Foundations
What is Relational Frame Theory?
A behaviour-analytic account of human language and cognition, and the reason a learner can know things nobody taught them.
The core idea: derived relational responding
Teach a child that the written word "cat" goes with a picture of a cat, and that the picture goes with the spoken word "cat". Without further teaching, most children will then match the written word to the spoken word, and go the other way too. Nobody taught those last relations. They were derived.
RFT calls this arbitrarily applicable relational responding. "Arbitrary" because the relation is set by social convention rather than physical resemblance: the written word "cat" looks nothing like a cat, and the relation holds anyway.
Three defining properties
A relational frame is identified by three features working together.
- Mutual entailment. If A relates to B in some way, then B relates back to A. Taught that A is the same as B, a learner derives that B is the same as A.
- Combinatorial entailment. Relations combine. Taught A is the same as B, and B the same as C, a learner derives the relation between A and C in both directions.
- Transformation of stimulus function. The psychological function of one stimulus transfers through the network. If C becomes frightening, A may become frightening too, with no direct experience of A.
Why it matters in practice
The third property carries the most practical weight. It means a learner can acquire responses to stimuli they have never directly encountered, which is the mechanism behind a great deal of generative language.
For practitioners, this reframes the goal. Rather than teaching every relation directly, the aim becomes establishing relational responding as a skill, so that a smaller amount of direct teaching yields a much larger network of derived responses.
The relational frames
RFT describes families of frames, each with its own entailment pattern.
Coordination
Relating stimuli as the same or equivalent. The basis of naming.
Distinction
Relating stimuli as different, without specifying how.
Opposition
Relating stimuli as opposites along a shared dimension.
Comparison
Relating stimuli along a graded dimension: more, less, bigger.
Hierarchical
Relating stimuli as member and category, part and whole.
Temporal
Relating events in time: before, after, then, now.
Spatial
Relating objects in space: above, beside, inside, behind.
Deictic
Relating from a perspective: I and you, here and there, now and then.
Causal
Relating events as cause and effect, if and then.
Where to read more
The foundational text is Relational Frame Theory: A Post-Skinnerian Account of Human Language and Cognition (Hayes, Barnes-Holmes, & Roche, 2001). Niklas Törneke's Learning RFT (2010) is the standard accessible introduction. The Association for Contextual Behavioral Science maintains a free online tutorial.
Common questions
- What is Relational Frame Theory in simple terms?
- It is a behaviour-analytic theory of language and thought. It says humans learn to relate things to each other in arbitrary ways, and once that skill exists, people derive relations they were never taught directly.
- What is derived relational responding?
- Responding to a relation between stimuli that was never directly trained. Taught that A goes with B and B goes with C, a learner derives the A-to-C relation without instruction.
- Who developed Relational Frame Theory?
- Steven C. Hayes, Dermot Barnes-Holmes, and Bryan Roche set it out in their 2001 book, building on earlier work on stimulus equivalence.
- How is RFT different from stimulus equivalence?
- Stimulus equivalence describes one pattern, coordination. RFT generalises the account to many relation types and treats relational responding itself as learned operant behaviour.