20 Counselling Psychology Review, Vol. 26, No. 1, March 2011 © The British Psychological Society – ISSN 0269-6975
Aim: The present study examined the view of the therapist as an attachment figure and the relationship between client adult attachment in close relationships and client attachment to the therapist.
Method: 27 clients in ongoing therapy were asked to complete measures of adult attachment and client attachment to their therapists. In addition, five participants were interviewed with regard to their perceptions of their therapist and the therapeutic relationship.
Findings: Quantitative analysis suggested that client attachment may be activated in the context of the therapeutic relationship. In particular, ‘Need for Approval’ was significantly associated with both ‘Fearful/Avoidant’ and ‘Preoccupied/Merger attachment to the therapist’; ‘Preoccupation in Relationships’ was positively correlated to ‘Preoccupied/Merger attachment to therapist’. Qualitative findings further suggested that the therapist serves as an attachment figure for clients
Implications for practice: The therapeutic relationship contains features of an attachment relationship and therapists are often seen by clients as a secure basis for exploration. In this context, the behaviour that clients present in therapy is seen as meaningful and its exploration can contribute to the deconstruction and re-appraisal of internal working models of self and others. Implications for the practice of counselling psychology are discussed.
Keywords: client attachment; therapist as attachment figure; transference; psychotherapy.
Original Research Paper
The therapeutic relationship from an attachment theory perspective Marina Christina Skourteli & Clare Lennie
CCORDING to attachment theory, infants develop relational patterns directly in response to the caregiver’s availability and consistency with the view to maximise closeness, security and ultimately survival (Bowlby, 1988). Attachment patterns impact significantly processes of affect regulation and cognitive processing (Goldberg, 2000), the development of internal working models (Bretherton & Munholland, 2008) and inner representations and are further implicated in adult psychopathology and psychotherapy (Bateman & Fonagy, 2004; Dozier et al., 2008). Attachment patterns are thought to persist throughout life and to be activated by any close relationship, including the therapeutic relationship. Attachment theory has marked a significant shift from a one-person to a two-person psychology and has had a profound influence on psychological and psychotherapy research (Beebe, Jaffe & Lachmann, 1992; Stephen & Pelham, 2000). By adopting an attachment theory perspective, the present study aims to understand the client’s experience of the therapist and the emerging dynamic between them as meaningful in the context of the client’s past and current relationships. This relational framework seems particularly relevant to the practice and knowledge-base of counselling psychology as it de-pathologises client difficulties and highlights the role of the therapist, ultimately promoting reflective and ethical practice. Within the context of attachment theory, both client and therapist are thought to contribute to processes of transference and countertransference and the emerging dynamic in therapy, yet relatively few studies have explored client attachment within the therapeutic relationship (Parish & Eagle, 2003; Woodhouse et al.,
A Counselling Psychology Review, Vol. 26, No. 1, March 2011 21 2003). Overall, counselling psychological research seems to employ mostly qualitative methods whilst few attempts have been made to systematically assess the client-therapist dynamic (Mallinckrodt, Gantt & Coble, 1995). The present research uses a mixed methodology to explore the ways in which client attachment orientation may influence perceptions of the therapist and the therapeutic relationship. Within this context, my research questions are formulated as follows: 1. How is client attachment orientation activated or enacted within the therapeutic relationship? 2. Is there a relationship between client adult attachment in close relationships and client attachment to the therapist? 3. How closely does the therapist fit the role of an attachment figure?
Background and literature review Self and Object representations Attachment theory accounts for the ways in which early child-parent interactions become abstracted into inner representations; in other words, it maintains that everything that is intrapsychic was once interpersonal (Maroda, 2010). From early studies like Ainsworth et al.’s (1978), it emerged that mere physical absence of the parent did not appear to be key to separation anxiety; rather it was the child’s evaluation of the mother’s departure on the basis of previous experience of her that was central for understanding responses to separations. Internal working models emerge as early attachment-related experiences become transformed into inner representations (Grossman & Grossman, 1991; Bretherton & Munholland, 2008).
Transference Internal working models are essentially the mechanisms by which early attachment experiences are carried forward and organise subsequent social experiences. Within this context, Westen and Gabbard (2002) discuss the concept of transference in terms of inner representations existing as potentials for activation. Internal representations of self and others exist in a latent state awaiting to be activated with varying intensity either consciously or unconsciously, particularly so within close relationships. Within therapy, transference is viewed as a template of early experience that can influence the client’s emotional relationship with the therapist (Casement, 1991; Bateman & Holmes, 1995). Mallinckrodt (2000) maintains that early experience is central not only to the development of schemas (implicated in view of oneself and others, conflict resolution style, affect regulation and cognitive appraisal) but further to the perception of events and people as stressful, rejecting or supportive. The above parameters are central to interpersonal processes (or themes) within relationships, including the therapeutic relationship (Luborsky & Crits-Christoph, 1998; Connoly et al., 2000). In considering interpersonal processes, Mallinckrodt (2000) suggests that people’s attachment style and thus their relatedness dispositions may be interacting with that of others in ways that maintain existing relationship patterns, posing thus interesting implications for the transference-countertransference dynamic in therapy.
