Adult Attachment Interview — A Primer

You may have heard the phrase Adult Attachment Interview and wondered what actually happens in one. The short answer is: about an hour of open, mostly unstructured questions about a person's childhood relationships with important others — and, more importantly, how the person tells the story.

Developed by Mary Main and colleagues in the 1980s, the Adult Attachment Interview (AAI) does something quietly clever. It does not ask what happened. It asks the person to describe their early experience of closeness, and then it codes not the content of the story but the coherence of it (Hesse, 2016) [Externally Validated Evidence].

The reason coherence matters is that a well-integrated attachment history tends to read as an organised narrative — the person can hold both positive and difficult memories together, offers concrete supporting detail, and is honest about complexity. A less-integrated history often shows up not in what is said but in how it is said: contradictions the speaker does not catch, dismissive brevity where fuller detail would help, or being pulled into vivid distress by memories that should be far away.

The classifications the AAI yields — secure, dismissing, preoccupied, unresolved — track the same underlying attachment landscape as the Strange Situation, but from the adult side. And crucially, the correspondence between an AAI classification and the child's Strange Situation classification is not perfect but is meaningful (van IJzendoorn, 1995), which is one of the strongest empirical bridges between infant and adult attachment research.

What the AAI is genuinely useful for is clinical work — particularly with parents whose own attachment history is showing up in how they respond to their child, and with adults trying to understand why closeness in the present carries such old weight. It is also useful in research where a rigorous adult measure is needed and a self-report survey is not enough.

The honest limits are worth naming. The AAI takes real training to administer and score. The classifications are not diagnoses and should not be used as such. And they describe a working model, not a fixed identity — a person's coherence around attachment history can shift as they do their own work.

The most useful frame is this: the AAI listens for how integrated a person's attachment story is right now. Integration is trainable, over time, in the presence of another attuned nervous system. That is not a promise. It is what the research suggests, and it is why the interview has held its place in the field for four decades.

For the paradigm the AAI extends, see Ainsworth's Strange Situation, Reconsidered. For how attachment patterns land in adult life, The Four Adult Attachment Patterns, Correctly is a useful next step.