Core Conflictual Relationship Theme (CCRT)

Research background and how it was developed

Where the method comes from

The idea that people repeat old relationship patterns with new people goes back to Freud, who noticed that his patients brought feelings to him that seemed to belong to earlier relationships, and who came to see that what cannot be remembered tends to be repeated (Freud, 1912, 1914). For decades that idea lived mainly in clinical judgment. In 1977 Lester Luborsky introduced the Core Conflictual Relationship Theme method to make it observable: instead of inferring a pattern from a whole treatment, judges read short accounts of specific interactions, called relationship episodes, and code each one into a wish, a response of the other and a response of self (Luborsky, 1977). The theme is the combination that recurs across the largest number of episodes.

Over the following two decades Luborsky, Paul Crits-Christoph and their colleagues built the method into a standardized procedure with fixed category lists, and in their research trained judges agreed well enough for the method to be used in studies of psychotherapy (Luborsky & Crits-Christoph, 1998). That work reported that the same theme tends to appear across different relationships, and it examined how the theme’s pervasiveness changes over the course of treatment (Crits-Christoph & Luborsky, 1990), and one study found that therapists’ interpretations that matched a patient’s theme were associated with better outcomes (Crits-Christoph, Cooper, & Luborsky, 1988). These findings describe trained human coding in clinical settings, not this online version.

How this version was built

This version keeps the structure of the method and automates the coding. It asks for three written episodes instead of the ten or so that clinical use usually gathers. Each story is read sentence by sentence against a fixed dictionary built on the method’s standard category names, with 35 categories for the wish, 30 for the response of the other and 31 for the response of self. When a sentence matches a category, the match is counted toward that part of the pattern, and each category also carries a fixed position on the two dimensions of warmth and dominance, which is what the pattern map plots. The scorer is a fixed set of rules rather than an artificial intelligence (AI) model, and nothing about you is compared with other people.

How it has been checked, and what has not been checked yet

Because the rules are fixed, the same stories always produce the same result, and the software’s automated tests re-run the scorer on a fixed set of example stories whenever the code changes. What has not yet been done is a comparison between this scorer and trained judges coding the same stories, so there is no figure for how closely the two agree, and no reference group of other people’s results stands behind the report. Keyword rules can also miss meaning that a clinician would catch, such as irony, negation or an unusual way of describing a feeling, and three stories give the pattern fewer chances to show itself than a full clinical set would. For these reasons the report describes themes in the words you wrote, and it is best read as a structured starting point for reflection or for a conversation with a clinician.