Emergent Relations and Generative Performance: How Limited Teaching Produces Unlimited Skills
When you teach a client one skill, do they automatically learn five others? Sometimes yes—and understanding why matters deeply for your practice. This article walks you through emergent relations and generative performance: two core ABA concepts that explain how learners can produce skills they were never directly taught.
If you’re a BCBA, RBT, clinic director, or caregiver trying to build efficient, durable teaching programs, this post will help you recognize when emergence is happening—and more importantly, when to test for it rather than assume it.
One-Paragraph Summary
Emergent relations are learned connections between stimuli that appear without direct teaching of those specific relationships. Generative performance is the ability to produce novel, untrained responses by combining or applying patterns already learned. Several processes produce emergence: stimulus equivalence, multiple-exemplar training, and derived relational responding. An ABA example: teaching a child to match a picture to a written word and the written word to a spoken label often leads to matching the picture to the spoken label—without training that pair directly. A non-ABA example: a student taught multiplication facts and the distributive property can solve novel multiplication problems using that rule. This matters because one well-designed teaching sequence can yield many untaught skills, saving time while building truly flexible abilities. For assessment and teaching, this means we must test for emergent relations rather than assume they’ve occurred.
What Are Emergent Relations and Generative Performance?
Emergent relations are connections between stimuli that develop as a byproduct of training other, related connections—without direct teaching or reinforcement of those specific pairs. They emerge from the learner’s history and the relational structure you’ve built into your teaching.
Generative performance is what happens next: the learner uses those emergent relations to produce responses they’ve never been taught. A child who learns letter sounds can read novel words. A teenager who learns a social rule can apply it to new peer situations. Growth extends far beyond the specific examples you trained.
The key difference: taught responses are directly reinforced during training; emergent responses appear without direct training of that exact response. This distinction changes how you design lessons, what you measure, and when you can ethically rely on a skill working in the real world.
The Processes That Produce Emergence
Several well-documented processes reliably produce emergent relations. Understanding each one helps you design teaching that’s not just effective, but generative.
Stimulus equivalence is the foundation. When you teach a learner that Picture A matches Written Word B, and Written Word B matches Spoken Label C, untrained relations can emerge. The learner may then match Picture A to Spoken Label C—a pair you never explicitly trained. This happens because stimulus equivalence involves three core properties: reflexivity (A relates to itself), symmetry (if A relates to B, then B relates to A), and transitivity (if A relates to B and B relates to C, then A relates to C). Testing these untrained relations—called “probing”—reveals whether emergence has occurred.
Derived relational responding is the broader, language-like process from which stimulus equivalence emerges. It’s the ability to respond to relations you didn’t directly train because they logically follow from relations you did teach. If you teach that “bigger than” relates apple to watermelon, and watermelon to planet, a learner may derive that planet is “bigger than” apple—without training that specific comparison.
Multiple-exemplar training (MET) builds generativity through breadth and variation. Instead of teaching one chair, one dog, one way to ask for help, you teach across many examples: different chairs, different dogs, different contexts. This reduces stimulus rigidity. A child who sees “dog” only in picture books may call every four-legged animal a dog. But a child who’s seen dogs in photos, videos, cartoons, plush toys, and real life learns the boundaries of the class.
Naming and bidirectional naming unlock rapid language growth. When a learner can both tact (say the name of an object when shown it) and listen (pick the object when told its name), these two functions often support each other. Teaching one direction well can lead to emergence of the other. Naming also creates a “bridge” that makes other derived relations more likely.
Transformation of stimulus functions explains how the meaning or impact of a stimulus changes through its relations with other stimuli. If you learn that Brand X is linked to a frightening story, the brand itself may trigger anxiety—without any direct pairing. The function transforms across the relational chain.
Emergence Versus Generalization: A Critical Distinction
Many clinicians use these terms interchangeably, but they’re not the same—and conflating them can lead to missed deficits.
Emergence is the appearance of new relations or skills that weren’t directly taught. The relation between A and C emerges from training A↔B and B↔C.
