Psychotherapy Volume 38/Winter 2001/Number 4
ATTACHMENT STYLE
BJORN MEYER PAUL A. PILKONIS
Louisian State University Western Psychiatric Institute and Clinic
This article provides a brief review of Whereas early studies focused primarily on at-
attachment theory and recent research tachment during childhood (e.g., Ainsworth,
Blehar, Waters, & Wall, 1978), recent years have
studies on its implications for individual witnessed a proliferation of research on adult at-
psychotherapy. Attachment theory offers tachment, especially in the area of romantic rela-
a conceptual framework that helps tionships (e.g., Kazan & Shaver, 1987, 1994;
illuminate how past experiences with Simpson & Rholes, 1998). Psychotherapy re-
caregivers might influence current searchers have also become active in exploring
attachment theory's implications (cf. Mace &
transactions between therapist and Margison, 1997; Strauss, 2000).
patient. Both patients and therapists may Adult attachment styles describe people's com-
form internal working models that are fort and confidence in close relationships, their
based, in pan, on early experiences of fear of rejection and yearning for intimacy, and
interpersonal responsiveness. Such their preference for self-sufficiency or interper-
working models are reflected in secure, sonal distance (for more technical definitions see
Cassidy & Shaver, 1999). Attachment styles are
anxious-ambivalent, and anxious- formed in response to real-life experiences with
avoidant attachment styles, which caregivers and other people, and they reflect men-
describe whether patients (and tal representations ("internal working models") of
therapists) tend to be comfortable and others, of oneself in relation to others, and of
confident in relationships, fearful of relationships in general.
There are at least two approaches to the mea-
abandonment, or defensively separate. surement of attachment in adults. The first ap-
Recent studies showed that attachment proach is based on the idea that the manner in
styles in the context of therapy can be which adults speak about past and present rela-
measured reliably and related to tionships provides clues about their internal work-
therapeutic process (e.g., the alliance) ing models. This "narrative approach" has culmi-
and outcome (e.g., treatment response). nated in the assessment of adult attachment via
the analysis of semistructured interviews; for ex-
ample, the Adult Attachment Interview (AAI;
Since the publication of John Bowlby's (1969) George, Kaplan, & Main, 1985; Solomon &
seminal works more than 30 years ago, attach- George, 1999). In this procedure, spoken narra-
ment theory has inspired an impressive amount tives about attachment experiences are tran-
of research and theoretical speculation on the scribed and analyzed, yielding classifications of
function and structure of close relationships. attachment "states of mind."
A second approach to the measurement of adult
attachment is the use of self-report instruments
(Brennan, Clark, & Shaver, 1998; Shaver, Bel-
Portions of this article are adapted from J. C. Norcross
(Ed.), Psychotherapy relationships that work. New York: Ox-
sky, & Brennan, 2000). These questionnaires are
ford University Press. often based on Hazan and Shaver's work (Kazan
Correspondence regarding this article should be addressed & Shaver, 1987, 1994); they, in turn, based their
to Bjorn Meyer, Ph.D., Department of Psychology, 236 Au- dimensions on Ainsworth's classification of se-
dubon Hall, Louisiana State University, Baton Rouge, LA cure, ambivalent, and avoidant attachment (Ains-
70803. E-mail: bmeyer@[Link] worth et al., 1978). Self-reports of adult attach-
466
Attachment Style
ment converge to a substantial degree, and it is though securely attached patients tended to func-
now possible to use brief (36 to 40 items) ques- tion more optimally than others at both admission
tionnaires containing items culled from many and discharge, those classified as dismissive ex-
measures (Fraley, Waller, & Brennan, 2000). hibited the greatest amount of relative improve-
Evidence suggests that self-reported romantic at- ment over the course of treatment.
