Haslam, and Semedar

Penn, D. L., Corrigan, P. W., Bentall, R. P., Racenstein, J. M., & Newman, L. S. (1997). Social cognition
in schizophrenia. Psychological Bulletin, 121, 114–132.
Penn, D. L., Spaulding, W. D., Reed, D., & Sullivan, M. (1996). The relationship of social cognition
towards behavior in chronic schizophrenia. Schizophrenia Research, 20, 327–335.
Rudolf, K. D., Hammen, C., Burge, D., Lindberg, N., Herzberg, D., & Daley, S. E. (2000). Toward an
interpersonal life-stress model of depression: The developmental context of stress generation. De-
velopment and Psychopathology, 12, 215–234.
Sacco, W. P. (1999). A social cognitive model of interpersonal processes in depression. In T. Joiner &
J. Coyne (Eds.), The interactional nature of depression: Advances in interpersonal approaches (pp.
329–362). Washington, DC: American Psychological Association.
Segrin, C., & Abramson, L. Y. (1994). Negative reactions to depressive behaviors: A communication
theories analysis. Journal of Abnormal Psychology, 103, 655–668.
Sloman, L., & Gilbert, P. (2000). Subordination and defeat: An evolutionary approach to mood disorders
and their therapy. Mahwah, NJ: Lawrence Erlbaum Associates.