The transfer of Cfunc contextual control through equivalence relations.

Researchers taught college students to match pictures and nonsense words into four-member equivalence classes. They then added a new cue that told the students when to press a key for money and when to hold back.

Next the team slipped brand-new pictures into the classes. The cue had never met these pictures, yet the students still followed its money rule. The cue's control had hopped through the equivalence network.

Contextual control traveled two steps. It jumped to newly added class members the students had never seen with the cue. It also reached line patterns that were only distant 'relatives' through equivalence.

The transfer was clean and fast. No extra training was needed. One cue taught in one corner of the class quietly governed the whole network.

Perez et al. (2021) built on this idea. They showed one background color could flip a stimulus between three jobs: reward, escape, or extinction. Both studies prove context can ride equivalence highways, but Perez extended the ride to multiple functions.

Almeida-Verdu et al. (2008) moved the work out of the lab and into a hospital. Deaf children with cochlear implants formed auditory-visual classes just like the college kids, showing equivalence training travels across populations.

Foti et al. (2015) ran a sister study the same year. They found 'select' training (picking the correct match) created stronger transfer than 'reject' training (crossing out the wrong one). Together the two papers tell us both what kind of training works best and how far its control can spread.

If you teach a safety cue to one picture in a class, the cue may automatically govern every other picture the learner links to it. That means fewer direct teachings for you and faster generalization for the client. Try adding a new stimulus to an existing class and probe whether your contextual cue still holds. If it does, you just saved valuable teaching time.