tachment styles can be captured by two continu- Attachment styles were also examined as po-
ous dimensions—anxiety and avoidance—that tential predictors of treatment response in a study
emerge from factor analyses of attachment- by Pilkonis and colleagues (e.g., Meyer, Pil-
related item sets (Brennan, Clark, & Shaver, konis, Proietti, Heape, & Egan, in press). Among
1998). The anxious dimension is thought to reflect 149 patients (80% outpatients) at a psychiatric
people's mental model of the self; the avoidant hospital in Pittsburgh, attachment dimensions
dimension, people's model of others (cf. Brennan were assessed with the use of a semistructured
etal., 1998). Self-reports of romantic attachment psychosocial interview, followed by consensus
only moderately with measures of the AAI-de- ratings of several attachment prototypes (cf. Pil-
rived attachment states of mind (Shaver et al., konis, 1988). Over the course of 6 months, secure
2000), suggesting that these instruments capture attachment predicted relative improvement, as
somewhat different facets of attachment. measured by a global rating of psychosocial func-
In addition to attachment-related narratives and tioning, whereas other attachment dimensions
self-report instruments, questionnaires measuring were unrelated to outcome.
attachment in the context of psychotherapy have In a third study on the link between attachment
been developed (Hoger, 1999; Mallinckrodt, and treatment outcome, Mosheim and colleagues
Gantt, & Coble, 1995). Therapy contains many of (2000) used a German version of Pilkonis's
the features that characterize the child-caregiver (1988) prototype methodology (Strauss, Lobo-
relationship, including, for example, the thera- Drost, & Pilkonis, 1999) to classify attachment
pist's comforting presence, the potential for emo- among 65 inpatients diagnosed with eating disor-
tion regulation, and the provision of a "secure ders, mood disorders, or anxiety disorders. Se-
base" that allows for the safe exploration of diffi- cure attachment emerged as a significant predictor
cult psychological territory (Bowlby, 1988). In of therapy-goal attainment over an average treat-
a way that is similar to how children attach to ment duration of 7 weeks. Patients who were
caregivers, and how adults attach to romantic rated as comfortable and confident in past and
partners, patients can be said to attach to their present relationships (i.e., securely attached pa-
therapist. tients) tended to benefit more than others from
treatment.
Research Review Differences in measurement, in treatment type,
Recent studies have begun to examine the mul- and in patient functioning make it difficult to di-
tifaceted relationships between attachment styles rectly compare the results from these three stud-
and psychotherapy. Our review focuses on links ies. Nonetheless, in all three studies, securely
(a) between patient attachment and treatment out- attached patients were functioning best at the end
come, (b) between patient attachment and the of treatment. Under some conditions, securely
therapeutic alliance, and (c) between therapist at- attached patients may benefit more than others
tachment and treatment process and outcome. from treatment (e.g., Mosheim et al., 2000), but
under other conditions—when patients are more
Patient Attachment and Outcome severely impaired and receiving longer-term in-
The question of whether patients' attachment tensive treatment—those with dismissive attach-
styles influence psychotherapy outcome was ad- ment may show the best relative trajectory (Fo-
dressed in three recent studies. In their longitudi- nagy et al., 1996). Securely attached patients
nal study at the Cassel Hospital in London, Fo- might engage productively in most therapy set-
nagy and his colleagues (1996) administered the tings, whereas those with dismissing attachment
AAI to 82 inpatients, most of whom were diag- may require concentrated or targeted interven-
nosed with mood disorders and with severe per- tions, helping them overcome their characteristic
sonality disorders. All patients received individ- detachment. Once they do connect emotionally
ual and group psychoanalytic therapy, with an with a therapist, however, improvement might be
average duration of more than 9 months. Al- all the more dramatic.
