Select and reject conditional control on matching to sample and stimulus equivalence

Plazas and colleagues asked adults to learn four three-member classes with pictures and nonsense syllables.

Some adults got detached MTS: every trial taught both pick-the-correct-item (select) and reject-the-wrong-item (reject).

Other adults got only select training, only reject training, or the usual MTS mix.

The team then tested who formed true equivalence classes without extra coaching.

Detached MTS beat the rest.

Most adults who practiced both select and reject showed full symmetry, transitivity, and equivalence right away.

Adults trained with only select, only reject, or standard MTS often looked like they had classes but failed stricter tests, a sign of pseudoequivalence.

Foti et al. (2015) already showed that select control alone gives stronger classes than reject control alone.

Plazas et al. (2025) now shows combining both beats select alone, so the two studies line up like steps on a ladder.

White et al. (1990) warned that kids can form fake equivalence when procedures hide extra cues; the new study gives a fix—train both relations from the start.

Cohen-Almeida et al. (2000) proved clients with little language can still build classes, hinting that detached MTS might help that group too.

If you run matching-to-sample lessons, add reject trials instead of only asking clients to point to the right card.

Mixing "pick the same" and "cross out the different" within the same lesson creates sturdier stimulus classes and cuts the risk of pseudoequivalence.

Try it next time you teach coin names, sight words, or category members—your learners may need fewer review blocks before the whole class clicks.