Effects of within-class differences in sample responding on acquired sample equivalence.

The team worked with pigeons on many-to-one matching tasks. Birds had to peck the center key first, then pick a side key that went with that sample.

For half the birds, every sample in one class needed the same pecking rhythm. For the other half, each sample in a class needed a different rhythm. After training, the birds saw new combinations to test if the samples acted as equivalent.

Only the birds that used one rhythm per class showed equivalence. The birds that switched rhythms did not.

Same response = new stimuli acted alike. Different responses = no equivalence.

van Schrojenstein Lantman-de Valk et al. (2006) saw no sign that response patterns entered equivalence classes. The 2008 study shows why: the earlier work let birds use different peck styles within a class. Uniform responding was the missing piece.

Eugenia Gras et al. (2003) taught pigeons to use DRL vs FR pecking as silent cues. The 2008 paper keeps that idea but adds the rule: one class, one style.

Tassé et al. (2013) later found symmetry and expansion with response members, backing the 2008 claim that responses can join classes when they stay the same across samples.

When you set up conditional-discrimination programs, ask learners to respond the same way to every stimulus in an equivalence class. One topography per class helps the stimuli become interchangeable. Try it next time you teach coin names, sight words, or social phrases: keep the prompt or response form steady across items that must act as the same.