Derived relational responding

The tested relation was never directly taught

Suppose a person is taught that an abstract symbol A goes with B and B goes with C. A later A-to-C or C-to-A response may emerge without direct teaching of that pair.

The history must be explicit. If an instructor, worksheet, prior lesson, or everyday experience already taught the tested relation, the response may be accurate without being derived in the experimental sense.

Stimulus equivalence is one important case

Stimulus equivalence is commonly evaluated through reflexivity, symmetry, and transitivity or combined equivalence tests. Teaching A-to-B and B-to-C relations can support untrained B-to-A, C-to-B, A-to-C, and C-to-A performances.

A review of derived stimulus relations describes the emergence of unreinforced matching from a smaller trained network. Actual outcomes vary with learning history, procedures, stimuli, and participants.

Derived relations extend beyond sameness

People can respond to relations such as larger than, before, opposite, contains, belongs to, here versus there, and self versus other. The relation may reverse in a predictable way. If A is larger than B, then B is smaller than A.

Two trained relations can also combine. If A is before B and B is before C, a person may derive that A is before C. The exact pattern depends on the relation and contextual cues.

The phenomenon and its theories differ

Stimulus-equivalence research, relational frame theory, naming accounts, and other approaches offer descriptions or explanations for derived performance. The observed phenomenon is broader than any one theory.

An article should identify whether it is reporting an observed untrained relation, using equivalence terminology, or making a theoretical claim. Treating these as interchangeable obscures the evidence.

Stimulus functions can change through relations

If a stimulus acquires an eliciting, reinforcing, aversive, or discriminative function, related stimuli may affect behavior without direct training of each one. Researchers describe this as transfer or transformation of stimulus function, depending on the relation and conceptual framework.

The BACB BCBA Test Content Outline covers processes that promote emergent relations and generative performance. The outline defines examination content, not a clinical protocol.

A fictional probe set

Owen chooses to learn symbols used in a transit app. Training establishes Route A-to-color B and color B-to-destination C. He meets the predeclared criterion on both trained relations.

During unreinforced probes, Owen selects B given A in 4 of 4, C given B in 3 of 4, C given A in 3 of 4, and A given C in 2 of 4. Report each relation separately. A pooled 12 of 16 would hide the weaker combined reverse relation.

These probes show a pattern under this task. They do not establish a generalized relational repertoire or explain which learning process produced it.

Probe design protects the inference

Document every trained and tested relation. Keep probe trials free of feedback that would teach the answer during assessment. Balance comparison positions, trial order, and exposure.

Include enough opportunities to distinguish stable responding from chance. Record prompts, self-corrections, omissions, invalid trials, and prior familiarity. Replicate with novel stimuli when the question concerns generativity.

Accessibility belongs in the design

Use stimuli the person can see, hear, touch, or otherwise access. Accept an effective response form such as touch, eye gaze with a validated access method, speech, sign, keyboard, switch, or augmentative and alternative communication.

Keep necessary communication and mobility supports available. Adaptations should preserve the relation being tested and be documented so results remain interpretable.

Clinical significance needs separate evidence

A laboratory matching result may inform assessment or instruction, yet it does not automatically identify a socially important goal. Ask whether the skill supports communication, learning, choice, safety, education, work, relationships, or another priority selected by the person.

Measure functional use, maintenance, generalization, burden, and social validity separately. A 2020 review of applied research with children found varied procedures and outcomes, reinforcing the need to describe exactly what was trained and tested.

Create a relation inventory before testing

List each stimulus, trained relation, training direction, mastery criterion, and proposed probe. Mark which reverse or combined relation each probe evaluates. This prevents a familiar or previously reinforced pair from being mislabeled as derived.

After testing, record relation-level counts, trial order, feedback status, and any repeated exposure. If a probe is taught after an error, move subsequent presentations into a post-teaching phase. Keep a clean distinction between initial emergence, later acquisition, and generalization to another stimulus set.