Attachment to the therapist The attachment organisation established in early life appears to be robust and can be activated by any close relationship, including the therapeutic relationship. The counselling relationship contains many features which may activate the client’s ingrained expectations; similar to a caregiver, the therapist is emotionally available, offers a comforting presence, affect regulation and a sense of a secure base from which the client can explore inner experience (Holmes, 1999). Attachment concepts can provide considerable insight into what happens in therapy, particularly the relationship aspects of therapeutic change (Bender, Farber & Geller, 1997; Woodhouse et al., 2003, Jordan, 2007). Mallinckrodt, Gantt and Coble (1995) developed and validated the Client Attachment to Therapist Scale (CATS), which explores the ways clients relate to their therapist. In their study clients who scored high on the CATS Secure subscale perceived their therapists as emotionally responsive, accepting and promoting a secure base from which they were able to explore their emotional experience; these clients seemed to have positive working models of self and others. Clients who scored high on the Preoccupied-Merger subscale displayed a desire for dissolution of boundaries in the therapeutic relationship. These clients appeared preoccupied with their therapist and had a tendency to become dependent. Overall, preoccupied clients seemed to maintain a negative working model of themselves but a positive working model of others. Finally, clients who scored high on the Avoidance subscale tended to mistrust their therapists and were fearful of rejection. These clients reported a sense of alienation, mistrust and hopelessness and seemed to maintain negative working models of both self and others. Mallinckrodt later tied the above dimensions of the CATS with constructs of anxiety and avoidance with hyperactivation being reflective of anxiety and deactivation of avoidance (Mallinckrodt, 2000). Parish and Eagle (2003) further explored the ways in which therapists function as attachment figures for clients and concluded that the relationships formed in long-term psychotherapy share many qualities of an attachment relationship. Mallinckrodt, Porter and Kivlighan (2005) explored the role of attachment in therapy and in particular the relationship between client attachment to therapist, depth of in-session exploration and object relations. Consistent with the view of the therapist as a secure base, results indicated that security in attachment to the therapist was significantly associated with greater session depth and smoothness. Insecurity in attachment was mirrored within the therapeutic relationship; avoidance was associated with weaker working alliances, a sense of alienation and social incompetence (Mallinckrodt, Porter & Kivlighan, 2005).
Assessment of adult attachment Narrative measures Narrative approaches have emerged from the developmental and psychoanalytic traditions within psychology and are broadly considered to be tapping into unconscious states of mind with regard to attachment (Hesse, 2008; Bifulco, 2002; Shaver & Mikulincer, 2002). There seems to be a close relationship between attachment classifications in childhood and narrative style in adulthood and narrative assessment in attachment is based on the notion that representational processes are reflected in language (Crowell, Fraley & Shaver, 2008). Secure attachment in adulthood is thought to be mediated by a single, consistent working model, which is communicated through coherent and collaborative narratives characterised by meta-cognitive monitoring (Bateman & Fonagy, 2004). On the other hand, inconsistent or incoherent narratives are thought to be indicative of multiple, often contradictory models of the same aspect of reality, consistent with the perceived insecurity and inconsistency underlying the experience of early attachment figures. (Main, 1991, 1993; Slade, 2008).
Self-report measures Self-report measures have emerged from the social psychology tradition and outline the individual’s conscious experiences with regard to current attachment relationships. From this perspective, attachment styles are conceptualised as enduring patterns of needs, expectations and affect-regulation strategies that emerge from earlier relationships with caregivers. Attachment classification according to self-report measures is best represented as regions across two dimensions, attachment-related anxiety and attachment- related avoidance (Shaver & Mikulincer, 2002). Security of attachment is the region where both anxiety and avoidance are low, whilst preoccupation corre- Marina Christina Skourteli & Clare Lennie Counselling Psychology Review, Vol. 26, No. 1, March 2011 23 sponds to a region where anxiety is high and avoidance is low. Avoidant attachment in adults seems more complex, as there is a further distinction (resulting thus in a fourcategory model of classification) based on the individual’s working model of self and other. Fearful-avoidant attachment is defined by negative working models of both self and other, indicating high anxiety over abandonment on one hand and high avoidance of intimacy on the other. In contrast, a dismissive-avoidant attachment is characterised by a positive self-model and a negative other-model, indicating low anxiety but high avoidance of intimacy (Feeney, 1999; Shaver & Mikulincer, 2002).
Basis for present study The present study employs both quantitative and qualitative methods to promote balance and integration in the research process (McLeod, 2003). As attachment, transference and internal working models often seem nebulous and abstract concepts, they are assessed quantitatively in an attempt to capture them more clearly. A possible limitation of self-report measures is their openness to distortion by participants lacking in selfawareness whilst defences such as denial or idealisation, may be further biasing participant responses (Bifulco, 2002). Nevertheless, self-report measures were deemed as ethically more appropriate for the purposes of the present research, as they promote a greater sense of control over responses and informed consent for participation (BACP, 2010; BPS, 2006). In addition to the quantitative part of the study, semi-structured interviews were conducted with five psychotherapy clients, in order to enrich findings generated through the use of questionnaires. Such qualitative analysis promoted a more thorough exploration of client responses and a greater understanding of the personal meanings and experiences of individual participants. Findings generated quantitatively and qualitatively are brought together through the use of researcher reflexivity.