467
B. Meyer & P. A. Pilkonis
It may be difficult to disentangle the effects of affect outcome per se, group differences were
patient attachment on treatment outcome, in part observed in the formation of the therapeutic alli-
because therapists respond flexibly to patient ance. Whereas securely attached patients formed
styles, changing interventions on a moment-to- relatively stable alliances throughout treatment,
moment basis to accommodate emerging clinical those with preoccupied attachment reported a
needs (cf. Stiles, Honos-Webb, & Surko, 1998). poor alliance in the middle but very strong alli-
Patients with dismissive attachment may require— ance in the later stages of therapy. By contrast,
and perhaps elicit—active interventions that fa- patients with dismissing attachment reported dete-
cilitate emotional expression and connection. riorating alliances toward the end of therapy. (See
Those with preoccupied attachment may require Zuroff et al., 2000, for a conceptually similar
supportive interventions that help contain over- finding: perfectionistic patients failed to form
whelming emotions. Patients with secure attach- strong alliances toward the end of treatment.)
ment, in turn, may be judged as able to work Another study that examined how patient at-
productively without "customizing" interventions. tachment style contributes to the therapeutic alli-
ance was conducted by Mallinckrodt et al. (1995),
Patient Attachment and Alliance who developed the Client Attachment to Thera-
Additional research has demonstrated that pa- pist Scale (CATS) and examined its correlations
tients' attachment styles also relate to therapy pro- with a self-report of therapeutic alliance. The
cess on an intermediate level of analysis. This CATS is a 36-item questionnaire measuring three
research has focused on linkages between attach- facets of the patient's attachment to the therapist:
ment styles and the therapeutic alliance, which the patient's perception of the therapist as emo-
in itself is a potent predictor of outcome (e.g., tionally supportive or responsive (secure), the pa-
Orlinsky, Grawe, & Parks, 1994). tient's appraisal of the therapist as disapproving
In a study conducted in London, Eames and Roth or rejecting (avoidant), and the patient's wish to
(2000) investigated the relationships between pa- be closer or have a more intimate relationship
tients' self-reported attachment styles and the with the therapist (preoccupied/merger).
quality of the alliance. Among 30 adult outpa- Among 129 outpatients from the Pacific North-
tients, the majority of whom received cognitive- west, both the secure and fearful-avoidant CATS
behavioral therapy, patient self-reports of attach- scales correlated with various aspects of the alli-
ment correlated with the quality of the therapeutic ance (Mallinckrodt et al., 1995). For example,
alliance and with ruptures in the alliance. securely attached patients, who perceive their
Whereas securely attached patients tended to form therapist as responsive, accepting, and caring,
effective alliances, fearful (anxious-avoidant) pa- were more likely to endorse a strong alliance. In
tients tended to experience problems. Interest- fact, the magnitude of correlations (around .80)
ingly, some evidence also suggested that both the suggested that some CATS scales share a com-
preoccupied and dismissing attachment dimen- mon conceptual core with the working alliance
sions were associated with positive alliance rat- (cf. Robbins, 1995). Interestingly, there was no
ings. Patients who yearn for intimacy and fear relationship between preoccupied attachment and
abandonment might strive with particular persis- a total alliance scale. Thus, yearning "to be at
tence to establish a close alliance, given their one" (Mallinckrodt et al., 1995, p. 311) with the
concern over potential rejection. In contrast, pa- therapist is no guarantee for successful therapy.
tients with dismissing styles might defensively Indeed, in their frantic efforts to avoid rejection,
deny problems in the alliance or establish only a patients with preoccupied attachment may try to
superficial relationship while remaining reluctant submissively please and appease their therapist
to connect and self-disclose on a genuine, per- without engaging in the more risky task of identi-
sonal level (Eames & Roth, 2000). fying and openly discussing difficult personal
Associations between attachment and the thera- problems (i.e., agreeing on therapeutic tasks
peutic alliance were also examined in a study and goals).
by Finnish psychotherapy researchers Kanninen, In summary, evidence is converging to suggest
Salo, and Punamaki (2000). Among 36 Palestin- that patients' attachment styles influence im-
ian political ex-prisoners, attachment was mea- portant parameters of the therapeutic alliance. Pa-
sured with a newly devised paper-and-pencil in- tients' mental representations of the self and others
strument. Although attachment styles did not in relationships may predict how their therapists
468
Attachment Style
respond to them within sessions (Hardy et al., 1999) complementarity between case managers' and pa-
and how effective they are at establishing a thera- tients' attachment styles worked best: preoccu-
peutic alliance (e.g., Kanninen, Salo, & Punamaki, pied patients fared best when they worked with
2000; Mallinckrodt et al., 1995). dismissing case managers, and—vice versa—
dismissing patients fared best when they worked
Therapist Attachment with preoccupied case managers.
