Equivalence class formation when responding is separated from sample and comparison stimuli: Working memory, priming, and sorting
Hide the choices before the response and you can double equivalence-class formation in typical adults.
01Research in Context
What this study did
Fields et al. (2021) tested a small timing trick. They asked adults to pick the matching picture after the pictures had already left the screen.
The task is called trace-stimulus-pairing with a two-response window. First the sample and choices flash on, then they vanish, then the person presses a key.
Eleven neurotypical adults built three- and four-member equivalence classes this way. The team also ran a regular matching task where choices stayed on while the person responded.
What they found
When the pictures disappeared before the response, every adult formed the classes. In the regular task only about half did.
Doubling the success rate took one simple change: separate the response from the pictures. The gap let working memory do its job.
How this fits with other research
Ayres-Pereira et al. (2025) seem to say the opposite. They found that keeping the pictures on the screen, side-by-side, is what helps people learn look-alike items. The two studies do not clash. Fields moved the response late; Ayres-Pereira moved the pictures close together. Timing and layout are different levers.
Boldrin et al. (2022) tweaked how many choices show up. They proved two choices work fine if you rotate the foils. Fields adds a second tweak: when the learner clicks also matters.
Murphy et al. (2014) boosted scores by overtraining to five hundred trials. Fields gets the same jump in half the trials just by inserting a short pause.
Why it matters
If you run equivalence-based language programs, slide in a brief pause. Hide or remove the array, then let the learner respond. This tiny edit can turn non-formers into formers without extra trials or fancy software. Try it next time you teach sight-word sets, category sorts, or intraverbal webs.
Want CEUs on This Topic?
The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.
Join Free →After the comparison array appears, clear the screen, then prompt the learner to click or say the answer.
02At a glance
03Original abstract
The experiment determined whether equivalence class formation required overlap of comparison stimuli and responding. Each trial contained a sample first, a single, nonoverlapping comparison second, and a nonoverlapping response-window (RW) third, during which the participant made one of two responses (2R). All 11 participants formed two 3-member ABC equivalence classes using these "trace-stimulus-pairing two-response with response window" (TSP-2R-RW) trials. After adding a fourth stimulus (D) by CD training, ABCD tests showed immediate expansion to 4-member ABCD classes. When 4-member probes (AD, DA, BD, DB, CD, DC) were administered without 3-member probes, many participants showed decrements in class-indicative responding that then resurged to mastery with test repetition. Thus, 3-member probes enhanced class expansion. Class formation occurred for all participants when responding was temporally dissociated from the comparisons. In a matched, contemporaneously published experiment, where responding occurred during comparisons, only 54% of participants formed the classes. Thus, the comparison-response-separation nearly doubled class formation. Additionally, a special post-class-formation sorting test documented the emergence of two explicit equivalence classes. Finally, we noted a 1:1 correspondence for TSP-2R-RW and priming trials. Since priming measures neural substrates of equivalence classes, TSP-2R-RW trials should do the same.
Journal of the Experimental Analysis of Behavior, 2021 · doi:10.1002/jeab.651