Method Research settings and participants The present study took place within two small counselling organisations. Counselling practitioners receiving ongoing supervision were approached through staff meetings and were informed about the nature of the research. The practitioner’s contribution to the study included accessing clients who were in ongoing therapy with them and identifying clients interested in participating. Their contribution thereafter was limited to the distribution of questionnaires supplied in a standardised manner by the researcher. It was explained that therapists would not be involved any further in the study and they would not be made aware of individual client responses in order to guard client confidentiality and anonymity. In approaching potential participants, therapists were advised to take into account that clients were not in a particularly vulnerable state and that, in their judgement (as informed by the supervision and personal experience of their clients) were not likely to be adversely affected as a result of participation. Individual practitioners maintained the right to refrain from participation; their role however, was central in assessing the level of vulnerability of their clients, ensuring that no harm was caused to clients and sensitively recognising client reluctance to participate. Consistently with relevant literature, therapists were asked to approach clients with whom they had worked for a minimum of three sessions and who were in ongoing therapy with them. Overall 18 therapists agreed to participate who were seeing between two and eight clients per week. Prior to collection of data, full ethical approval was granted by the directors of both organisations and the University of Manchester’s ethics committee. Measures ● Attachment Style Questionnaire (ASQ) (Feeney, Noller & Hanrahan, 1994) This is a 40-item self-report measure of adult attachment based on models of positive versus negative view of self and positive versus negative view of others. These constructs cover the major features described across different models of adult attachment, together with basic themes of infant-attachment theory. The three-factor solution yields factors labelled as Security, Avoidance, and Anxiety. The five-factor solution yields the factors of Confidence (in self and others), Discomfort with Closeness, Need for Approval, Preoccupation with Relationships and Relationships as Secondary. These five attachment scales define four attachment groups based on the notion of model of self versus model of other, resulting in Secure, Preoccupied, Dismissing-Avoidant and Fearful-Avoidant classifications. (Feeney, 1999) ● Client Attachment to Therapist Scale (CATS) (Mallinckrodt, Gantt & Coble, 1995) This is a 36-item, three-factor measure of client attachment to their therapist. ‘The labels and interpretations of the three subscales are as follows: (a) Secure, experiencing the therapist as responsive, sensitive, understanding and emotionally available; feeling hopeful and comforted by the therapist; and feeling encouraged to explore frightening or troubling events; (b) Avoidant-Fearful, suspicious that the therapist is disapproving, dishonest, and likely to be rejecting if displeased; reluctant to make personal disclosures and feeling threatened, shameful and humiliated in sessions; and (c) Preoccupied-Merger, longing for more contact and to be ‘at one’ with the therapist, wishing to expand the relationship beyond the bounds of therapy and being preoccupied with the therapist and the therapists other clients’ (Mallinckrodt, Gantt & Coble, p.310). ● A brief semi-structured interview exploring clients’ perceptions of their therapist and the therapeutic relationship. The interview was developed after reviewing the literature on attachment, particularly so with respect to the therapist as an attachment figure (Farber, Lippert & Nevas, 1995; Mallinckrodt, Gantt & Coble, 1995). Items include ‘How would you describe your relationship with your counsellor?’, ‘In what ways do you find that your counsellor helps you explore difficult feelings or experiences?’, ‘What are the aspects of your relationship with your counsellor that you find most/least helpful?’, ‘What qualities of your counsellor do you find most/least helpful?’, ‘What are your feelings towards your counsellor?’ and ‘How do you find that these feelings may have changed over time?’ Procedure After identifying clients willing to participate, therapists provided participants with identical, sealed and unmarked envelopes containing a participant information sheet explaining the nature and aims of the research, the two self-report measures described above and a participant debriefing form. In order to preserve client anonymity no written consent was obtained; participants interested in participating in further interviews were asked to provide contact details. Research packs were to be completed immediately after a session with the therapist, but procedure also allowed for them to be taken away. Although this possibly carries some methodological limitations over the immediacy and proneness to distortion of responses, it was deemed more ethical, as it could minimise coercion to participate and would promote greater autonomy over responses and right to withdraw consent.