Generalization is stable performance of a learned skill across new contexts, settings, or similar examples. A child who learned to greet the teacher can also greet the paraprofessional—that’s generalization. A child who learned A↔B and B↔C can now match A↔C without training—that’s emergence.
You can have generalization without emergence, and emergence without generalization. Good teaching often produces both, but testing them requires different probes.
How to Recognize and Test for Emergence
Emergent relations aren’t always obvious. Probing—deliberately testing untrained relations—is the only reliable way to know.
A probe presents an untrained relation and records whether the learner responds correctly, without prompting or feedback. If you’ve taught A↔B and B↔C, a probe for A↔C might look like: present the picture (A), ask “What do you hear?” (or match to C), and score the response with no correction. Cold probes—given without the lead-up of trained trials—give the clearest picture of what the learner can actually do.
Good probing is systematic. Probe across multiple sessions, in varied contexts, with different people, and at different times. If A↔C emerges consistently after A↔B and B↔C training, that’s strong evidence. If it doesn’t emerge, the learner needs more training, different exemplars, or a different approach.
Documentation matters too. Record which relations you trained, which ones you probed, the results, and when emergence appeared. This builds a clear picture of your learner’s acquisition pattern.
Why This Matters in Your Clinic or Classroom
Emergence is practically valuable and ethically essential.
On the practical side, designing teaching that produces emergence means you can teach less and accomplish more. One well-structured stimulus equivalence sequence might yield five or six untrained skills. Multiple-exemplar training helps learners apply skills in real life, not just in your clinic. This is what caregivers and teachers actually need: learners who can think, problem-solve, and adapt.
Ethically, emergence creates responsibility. If you assume emergence without testing, you might believe a learner can do something they actually can’t. A client who learned “hello” in three contexts might not have the flexible greeting skills you think they do. Relying on untested emergent skills can be unsafe, especially for high-stakes tasks like following safety rules or recognizing danger.
This is why the field emphasizes verification. Before you count a skill as learned and ready for real-world use, probe it. Document the probe. Repeat it across contexts and times.
When and How to Use Emergence in Your Practice
Emergence becomes a teaching strategy when you’re aiming for flexible, generative skills—which is most of the time.
Language and literacy are prime examples. Teaching letter-sound correspondences and phonetic rules sets up emergence of reading new words. Teaching tacts across multiple contexts often leads to emergence of mands and intraverbals.
Social skills and problem-solving benefit heavily. Teaching a conflict-resolution rule and practicing it with a few peer scenarios can lead to emergence of appropriate responding with new peers in novel situations.
Efficiency in large curricula is another clear use. If you’re teaching reading, math, or communication across many words, objects, or concepts, designing your teaching to produce emergence means fewer direct lessons and faster progress.
Before you design a lesson, ask yourself, “Can I set this up so emergence is likely?” If there’s a relational structure you can teach that will yield untaught skills, invest in that structure. Teach systematically, vary your exemplars, use probes, and document.
Common Mistakes and How to Avoid Them
Even experienced clinicians sometimes skip verification. Here are the most common pitfalls.
Assuming emergence without probing is the biggest risk. A caregiver might report, “My son learned to say ‘dog’ and now he says it for all animals.” But if you probe, you might find he’s overgeneralizing—calling cats and cows “dog” because the training didn’t establish firm boundaries. That’s not correct emergence; it’s a training error.
Using too few exemplars is another common mistake. If you teach “sit” only in the therapy room with one prompt style, you’ve set up a brittle skill. Expand to multiple contexts, people, and prompt styles.
Confusing simple transfer with derived relations happens when clinicians see a skill generalize to a new context and call it emergence. If a learner can match a picture to a word in both the clinic and home, that’s generalization—valuable, but not the same as deriving a new relation.
Ignoring reinforcement history and context control undermines emergence. A learner whose history includes reinforcement for narrow, rote responses may not show emergence even with good teaching. You may need to actively teach flexibility or reward novel responding.