Although patients' working models are im- Consistent with previous research on client-
portant factors in psychotherapy, it may be clinician matches, dissimilarities between pa-
equally important to consider therapists' attach- tients' and therapists' interpersonal styles ap-
ment styles. For example, clinicians with secure peared advantageous, supporting the view that
attachment may handle ruptures in the relation- patients benefit from interventions that counteract
ship with patients more easily, whereas those with their problematic style of relating to others.
anxious or preoccupied attachment—given their Overly emotional patients may require emotion-
characteristic fear of rejection—may have diffi- containing interventions, whereas emotionally de-
culties in this regard. tached patients may need interventions that facili-
In Rubino, Barker, Roth, and Fearon's (2000) tate affective expression and connection (cf.
study, this hypothesis was tested and partially Hardy et al., 1999; Stiles et al., 1998). To the
supported. Seventy-seven trainee-therapists from degree that patients and therapists happen to be
England viewed videotaped vignettes of simu- matched in such complementary constellations,
lated ruptures in the therapeutic alliance and were therapy may have a natural advantage. Tyrrell
asked to respond as if they were interacting with and colleagues (1999) noted appropriately that
actual patients. Each vignette depicted a patient caution is warranted in interpreting their findings,
with either a secure, dismissing, fearful, or preoc- and that studies with different patient and thera-
cupied attachment, and therapists were asked to pist populations, alternative measurement ap-
complete self-reports of attachment in close rela- proaches, and longitudinal designs are needed to
tionships. Independent raters judged the thera- establish the replicability and generalizability of
pists' empathy and depth of interpretation in re- such interactive effects.
sponse to the vignettes. Although results were
complex, and some hypotheses were not sup- Summary
ported, findings suggested that therapists scoring In summary, a growing literature suggests that
higher on an anxious-attachment dimension both patients' and therapists' attachment styles
tended to respond less empathically. This was influence important aspects of process and out-
especially true in response to patients with secure come in individual adult psychotherapy. People's
and dismissing attachment. Based on the results mental representations of themselves and of at-
of their study, Rubino and colleagues (2000) con- tachment figures—whether they are reflected in
cluded that "more anxious therapists might inter- relationship narratives or in self-reports—make
pret ruptures as an indication of their patients' a difference in terms of how therapists respond
intention to leave therapy, and their own sensitiv- to patients and in terms of treatment response.
ity toward abandonment might diminish their Not surprisingly, much of the evidence points to
ability to be empathic" (p. 416). the benefits of a secure attachment style. Patients
Therapists' attachment was also considered in who feel comfortable in relationships and are not
a study by Tyrrell, Dozier, Teague, and Fallot overly concerned about rejection tend to form sta-
(1999). Adult attachment interviews were admin- ble and effective alliances (Eames & Roth, 2000;
istered to a sample of 54 patients with severe Kanninen et al., 2000) and may thus benefit more
mental disorders and to their 21 case managers. from treatment (Meyer et al., in press; Mosheim,
Whereas most patients were rated as insecurely 2000). At the same time, greater relative improve-
attached, most case managers were rated as se- ments may sometimes occur for patients with in-
cure or autonomous. In addition, this study re- secure forms of attachment. This is illustrated
vealed interactions between patients' and case by Fonagy and colleagues' (1996) finding that
managers' attachment styles when these styles dismissing patients improved more than others in
were characterized along a preoccupied versus psychoanalytic treatment.
