Four Patterns of Adult Discourse Observed in the Adult Attachment Interview

Coherence is a central construct in attachment interviews. Coherent discourse is based on what the linguistic philosopher Grice calls the ‘Cooperative Principle’. This has four maxims, namely:

Quality: be truthful and have evidence for what you say
Quantity: be succinct, yet complete
Relevance: be relevant
Manner: be clear, brief and orderly

The Adult Attachment Interview (AAI) is a semi-structured interview devised by George, Kaplan and Main, based on Grice’s principle. It provides researchers with a standardized method to assess adult mental representations of childhood attachment experiences, the influence of these experiences as perceived by the interviewee and the current relationship with one’s parents. The participant is also asked about loss of loved ones and about other traumatic experiences. During the interview, the interviewee is faced with the dual tasks of producing and reflecting upon memories related to attachment while simultaneously maintaining coherent discourse with the interviewer.

Bowlby drew attention to the ways in which information is stored in distinct systems of memory. Episodic or explicit memory consists of information that is stored in the form of temporally dated autobiographical details. Each remembered event or episode has its own distinctive place in the person’s life history. By contrast, semantic or implicit memory consists of generalised information about the world and the person’s sense of self in relation to significant others. Such generalised information is encoded in internal working models and mediates the person’s attachment-related thoughts, feelings and behaviour in a largely non conscious or procedural way. Implicitly encoded information may be at great variance with information stored in the explicit memory systems. This gives rise to cognitive and emotional conflict and to gross inconsistencies between the generalisations a person makes about his or her parents and what is explicitly implied or actually recalled in terms of specific episodes. Such conflict and inconsistencies indicate the operation of parallel memory systems and the dissociation of painful affect. The AAI is designed to detect conflict and inconsistencies in the discourse and narrative style of the interviewee.

The AAI operationalizes Bowlby’s construct of the “internal working model” as a “state of mind with respect to attachment”, as expressed in discourse about early relationships. The researcher shifts attention from the content of autobiographical memory to the form of discourse in which those memories are presented. For example, the mother’s state of mind in respect of her attachment history may be classified as secure-autonomous and her child as securely attached, despite her having experienced early trauma in the form of separation, loss and/or abuse. Such findings indicate the resolution of trauma and the attainment of ‘earned security’ via subsequent secure attachment experiences which, of course, may include a therapeutic relationship. AAI classifications, then, reveal differences in discourse style, in access to attachment memories, and in ability to coherently discuss past attachment experience.

The following four patterns of adult discourse in the AAI have been observed:

Secure-Autonomous: Adults termed secure-autonomous provide discourse that is open, free, coherent and collaborative, presenting even difficult early attachment experiences in clear and vivid ways. Discourse includes no contradictions between semantic and episodic memories of childhood attachments, a focus on the goal of the discourse task and rich use of language and expression. The interviewee demonstrates an ability to discuss and reflect upon personal attachment experiences in collaboration with the interviewer without disorganization, lack of memory or passivity of thought. These interviews are characterized by recognition, acceptance and forgiveness of imperfections and injustices in parents and in self, reflecting an integration of positive and negative feelings. As noted above, even adults with extreme and abusive attachment histories, who have come to understand coherently their early difficulties, may provide a coherent and autonomous narrative.

Discourse termed insecure or non-autonomous may show one of three patterns:

Dismissing: Transcripts coded as dismissing tend to be excessively brief and are characterized by notable contradictions in the interviewee’s discourse about early attachments, with generalised representations of history being unsupported or actively contradicted by episodes. Strong idealization of caretakers is common, along with contradictory and impoverished memories of actual events. The interviews are notable for restriction in coherence and content, indicating a deactivating strategy with respect to potentially painful memories. Some adults in this group minimize the importance of close relationships and derogate or dismiss the influence of attachment experiences, emphasizing, instead, extraordinary self-reliance.

Preoccupied: The transcripts of adults termed preoccupied may be excessively long and embellished, including information that is irrelevant to the discourse task. Interviewees are not able to describe their attachment biography coherently and show an inability to move beyond an excessive preoccupation with attachment relationships. There are frequent examples of passive speech, sentences begun and left unfinished and specific ideas that disappear in vague expressions. The boundaries between present and past and self and other are often confused. There is a diffuse self-concept and a notable inability to reflect upon experience. In some transcripts coded as preoccupied there is notable anger, passivity or fear, which is displaced from past childhood events to the present discourse task, indicating a continuing intense involvement and preoccupation with attachment experiences. The reliving of the affective experience of historical events interferes with the interviewee’s consciousness of the current discourse task.

Unresolved: Transcripts of adults are termed unresolved/disorganized when there is evidence of substantial lapses in the monitoring of reasoning and discourse, specifically surrounding the discussion of traumatic events involving loss, physical or sexual abuse. The interviewee may briefly indicate a belief that a dead person is still alive in the physical sense, or that this person was killed by a childhood thought. The individual may lapse into prolonged silence, engage in eulogistic speech or enter a trance-like dissociated state. It should be noted that the unresolved classification is made solely on the discussion of trauma, abuse or loss experiences and is superimposed on one or other of the three main attachment classifications.
Findings from research utilizing the AAI show that psychopathology is associated with non-autonomous patterns of attachment and that people classified as preoccupied and unresolved/disorganized are strongly over-represented in clinical samples.