ABA Fundamentals

Optimizing equivalence-based instruction: Effects of training protocols on equivalence class formation.

Fienup et al. (2015) · Journal of applied behavior analysis 2015
★ The Verdict

Teach equivalence classes step-by-step with probes after each new stimulus, not in one big block.

✓ Read this if BCBAs who use matching-to-sample or equivalence-based instruction with teens or adults.
✗ Skip if Clinicians focused only on early childhood verbal behavior with no equivalence component.

01Research in Context

01

What this study did

The team compared two ways to teach college students new equivalence classes. One group got simple-to-complex training: easy matches first, then probes mixed in. The other group got all matches at once, then a big test block.

They used an alternating-treatments design so each learner tried both methods. The goal was to see which path built three- and four-member classes faster.

02

What they found

Simple-to-complex won. Every adult formed classes right away with that method. The simultaneous block method worked for only three out of four people on three-member sets and fewer than half on four-member sets.

In plain words, sprinkling quick checks between easy steps beats cramming everything and hoping it sticks.

03

How this fits with other research

Frampton et al. (2023) extends the win: when learners also drew graphic organizers during simple-to-complex training, scores jumped from 75% to 100%. The same basic sequence, just with a note-taking boost.

Williams et al. (2025) conceptually replicates the pattern. Their rule-plus-examples sequence (another step-by-step approach) beat massed practice every time, showing the wider rule: feed material in small, ordered bites.

Whitehouse et al. (2014) seems to clash at first glance. That study found massed trials beat spaced trials for teaching tacts to kids with autism. The difference is content and probes. Whitehouse et al. (2014) massed tact repetitions with no early tests; the target paper interspersed probes during simple-to-complex training. Early checks, not just massing, appear key for equivalence.

04

Why it matters

If you run matching-to-sample lessons for staff training or client academics, swap from big test blocks to a simple-to-complex flow. Start with easy relations, drop in quick probe trials, then grow the set. You will likely see mastery sooner and cut re-teach time. Try it next session: teach two new stimuli, probe once, add the third, probe again—keep the cycle tight.

Free CEUs

Want CEUs on This Topic?

The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.

Join Free →
→ Action — try this Monday

Insert a quick probe trial right after the first new relation, then keep adding stimuli one at a time with probes.

02At a glance

Intervention
stimulus equivalence training
Design
alternating treatments
Population
neurotypical
Finding
positive
Magnitude
large

03Original abstract

Two experiments evaluated the effects of the simple-to-complex and simultaneous training protocols on the formation of academically relevant equivalence classes. The simple-to-complex protocol intersperses derived relations probes with training baseline relations. The simultaneous protocol conducts all training trials and test trials in separate portions of the protocol. In Experiments 1 and 2, participants formed 4 3- and 4-member neuroanatomy classes, respectively. When trained with the simple-to-complex protocol, 100% of participants immediately formed the 3- or 4-member classes. When trained with the simultaneous protocol, the 3- and 4-member classes were formed immediately by 75% and 42% of participants, respectively. Thus, the immediate emergence of equivalence classes was an interactive function of training protocol and class size. The remaining participants eventually formed classes after a few cycles of retraining. The incorporation of these training and testing parameters could optimize the use of equivalence-based instruction for teaching college-level course content.

Journal of applied behavior analysis, 2015 · doi:10.1002/jaba.234