Results 1. Quantitative analysis Descriptive statistics All participants were female and they were all in ongoing therapy with female therapists. Overall 55 research packs were distributed, of which 29 were returned, indicating a response rate of 52.2 per cent. Of the returned packs, two were excluded from analysis, as participants had failed to complete all the questions. Therefore, in total 27 packs were retained for analysis. Of the 27 participants that successfully returned their questionnaires, 11 volunteered for further individual interviews, indicating a response rate of 40.7 per cent. Of those, five participants were randomly selected for individual interviews, in the order in which they returned their research packs. Inferential statistics In completing the ASQ (Feeney, Noller & Hanrahan, 1994), participants obtained a score in the following dimensions corresponding to working models of self and others. These included Confidence (indicating low anxiety and low avoidance), Discomfort with Closeness (high anxiety and high avoidance), Relationships as Secondary (low anxiety, high avoidance), Need for Approval (low avoidance, high anxiety) and Preoccupation in Relationships (high anxiety, low avoidance). In completing the CATS (Mallinckrodt, Gantt & Coble, 1995) participants produced a score describing their relationship to their therapist in the following dimensions; Secure Attachment, Avoidant/Fearful Attachment and Preoccupied/Merger Attachment to Therapist. Statistical correlations were performed with regard to dimensions between the ASQ (Feeney, Noller & Hanrahan, 1994) and CATS (Mallinckrodt, Gantt & Coble, 1995), using Spearman’s r. Need for Approval was positively associated with both Avoidant/Fearful Attachment to Therapist (r=0.56, p=0.002, p<0.01) and more weakly with Preoccupied/Merger Attachment to Therapist (r=0.41, p=0.03, p<0.05). A significant positive correlation was found between Preoccupation in Relationships and Preoccupied/Merger Attachment to Therapist (r=0.41, p=0.03, p<0.05). Contrary to expectation, no significant relationships were found between length of therapy (Number of Sessions) and Secure Attachment to Therapist. Further, no significant relationships were found between Discomfort with Closeness (that can be indicative of both avoidance and anxiety) and Preoccupation in Relationships, Preoccupied/ Merger or Avoidant/Fearful Attachment to Therapist. To summarise, the statistical conclusions from this study suggest that some features of client attachment may be re-enacted within the therapy relationship. Table 1: Summary of correlations between ASQ and CATS. Confid. Discomfort Need for Preoccup. Secure Avoidant/ Preoccup./ with Approval in Attach. Fearful Merger Closeness Relations. Attach. Attach. Number of Sessions .076 .207 Confidence –.486* –.521** Need for Approval .565** .415* Preoccupation in .414* Relationships Discomfort with .111 .166 .051 Closeness * Correlation is significant at the 0.05 level (2-tailed) ** Correlation is significant at the 0.01 level (2-tailed)
2. Qualitative analysis The purpose of the interview was to explore participant’s perceptions of their therapists and their therapeutic relationship at the time of the interview and possible variations of these perceptions over time. Participant transcripts were analysed using Content Analysis (Patton, 2002; McLeod, 2003). The resulting categories were imposed on the data by the researcher primarily in accordance to the literature on the therapist as an attachment figure (Farber, Lippert & Nevas, 1995; Mallinckrodt, Gantt & Coble, 1995; Parish & Eagle, 2003); these categories were kept intact and have not been divided further in order to maintain consistency and congruence with the existing attachment literature. A further category was created to capture a pronounced construct that emerged from participant data and using researcher reflexivity. Overall, five categories emerged that appear indicative and consistent with the notion of the therapist as an attachment figure. 1. The therapist as a secure base for exploration (82 items) The notion of the therapist as a secure base for exploration is consistent with attachment theory; within the security of the therapeutic relationship clients have the opportunity to explore their relationship with self and others (Farber, Lippert & Nevas, 1995; Holmes, 1999, 2001). This is the first most pronounced category emerging from participant interviews. Client perceptions of their therapists’ constancy, availability, sensitivity and responsiveness can be viewed as supporting of the notion of the therapist as a secure base for exploration. ‘She was helping me along by being there.’ (P2) ‘The fact that she’s predictable, it would make me feel very uncomfortable if that was to change from week to week.’ (P3) 2. Transference – relatedness (32 items) Transference within the therapeutic relationship is thought to be mediated through clients’ internal working models of how the therapist is viewed or experienced. This category, the second most pronounced within participant interviews, refers to the notion of the therapist as a specific focus of intense affect, cognition and behaviour (Bowlby, 1988; Farber, Lippert & Nevas, 1995). The notions of transference and internal working models suggest that clients may respond to the therapist in ways that are similar to previous or other current relationships outside therapy. ‘I did at first worry about her, cause when I meet new people I tend to want to look after them.’ (P3) ‘I didn’t trust her at all when I first met her, not because of her but just because I don’t until they prove me wrong.’ (P3) 3. The therapist as containing and as providing a holding environment (26 items) The notion of the therapist as containing and as providing a holding environment (Casement, 1991) refers to processes of holding, metabolising and feeding back affective material to the client. The therapist’s survival and robustness provides a sense of felt security and trust as s/he is neither destroyed nor responds anxiously in an attempt to rescue clients from their distress (Casement, 1991). ‘Sometimes I feel really negative about myself … and she just stays there, she doesn’t try to show me the opposite.’ (P5) ‘… she is not just giving in to my feelings.’ (P5) 4. The therapist as wiser and stronger (24 items) The notion of the therapist as wiser and stronger refers to clients’ perception of the therapist as knowing more than they do, as being able to be more objective and as having greater clarity but also as being understanding and sensitive to client experience. In addition to clients’ expectations, the view of the therapist as wiser and stronger may further emerge from the thera- Marina Christina Skourteli & Clare Lennie Counselling Psychology Review, Vol. 26, No. 1, March 2011 27 pist’s skill, knowledge, education and social status (Farber, Lippert & Nevas, 1995). ‘… I think she actually picked up that the time was right for her to back down but it wasn’t what I wanted at the time, but it was actually what I needed.’ (P1) ‘… she has a lot of insight and an ability to read between the lines and sort of, coax things out of me.’ (P2)
5. Therapeutic boundaries (5 items) The notion of a boundaried relationship emerged as an important element facilitating security in addition to previous categories that were informed by relevant attachment literature (Farber, Lippert & Nevas, 1995). This is the least pronounced category across participant accounts yet the notion of therapeutic boundaries appeared to be particularly important for clients in the interview population. ‘… I’ve got the boundaries really there in place … I think you do need to know where the line is.’ (P1) ‘And she is very respectful of that boundary and that makes me trust her even more.’ (P3)