To avoid these mistakes, treat probing as non-negotiable. When you’re tempted to say “The learner has generalized this skill,” pause and ask: “Have I actually tested the untrained relations?”
Emergent Relations in Real ABA Practice
Stimulus equivalence with sight words: You teach a child to match a picture of an apple to the written word “APPLE,” and then to match the written word “APPLE” to a spoken label. On a probe, you present just the spoken label and ask the child to pick the picture. They do—without ever training that specific pair. The relation emerges from the A↔B and B↔C relations you taught.
Why this matters: the child isn’t memorizing word-picture pairs; they’re learning the relational structure that lets them decode and recognize words flexibly.
Multiple-exemplar training with verbs: You teach “run” by showing different people running in different contexts. You vary your prompts. Once the child demonstrates mastery across these examples, you present a novel actor in a novel context and ask, “What is it doing?” The child says “running”—a response to a combination they’ve never seen.
Why this matters: the child has learned the feature that defines running, so they can recognize and label it anywhere.
Examples Beyond ABA
Emergence and generativity aren’t unique to ABA.
Math and the distributive property: A student learns that 3 × 4 = 12 and 3 × 5 = 15, and is taught the rule: 3 × (4 + 5) = (3 × 4) + (3 × 5). Now given 3 × 9, they can solve it using the rule, even though no one taught them “3 × 9” directly.
Vocabulary building: A reader learns several synonyms and learns that these words share a general meaning. Later, they encounter a new word and, using the pattern they’ve learned, infer it also means something similar. They may even use it correctly in an original sentence—without explicit instruction.
These examples show that emergence and generativity are fundamental learning principles. When you design teaching around relational structure and varied examples, learners naturally produce skills you didn’t directly teach.
How to Verify Emergent Skills Ethically
Start with systematic probing. Before you tell a parent, “Your child can do X,” run cold probes in at least two different sessions and contexts. Run a few more probes over the next week. Consistency across time and context builds confidence.
Document clearly. Write down what relations you trained, what you probed, the dates and results, and any contextual notes. This protects the learner and your practice.
For high-stakes skills, require extra proof. If the skill involves safety, functional independence, or major life domains, don’t rely on a few probes. Build in multiple probes, varied people, varied contexts, and time delay. A skill that emerged in the clinic should be re-tested at home, at school, and in the community.
Stay humble about emergence. Not every learner will show emergence for every relation. If emergence doesn’t appear, the learner may need more direct teaching, different exemplars, or additional support. Adjust and try again.
FAQs: Quick Answers
What if a learner doesn’t show emergent relations even after good teaching?
Add more exemplars, try different teaching formats, or scaffold the relation more explicitly. Some learners benefit from naming practice, video modeling, or peer examples. Document what you try and what works.
How many exemplars do I need before emergence is likely?
There’s no magic number. Use probes to guide you. If probes show no emergence after two exemplars, add a third or fourth. Once you see emergence, document how many exemplars it took.
Can I use emergent skills immediately in the real world?
Only after you’ve probed them successfully across contexts and time. Always probe in the real-world context before you rely on the skill.
What’s the difference between emergence and just remembering something?
Emergence is the logical derivation of a new relation from relations that were taught. Testing with novel combinations or after time delays helps distinguish true derived relations from recall.
Key Takeaways
Emergent relations and generative performance are the engine of flexible, efficient learning. When you teach relational structure—using stimulus equivalence, multiple-exemplar training, naming, and systematic variation—learners can produce skills far beyond what you directly taught.
But emergence must be verified, not assumed. Probing untrained relations is the only reliable way to know whether a learner truly acquired a skill. Ethical practice means documenting these probes, especially for skills that matter for safety or independence.
As you review your current teaching programs, ask: Where could I design for emergence? Which skills would benefit from multiple-exemplar training? Which relations should I be probing? Starting with those questions will help you build stronger, more generative learner repertoires—and practice with integrity.
