Constructed-response matching to sample and spelling instruction.

Dube et al. (1991) built a computer spelling game for two people with intellectual disability.

The screen showed a spoken word, then the learner picked letters one by one to spell it.

Hints faded out as the player got better, and old words came back for review.

Both learners spelled new words correctly after the game training.

They kept the words when the computer mixed them with older lists later.

LeFrancois et al. (1993) copied the idea but used a delayed matching format. Words appeared first, then the choice came five seconds later. Their students also learned to spell and even matched words they had never been taught, showing the 1991 method can bloom into full equivalence classes.

DeRoma et al. (2004) moved the same delayed-cue trick to children with autism. Instead of spelling, they taught picture-symbol links. The kids learned faster than with an older exclusion method, proving the timing tweak travels across diagnoses and targets.

Tenneij et al. (2009) flipped the timing the other way: they showed choices first, then the sample five seconds later. Error rates for adults with ID dropped sharply. Together these three papers show that a small pause—before or after the sample—can boost matching accuracy no matter the content.

If you run matching-to-sample lessons, slip in a short delay. A three-to-five second gap before or after the sample can cut errors and build new relations without extra drills. Try it next session: present the comparison cards, wait five seconds, then show the sample. Watch accuracy rise.