Discussion The purpose of the present study was to explore the ways in which client attachment orientation in close relationships may be enacted within the therapeutic relationship and the extent to which the therapist may serve as an attachment figure for clients. Consistent with expectation, it appears that some of the features characterising client adult patterns of relatedness outside therapy manifest themselves within the therapeutic relationship. 1. Quantitative analysis A positive association was found between Need for Approval and both Avoidant/ Fearful and Preoccupied Attachment to the therapist. Whilst Need for Approval is indicative of anxiety and implies a negative internal model of self, it seems to be implicated in different patterns of relatedness. Mallinckrodt, Gantt and Coble (1995) maintain that fearful/avoidant clients in their study tended to mistrust their therapists and be fearful of rejection; in spite of these clients maintaining negative working models of themselves and others, they displayed a strong need for emotional closeness on one hand, but were uncertain over their capacity to establish supportive and fulfilling relationships on the other, due to their negative expectations of themselves and others. In this context, the association between Need for Approval and Avoidant/Fearful attachment in this study is consistent with Mallinckrodt, Gantt and Coble’s (1995) findings and may be important in demonstrating that the avoidance dimension in relationships may be particularly complex, often mediated by a degree of anxiety or ambivalence. The positive association between Need for Approval and Preoccupied Attachment to therapist is less surprising; preoccupation is thought to be mediated by a negative working model of oneself but a positive working model of others. From this may follow a tendency for preoccupied individuals to idealise others in relationships, including the therapeutic relationship and to view others as sources of self-validation. It appears then that anxious individuals’ perception of others may be biased by their own need for connectedness, which in the context of therapy may be translated as a wish for dissolution of boundaries and a tendency towards dependency (Feeney, 1999; Shaver & Mikulincer, 2002). These results are consistent with previous studies that report that both the Avoidant/Fearful and Preoccupied/Merger subscales in the CATS are negatively associated with self-efficacy (Mallinckrodt, Gantt & Coble, 1995; Mallinckrodt, 2000). Finally, a positive association was found between Preoccupation in Relationships and Preoccupied attachment to therapist. This finding appears significant for the notion of a re-enactment of clients’ patterns of relatedness (the anxiety dimension in particular) within the therapeutic relationship and is in line with previous attachment and inner representations literature. Bender, Farber and Geller (1997) reported that clients’ representations of parents were significantly correlated with representations of the therapist, although the latter could be modified over time, as a result of therapy. Connolly et al (2000) similarly reported a significant relationship between the most prominent interpersonal themes in client pre-therapy narratives and later those relating to their perception of their therapist. Taken together, these studies are supportive of the notion that client patterns of relatedness outside therapy may be replicated to some extent within the therapeutic relationship. 2. Qualitative analysis and Triangulation of findings The qualitative analysis in this study appears to yield support for the hypothesis that the therapeutic relationship contains features of an attachment relationship and is consistent with relevant literature (Farber, Lippert & Nevas, 1995; Parish & Eagle, 2003; Holmes, 2001). Participants in this study perceived their therapists as consistent and emotionally responsive, attentive and available and generally as reliable and trustworthy figures they could return to during the exploration of their emotional experience. Therapists were generally perceived as being wiser or stronger than clients and further as providing a containing and holding environment and a safe space that enhanced clients’ sense of felt security. Therapists overall served as figures, which clients sought emotional proximity to and were further the specific focus of intense emotional responses that often paralleled clients’ experiences in other relationships. Some of the participants in the study acknowledged that their relatedness to their therapist, particularly during the earlier stages of their therapy was determined by previous positive or negative experiences with others and that their perceptions of their therapists were initially formed on the basis of internalised expectations of others’ behaviour. The themes emerging from client transcripts suggest that the therapeutic relationship can provide the necessary relational conditions for a corrective emotional experience for clients, which they can utilise as a secure framework for the exploration of their presenting difficulties (Holmes, 1999). Whilst findings indicate that the therapist may serve as a secondary attachment figure for clients, there has been no systematic attempt in this study to establish a link between individual clients’ attachment orientation (as assessed by self-report measures) and their perception of the therapist or the therapeutic relationship; yet it seems important not to overlook a brief discussion of a possible relationship between the two primarily through means of researcher reflexivity. Whilst some of the participants reported relatively high scores in the constructs of confidence and secure attachment to therapist, closer examination of their interview transcripts revealed some inconsistencies or conflict with regard to the relationship with their therapist. For example, in describing an experience with a previous therapist, one participant reported feeling uncomfortable with her therapist’s self-disclosure but later on expressed frustration towards the therapist for not revealing enough about herself; the participant herself seemed unaware of contradictions in her account. Such a flavour of preoccupation was often manifested in the subjective experience of the researcher during the process of conducting interviews. For example, some participants produced narratives that seemed excessively long and disorganised; these clients appeared to have difficulty in keeping a clear focus on the interview process whilst constructing narrative accounts of their experience. Although such reflection on participant transcripts in the present study is highly speculative and unsystematic, it nevertheless appears in line with the literature utilising narrative assessment of adult attachment that considers meta-cognitive functioning or the ability to Marina Christina Skourteli & Clare Lennie Counselling Psychology Review, Vol. 26, No. 1, March 2011 29 reflect on current or past experience whilst simultaneously maintaining a coherent discourse with the interviewer, as indicative of security (Main, 1993; Hesse, 2008). It may be interesting to consider the additional theme of a boundaried relationship within this context. A strong need for a boundaried relationship for clients who may be experiencing anxiety in relationships may shed some light in the