dismissing dimension (i.e., "activating vs. deacti- Research is also beginning to demonstrate that
vating attachment"). Findings suggested that therapists' attachment influences treatment. Ther-
469
B. Meyer & P. A. Pilkonis
apists who tend to worry about rejection may be pulls but also countering them by complementary
prone to respond less empathically to some pa- therapeutic action. Preoccupied patients pull for
tients (Rubino et al., 2000), at least in analogue emotional-experiential interventions, but they
research, and complementary matches between may benefit as well from cognitive-behavioral
therapists' and patients' attachment may affect strategies that help modulate overwhelming feel-
the quality of the alliance (Tyrrell et al., 1999). ings. Similarly, avoidant patients pull for rational-
What conclusions might therapists draw from this cognitive interventions but may benefit also from
research for clinical practice? strategies that facilitate emotional engagement
(Hardy et al., 1999). Therapists ought to recog-
Therapeutic Practices nize the potential influence of patients' attach-
ment styles on the quality of the therapeutic alli-
Assess Patients' Attachment Styles ance. Building a strong alliance may require an
Several reliable and valid measures of adult individualized balancing of patient attachment
attachment have been developed in recent years, style with congruent and complementary thera-
and many of these instruments can be used effi- peutic action.
ciently in clinical practice. These instruments in-
clude Brennan, Clark, and Shaver's (1998) Expe- Consider Modification of Attachment Style as a
riences in Close Relationships questionnaire and Treatment Goal
Mallinckrodt et al.'s (1995) Client Attachment Because there are no empirical data on this
to Therapist Questionnaire. These questionnaires question, therapists would be well advised to ex-
permit clinicians to obtain quantitative estimates ploit their existing repertoire of interventions in
of patients' attachment styles in romantic relation- the pursuit of attachment-related treatment goals.
ships and in the context of therapy. For instance, if a therapist assesses a patient's
Patient attachment styles suggest specific con- attachment style, identifies a preoccupied pattern,
stellations of internal working models; for exam- and sets the goal of bolstering the inadequate self-
ple, preoccupied attachment suggests a negative model, it would seem advisable to employ any
model of the self combined with a positive model interventions serving this goal. Behavioral thera-
of others. Given the assumption that positive self pists might reinforce desirable in-session behav-
and other models are most desirable, therapists ior (e.g., moments of mature, autonomous action)
who identify problematic working models gain by differentially responding to such events (cf.
information about specific treatment goals. For Kohlenberg & Tsai, 1991). Cognitively oriented
instance, in working with a patient showing signs therapists might aim to challenge or refute nega-
of dismissing-avoidant attachment (positive self tive attachment-related thoughts. Regardless of
and negative other model), the goal would be to the specific strategies, the therapeutic goal could
increase the positivity of the "other" model. be derived from attachment considerations, and
Given the importance of cultural differences the effectiveness of intervention strategies would
(see Rothbaum, Weisz, Pott, Miyake, & Morelli, be judged in reference to progress toward those
2000), however, therapists should consider that goals.
it is not always advisable to facilitate autonomy
and individuation. Family and cultural context Assess Supervisees' Attachment Styles
needs to be evaluated, and attachment-related Clinical supervision might be enhanced by in-
therapy goals must be adjusted accordingly. For cluding assessments of clinicians' attachment
example, in cultures where communal integration styles. This information could then be used to
is valued, dependency-related traits may not be match therapists with patients. The benefits of
maladaptive (Rothbaum et al., 2000). such matching have not been demonstrated empir-
ically, however, and it would be premature to
Address Patients' Attachment Style as it Relates adopt this in routine clinical practice. But supervi-
to Therapeutic Alliance sors should recognize as well that therapists' at-
Patients present to therapy with different interper- tachment can influence interventions on a micro-
sonal or attachment style, which then pull for differ- scopic level. For example, therapists with anxious
ent interventions (Hardy, Stiles, Barkham, & forms of attachment might tend to respond less
Startup, 1998; Hardy, 1999). Being adequately re- empathically than others (Rubino et al., 2000).
sponsive to these styles means "giving in" to these It would seem advisable, then, to measure and
470
Attachment Style
monitor therapists' as well as clients' attachment Questionnaire on Client Expectations: A method for as-
sessing attachment styles among psychotherapy patients].
over the course of therapy, and to suggest adjust- Psychotherapeut,44, 159-166.
ments in interventions accordingly. These adjust- KANNINEN, K., SALO, J., & PUNAMAKI, R. L. (2000). Attach-
ments might involve establishing optimal comple- ment patterns and working alliance in trauma therapy for
mentarity between patients' and therapists' victims of political violence. Psychotherapy Research,
interpersonal styles, such that preoccupied pa- 10, 435-449.
