Research Cluster

Stimulus Equivalence and Emergent Learning

This cluster shows how to teach kids to match new things without extra work. You learn how to set up lessons so a child who is taught A=B and B=C can suddenly pick A=C, even though you never drilled that last pair. It gives easy rules, like using side-by-side pictures or tiny delays, to make these "smart jumps" happen faster. For a BCBA, this means fewer teaching hours and bigger skill sets for learners.

120articles
1984–2026year range
5key findings
Research Synthesis

What the research says

Stimulus equivalence is one of the most useful findings in behavior analysis for practitioners who teach language and academics. It is the ability of a learner, after being taught that A = B and B = C, to derive that A = C — without ever being directly taught that last relation. This is called transitivity. Two other emergent relations are also possible: reflexivity (A = A) and symmetry (if A = B, then B = A). Together, these properties let a skilled teacher get much more out of each teaching session.

Research in this cluster is largely optimistic. Studies with children, college students, and even animals show that equivalence classes can be built efficiently and then used to teach complex skills — categorization, statistical reasoning, reading, and more — with fewer direct trials than traditional instruction. Equivalence-based instruction (EBI) has been tested on topics ranging from preschool categories to graduate-level hypothesis testing, and it consistently outpaces or matches conventional teaching on both acquisition and retention.

Key Findings

What 120 articles tell us

  1. Teaching A=B and B=C can yield A=C, B=A, C=B and more without direct training — significantly multiplying the output of each teaching session.
  2. Equivalence-based instruction works across a wide range of skills and ages, from preschool categories to graduate-level statistics.
  3. Similar stimuli must be presented side-by-side as comparisons, not just sequentially, for equivalence class formation to succeed.
  4. Having learners create graphic organizers or take notes during training doubles class formation yields.
  5. Talk-aloud protocols should be used only after initial equivalence tests — verbal mediation during training can suppress class formation.
Free CEUs

Get 60+ CEUs Free in The ABA Clubhouse

Live CEU every Wednesday — ethics, supervision, and clinical topics. Always free.

Join Free →
Deeper Dive

What else the research shows

The research also identifies failure points. Equivalence does not automatically emerge just because reinforcement is programmed. Some learners need multiple exemplar training to learn which emergent relations pay off. Others fail because of procedural details: showing similar stimuli only sequentially instead of side by side, using clustering sorts instead of stacking sorts, or having learners talk aloud before completing initial tests. Each of these is a fixable problem, not evidence that the learner cannot form equivalence classes.

Recent studies add new tools. Using graphic organizers, note-taking procedures, math problems as precurrent responses, and sorting tests have all been shown to expand or test equivalence class formation efficiently. These additions are easy to implement in clinical settings and can make the difference between a learner who forms classes quickly and one who needs many more sessions to achieve the same outcome.

Monday Morning Actions

How to apply these findings

When you are teaching any skill that involves matching or categorizing, consider whether EBI could save you trials. Structure your teaching so that you directly train a sample of relations (A-B, B-C) and then probe for derived relations (A-C, C-A) before teaching them directly. Research shows that many learners will derive untrained relations after a small amount of training. If they do, move on. Only directly teach the relations that did not emerge. This approach can cut your total trial count significantly for vocabulary, category, and concept programs.
If a learner fails equivalence probes, check your procedure before concluding they cannot form classes. Research identifies several fixable problems: stimulus location inconsistency, format of comparison presentation, and the presence of verbal mediation tasks during training. Adjust one variable at a time. Try showing similar stimuli side-by-side during conditional discrimination training. Try a stacking sort format instead of clustering. Remove any talk-aloud prompts during initial training and add them only after probes are complete.
Use graphic organizers during EBI whenever you are teaching complex multi-member classes. Research shows that having learners draw their own visual representation of the relations being trained dramatically increases class formation even with five-member abstract classes. This is easy to implement: give the learner a blank organizer template and have them fill it in as training progresses. The act of organizing the information builds the mediation that drives emergent responding.

Frequently Asked Questions

Common questions from BCBAs and RBTs

In regular discrete trial teaching, you directly teach every relation you want the learner to know. In equivalence-based instruction, you teach a subset of relations and then probe to see which untrained relations emerged. Research shows that many learners derive the remaining relations without direct training. EBI uses the same trial structure but reduces total teaching time by exploiting emergent learning.

Probe for it after training A-B and B-C relations. Test for C-A, A-C, and B-A without any training. If the learner passes those probes at or above your mastery criterion, equivalence emerged. If not, check the procedural variables — stimulus presentation format, sort format, verbal mediation — before concluding the learner cannot form classes.

Research identifies several culprits: similar stimuli presented sequentially instead of side-by-side, clustering sorts instead of stacking sorts, and learners talking aloud during training. Each of these is a procedural fix. Change one variable at a time, re-run the training, and probe again before drawing conclusions about the learner.

Yes. Research shows EBI has been used to teach statistical test selection to graduate students, categorization to preschoolers with ASD, and note-taking to college students — all with strong results. The method works wherever you can define a set of relations among stimuli, which covers most academic and language content.

Research shows that having learners draw their own graphic organizers during training doubles class formation yields, even with complex five-member abstract classes. The act of visually organizing the relations builds the mediation needed for emergent responding. Provide a blank template and have the learner fill it in as they learn each relation.