often-obscured avoidance dimension within preoccupation; this may further be relevant to the understanding of avoidant/fearful attachments. For clients experiencing high need for approval and closeness, relationships themselves (although high in the hierarchy of psychological needs) may also be a source of considerable anxiety, as self and others may be viewed as demanding, overprotective, unreliable or dependent. The intense emotional experience accompanying relationships may lead to a sense of exhaustion and a desire for greater emotional distance, as a respite from anxiety; these clients may be finding themselves feeling either suffocating or being suffocated by others in relationships. Within the therapeutic relationship this desire for distance may indeed manifest itself through a strong need for clarification of the boundaries between client and therapist, as something that could possibly counteract the element of ambivalence. For these clients, an awareness of formal, external boundaries may be reassuring as their capacity for establishing boundaried relationships from within (through well differentiated representations of self and others) may be depleted. The possible discrepancies between selfreport and interview measures in the present study may be highlighting the role of defence mechanisms, particularly so for clients with insecure patterns of attachment. Shaver and Mikulincer (2002) report that attachment security, as assessed by selfreport measures has been repeatedly found to be negatively associated with the function of defensive distortions in participant perceptions of themselves and others. Similarly, Bifulco (2002) comments that self-classification measures of attachment may be prone to distortions emerging from defence mechanisms such as denial or idealisation and further ‘…by the very cognitive biases constituting vulnerability…’ (p.183) for clients’ presenting difficulties (Bifulco, 2002). It appears then that whilst self-report measure may be useful in assessing attachment security, they may be limited in accurately reflecting insecure patterns of attachment; such limitations may further account for the lack of significant relationships between the constructs of Discomfort with Closeness and Preoccupation in Relationships, Avoidant/ Fearful or Preoccupied/ Merger attachment to therapist within the present study. This suggests that narrative approaches may indeed be more sensitive in capturing aspects of attachment insecurity and the defence mechanisms that underlie them.
3. Implications for Practice If the therapeutic relationship contains qualities of an attachment relationship and the therapist acts as a secondary attachment figure for clients, this holds considerable implications for case conceptualisation and clinical practice. Research suggests that therapy can facilitate security in clients’ attachment through a deconstruction and reappraisal of their internal working models with regard to perceptions of themselves and others (Mallinckrodt, 2000; Slade 2008). Therapy essentially provides the emotional arena within which the transference-countertransference dynamic takes place; this unfolding relational exchange between client and therapist can lend a valuable context for an in-depth understanding of the former’s presenting difficulties and the latter’s therapeutic use of self (Wosket, 1999; Steven & Peltham, 2000). Clients’ patterns of relatedness manifest within the therapy relationship, often along with potential responses that the client attempts to pull from the therapist, in order to confirm or perpetuate their working model of self and others. For example, clients with deactivating attachment strategies may seek avoidant attachments with their therapist and to increase interpersonal distance; this may elicit disengaging or distancing responses in the therapist’s countertransference where s/he may feel locked out or hopeless about the prospects of attaining emotional contact with the client. In contrast, clients with hyperactivating attachment strategies may display preoccupied attachments to their therapist by seeking to reduce interpersonal distance through communicating their sense of helplessness and dependency; this in turn may elicit rescuing responses in the countertransference or may leave the therapist feeling swamped or overwhelmed by the client’s experience (Mallinckrodt, 2000; Shilkret, 2005). Within the framework of attachment theory, such client behaviour is seen as meaningful, aiming to maximise emotional safety in the relationship with the therapist; on the other hand therapist attentiveness to countertransference responses similarly may provide insight into the client’s early emotional experience. Importantly, the therapist’s awareness of such countertransference responses can be used therapeutically, facilitating empathy and offering insight into appropriate ways of being with clients. Many theorists suggest that a corrective emotional experience may often involve the therapist’s withholding of responses that the client may seek to obtain and instead providing counter-complementary or contrasting responses (Dozier, Cue & Barrett, 1994; Bernier & Dozier, 2002; Slade, 2008). For example, with clients who seek to increase interpersonal distance and are dismissive of closeness, the therapist may attempt to gently increase their awareness of emotional processes and deepen their emotional engagement in therapy, while sensitively monitoring their tolerance for anxiety and cognitively containing them. On the other hand, for clients with a tendency to exaggerate their attachment needs in relationships, the therapist may attempt to increase clients’ sense of autonomy and self-efficacy by resisting pulls for rescuing and maintaining appropriate boundaries, again whilst sensitively monitoring and containing client anxiety (Bernier & Dozier, 2002; Shilkret, 2005). In short, a corrective emotional experience may be achieved in therapy by introducing dissonance in client’s stereotyped expectations for others and subsequent self representations. The present study has focused solely on client’s patterns of relatedness, yet it is important to acknowledge that therapy is a mutual process of exchange that involves both the client’s and the therapist’s internal working models. In that respect, the therapist’s own experiences and areas of unresolved conflict may influence the therapeutic relationship in ways equally significant to those of clients (Maroda, 2010; Meszaros, 2004). In their study, Dozier, Cue and Barrett (1994) found that securely attached clinical case managers tended to provide more balanced countercomplementary interventions in their work with clients whereas those less securely attached tended to confirm or perpetuate their clients’ working models by either gratifying preoccupied clients’ attempts for closeness (and thus their perceptions of themselves as fragile and needy) or by working more superficially with deactivating clients thus reinforcing their belief in others as unavailable or unresponsive. Further research suggests that therapists’ unresolved attachment issues may be confounding countertransference reactions (Mallinckrodt, 2000). It seems critical, therefore, that counselling psychologists are aware of their own patterns of relatedness in order to be able to identify and distinguish between areas of personal conflict and genuine countertransference responses and ultimately to ensure ethical and competent practice. In this context, supervision and personal therapy gain great importance for the training and practice of counselling psychology.