KOHLENBERG, R. J., & TSAI, M. (1991). Functional Analyti-
tients' overwhelming emotions are contained, and cal Psychotherapy: Creating intense and curative therapeu-
dismissive patients' emotional disengagement is tic relationships. New York: Plenum.
effectively challenged. MACE, C., & MARGISON, F. (1997). Attachment and psycho-
therapy: An overview. British Journal of Medical Psychol-
ogy, 70, 209-215.
References MALLINCKRODT, B., GANTT, D. L., & COBLE, H. M. (1995).
AINSWORTH, M. D. S., BLEHAR, M. C., WATERS, E., & Attachment patterns in the psychotherapy relationship: De-
WALL, S. (1978). Patterns of attachment: A psychological velopment of the Patient Attachment to Therapist Scale.
study of the strange situation. Hillsdale, NJ: Lawrence Journal of Counseling Psychology, 42, 307-317.
Erlbaum. MEYER, B., PILKONIS, P. A., PROIETTI, J. M., HEAPE, C. L.,
BOWLBY, J. (1969). Attachment and loss. New York: Basic & EGAN, M. (2001). Attachment styles and personality
Books. disorders as predictors of symptom course. Journal of Per-
BOWLBY, J. (1988). A secure base: Parent-child attachment sonality Disorders, 15, 371-389.
and healthy human development. New York: Basic Books. MOSHEIM, R., ZACHHUBER, U., SCHARF, L., HOFMANN, A.,
BRENNAN, K. A., CLARK, C. L., & SHAVER, P. R. (1998). KEMMLER, G., DANZL, C., KINZE, J., BIEBL, W., & RICH-
Self-report measurement of adult attachment: An integ- TER, R. (2000). Bindung und Psychotherapie: Bindungs-
rative overview. In J. A. Simpson & W. S. Rholes (Eds.), qualitaet und interpersonale Probleme von Patienten als
Attachment theory and close relationships (pp. 46-76). moegliche Einflussfaktoren auf das Ergebnis stationaerer
New York: Guilford. Psychotherapie. [Quality of attachment and interpersonal
CASSIDY, J.,, & SHAVER, P. R. (1999). Handbook of attach- problems as possible predictors of inpatient-therapy out-
ment: Theory, research, and clinical applications. New come]. Psychotherapeut, 45, 223-229.
York: Guilford. ORUNSKY, D. E., GRAWE, K., & PARKS, B. K. (1994). Process
EAMES, V., & ROTH, A. (2000). Patient attachment orientation and outcome in psychotherapy—noch einmal. In A. E. Ber-
and the early working alliance: A study of patient and gin & S. L. Garfield (Eds.), Handbook of psychotherapy and
therapist reports of alliance quality and ruptures. Psycho- behavior change (pp. 270-376). New York: Wiley.
therapy Research, 10, 421-434. PILKONIS, P. A. (1988). Personality prototypes among de-
FONAGY, P., LEIGH, T., STEELE, M., STEELE, H., KENNEDY, pressives: Themes of dependency and autonomy. Journal
R., MATTOON, G., TARGET, M., & GEIBER, A. (1996). The of Personality Disorders, 2, 144-152.
relation of attachment status, psychiatric classification, and ROBBINS, S. B. (1995). Attachment perspectives on the coun-
response to psychotherapy. Journal of Consulting and Clini- seling relationship: Comment on Mallinckrodt, Gantt, and
cal Psychology, 64, 22-31. Coble. Journal of Counseling Psychology, 42, 318-319.
FRALEY, R. C., WALLER, N. G., & BRENNAN, K. G. (2000). ROTHBAUM, F., WEISZ, J., POTT, M., MIYAKE, K., & Mo-
An item response theory analysis of self-report measures RELLI, G. (2000). Attachment and culture: Security in the
of adult attachment. Journal of Personality and Social Psy- United States and Japan. American Psychologist, 55, 1093-
chology, 78, 350-365. 1104.
GEORGE, C., KAPLAN, N., & MAIN, M. (1985). Adult Attach- RUBINO, G., BARKER, C., ROTH, T., & FEARON, P. (2000).
ment Interview (2nd ed.). Berkeley, CA: University of Cali- Therapist empathy and depth of interpretation in response
fornia at Berkeley. to potential alliance ruptures: The role of therapist and
HARDY, G. E., ALDRTOGE, J., DAVIDSON, C., ROWE, C., RE- patient attachment styles. Psychotherapy Research, 10, 408-
ILLY, S., & SHAPIRO, D. A. (1999). Therapist respon- 420.
siveness to patient attachment styles and issues observed SHAVER, P. R., BELSKY, J., & BRENNAN, K. A. (2000). The
in patient-identified significant events in psychodynamic- adult attachment interview and self-reports of romantic at-
interpersonal psychotherapy. Psychotherapy Research, 9, tachment: Associations across domains and methods. Per-
36-53. sonal Relationships, 7, 25-43.
HARDY, G. E., STILES, W. B., BARKHAM, M., & STARTUP, SIMPSON, J. A., & RHOLES, W. S. (1998). Attachment theory
M. (1998). Therapist responsiveness to client interpersonal and close relationships. New York: Guilford.
styles during time-limited treatments for depression. Jour- SOLOMON, J., & GEORGE, C. (1999). The measurement of
nal of Consulting and Clinical Psychology, 66, 304-312. attachment security in infancy and childhood. In J. Cassidy
HAZAN, C., & SHAVER, P. R. (1987). Romantic love concep- & P. R. Shaver (Eds.), Handbook of attachment: Theory,
tualized as an attachment process. Journal of Personality research, and clinical applications (pp. 287-316). New
and Social Psychology, 52, 511-524. York: Guilford.
HAZAN, C., & SHAVER, P. R. (1994). Attachment as an orga- STILES, W. B., HONOS-WEBB, L., & SURKO, M. (1998). Re-
nizational framework for research on close relationships. sponsiveness in psychotherapy. Clinical Psychology: Sci-
Psychological Inquiry, 5, 1-22. ence and Practice, 5, 439-458.
HOGER, D. (1999). Der Bielefelder Fragebogen zu Klienten- STRAUSS, B. M. (2000). Attachment theory and psychother-
erwartungen (BFKE). Ein Verfahren zur Erfassung von apy research: Editor's introduction to a special section.
Bindungsstilen bei Psychotherapiepatienten. [The Bielefeld Psychotherapy Research, 10, 381-389.
471
B. Meyer & P. A. Pilkonis
STRAUSS, B. M., LOBO-DROST, A. I., & PILKONIS, P. A. tachment states of mind. Journal of Consulting and Clinical
(1999). Einschatzung von Bindungsstilen bei Erwachsenen: Psychology, 67, 725-733.
Erste Erfahrungen mil der deutschen Version einer Proto- ZUROFF, D. C., BLATT, S. J., SOTSKY, S. M., KRUPNICK,
typenbeurteilung. Zeitschrift fur Klinische Psychologic, J. L., MARTIN, D. J., SANISLOW, C. A., & SIMMENS, S.
Psychiatric und Psychotherapie, 47, 347-364. (2000). Relation of therapeutic alliance and perfectionism
TYRRELL, C. L., DOZIER, M., TEAGUE, G. B., & FALLOT, to outcome in brief outpatient treatment of depression. Jour-
R. D. (1999). Effective treatment relationships for persons nal of Consulting and Clinical Psychology, 68, 114-124.
with serious psychiatric disorders: The importance of at-
472