4. Limitations and further research Within the present study, no significant relationship was found between length of therapy (i.e. number of sessions) and Secure Attachment to therapist. Although the variance in participants length of therapy was considerable (between three and 100 sessions) and could have provided sufficient conditions for documentation of significant effects, the lack of a unified theoretical approach employed by participating therapists and their varying level of training and experience may have been problematic in observing linear effects of time in therapy. Therapist attachment was not explored within the present study, although it has been shown to influence ways of being with clients and neither were client perceptions of the therapist and the relationship (Dozier, Cue & Barrett, 1994). The use of self-report measures in this study was deemed as more ethical, however, this may have posed considerable limitations particularly so for clients who were less secure and thus more prone to defensive distortions or bias in their self-classification. Further research utilising narrative measures of client attachment such as the AAI (Hesse, 2008) or the CCRT method (Luborsky & Crits-Cristoph, 1998) may be more sensitive in capturing unconscious processes underlying attachment classification. The contribution of the therapist’s patterns of relatedness should not be overlooked when exploring the relational dynamic unfolding in therapy; it would thus be interesting for future research to explore possible interactions between processes of client transference and therapist countertransference. 5. Conclusion In conclusion, it appears that attachment theory can provide an invaluable framework for an understanding of what happens in therapy. It can serve as a basis for integration in counselling psychology practice, as it facilitates the use of narrative, cognitive and psychodynamic approaches whilst remaining essentially humanistic and relationshiporiented (Holmes, 2001; Jordan, 2007). Attachment theory can provide a framework for practice that although unified, it can remain flexible and adaptable to particular client needs, as these are informed by their unique attachment experiences. It seems important for practitioners and researchers informed by attachment theory to remain critical, reflexive and sensitive to the particular contexts, such as cultural factors, that may determine the formation of relational patterns (both their clients’ and their own), nevertheless attachment theory seems to be a particularly rich and constructive theoretical framework in informing clinical work in counselling psychology.
About the authors
Marina Christina Skourteli is a Chartered Counselling Psychologist currently completing her DPsyc in Counselling Psychology at City University, London. She lives and works in Athens, Greece. Dr Clare Lennie is Joint Programme Director of the Doctorate in Counselling Psychology at the University of Manchester.
Correspondence E-mail: [email protected] 32 Counselling Psychology Review, Vol. 26, No. 1, March 2011 Marina Christina Skourteli & Clare Lennie Ainsworth, M.D, Blehar M.C., Waters, E. & Wall, S. (1978). Patterns of attachment: A psychological study of the strange situation. Hillsdale, NJ: Erlbaum. BACP (2010). Ethical framework for good practice in counselling and psychotherapy. Rugby: BACP. Bateman, A. & Fonagy, P. (2004). Psychotherapy for borderline personality disorder: Mentalisation-based treatment. Oxford: Oxford Medical Publications. Bateman, A. & Holmes, J. (1995). Introduction to psychoanalysis – contemporary theory and practice. London: Routledge. Beebe, B., Jaffe, J. & Lachmann, F.M. (1992). A dyadic systems view of communication. In N.J. Skolnick & S.C. Warshaw (Eds.), Relational perspectives in psychoanalysis. London: The Analytic Press. Bender, D.S., Farber, B.A. & Geller, J.D. (1997). Patient’s representations of therapist, parents and self in the early phase of psychotherapy. Journal of the American Academy of Psychoanalysis, 25(4), 571–586. Bernier, A. & Dozier, M. (2002). The client-counsellor match and the corrective emotional experience: Evidence from interpersonal and attachment research. Psychotherapy: Theory, research, practice, training, 39(1), 32–43. Bifulco, A. (2002). Attachment style measurement: A clinical and epistemological perspective. Attachment and Human Development, 4(2), 180–188. Bowlby, J. (1988). A secure base: Clinical applications of attachment theory. London: Routledge. British Psychological Society (BPS) (2006). Code of Ethics and Conduct. Leicester: BPS Publications. Bretherton, I. & Munholland, K.A. (2008). Internal working models in attachment relationships. In J. Cassidy & P.R. Shaver (Eds.), Handbook of attachment: Theory, research and clinical applications. New York: The Guilford Press. Casement, P.J. (1991). Learning from the patient. New York: The Guilford Press. Connoly, M.B., Crits-Cristoph, P., Barber, J.P. & Luborsky, L. (2000). Transference patterns in the therapeutic relationship in supportive expressive psychotherapy for depression. Psychotherapy Research, 10(3), 223–233. Crowell, J.A., Fraley, R.C. & Shaver, P.R. (2008). Measurement of individual differences in adolescent and adult attachment. In J. Cassidy & P.R. Shaver (Eds.), Handbook of attachment: Theory, research and clinical applications. New York: The Guilford Press. Dozier, M., Cue, K.L. & Barrett, L. (1994). Clinicians as caregivers: The role of attachment organisation in treatment. Journal of Consulting and Clinical Psychology, 62(4), 793–800. Dozier, M., Stovall-McClough, K.C. &Albus, K.E.(2008). In Cassidy, J. & Shaver, P.R. (Eds.). Handbook of Attachment: theory, research and clinical applications. NY: The Guildford Press Farber, B.A., Lippert, R.A. & Nevas, D.B. (1995). The therapist as attachment figure. Psychotherapy, 32(2), 204–212. Feeney, J. (1999). Adult romantic attachment and couple relationships. In J. Cassidy & P.R. Shaver (Eds.), Handbook of attachment: Theory, research and clinical applications. New York: The Guilford Press. Feeney, J.A., Noller, P. & Hanrahan, M. (1994). Assessing adult attachment. In M.B. Sperling & W.H. Berman (Eds.), Attachment in adults: Clinical and developmental perspectives. New York: The Guilford Press. Goldberg, S. (2000). Attachment and development. London: Arnold. Grossman, K.E. & Grossman, K. (1991). Attachment quality as an organiser of emotional and behavioural responses in a longitudinal perspective. In C.M. Parkes, J. Stevenson-Hinde & P. Marris (Eds.), Attachment across the life cycle. London: Routledge. Hesse, E. (2008). The Adult Attachment Interview: Historical and current perspectives. In J. Cassidy & P.R. Shaver (Eds.), Handbook of attachment: Theory, research and clinical applications. New York: The Guilford Press. Holmes, J. (1999). Ghosts in the consulting room – an attachment perspective on intergenerational transmission. Attachment and Human Development, 1(1), 115–131. Holmes, J. (2001). The search for the secure base: Attachment theory and psychotherapy. Hove: Brunner- Routledge. Jordan, L. (2007). Attachment-based counselling. Could this be an ABC of psychological therapies? Counselling Psychology Review, 22(1), 58–61. Luborsky, L. & Crits-Cristoph, P. (1998). Understanding transference: The Core Conflictual Relationship Theme method. Washington, DC: American Psychological Association. Main, M. (1991). Metacognitive knowledge, metacognitive monitoring and singular (coherent) vs. multiple (incoherent) models of attachmentfinding and directions for future research. In C.M. Parkes, J. Stevenson-Hinde & P. Marris (Eds.), Attachment across the life cycle. London: Routledge. Main, M. (1993). Discourse, prediction and recent studies in attachment: Implications for psychoanalysis. Journal of the American Psychoanalytic Association, 41, 209–245.
References Counselling Psychology Review, Vol. 26, No. 1, March 2011 33 The therapeutic relationship from an attachment theory perspective Mallinckrodt, B., Gantt, D.L. & Coble, H.M. (1995). Attachment patterns in the psychotherapy relationship: Development of the Client Attachment to Therapist Scale. Journal of Counselling Psychology, 42(3), 307–317, Mallinckrodt, B. (2000). Attachment, social competencies, social support and interpersonal process in psychotherapy. Psychotherapy Research, 10(3), 239-266 Mallinckrodt, B., Porter, M.J. & Kivlighan, D.M. (2005). Client attachment to therapist, depth of in-session exploration and object relations in brief psychotherapy. Psychotherapy: Theory, research, practice, training, 42(1), 85–100. Maroda, K.J. (2010). Psychodynamic techniques: Working with emotion in the therapeutic relationship. New York: The Guilford Press. McLeod, J. (2003). Doing counselling research. London: Sage. Meszaros, J. (2004). Psychoanalysis is a two-way street. International Forum Psychoanalysis, 13, 105–113. Parish, M. & Eagle, M.N. (2003). Attachment to the therapist. Psychoanalytic Psychology, 20(2), 271–286. Patton, Q.M. (2002). Qualitative research and evaluation methods. Thousand Oaks, CA: Sage. Shaver, P.R. & M. Mikulincer (2002). Attachmentrelated psychodynamics. Attachment and Human Development, 4(2), 133–161. Shilkret, C.J. (2005). Some clinical applications of attachment theory in adult psychotherapy, Clinical Social Work Journal, 33(1), 55–68. Slade, A. (2008). Attachment theory and research: Implications for the theory and practice of individual psychotherapy with adults. In J. Cassidy & P.R. Shaver (Eds.), Handbook of attachment: Theory, research and clinical applications. New York: The Guilford Press. Steven, P.A. & Pelham, G. (2000). A relational approach to therapy. In S. Palmer & R. Woolfe (Eds.), Integrative and eclectic counselling and psychotherapy. London: Sage Publications. Stevenson, C. & Cooper, N. (1997). Qualitative and quantitative research. The Psychologist, 10(4), 159–160. Westen, D. & Gabbard, G.O. (2002). Developments in cognitive neuroscience II: Implications for theories of transference. Journal of the American Psychoanalytic Association, 50, 99–134. Woodhouse, S.S., Schlosser, L.Z., Cook, R.E., Ligiero, P.D. & Gelso, C.J. (2003). Client attachment to therapist: Relations to transference and client recollections of parental caregiving. Journal of Counselling Psychology, 50(4), 395–408. Wosket, V. (1999). The therapeutic use of self: Counselling practice, research and supervision. London: Routledge.