Derived manding in children with autism: synthesizing Skinner's verbal behavior with relational frame theory.

Murphy et al. (2005) worked with three children with autism.

They first taught each child to mand for one item by name.

Next they used matching-to-sample to build equivalence classes.

Each class linked the mand object to two new pictures.

Finally they tested if the kids would mand for the new pictures without further teaching.

All three children asked for the untrained items right away.

The mands worked even though no one had directly taught them.

The kids also showed the items could still work as reinforcers.

Equivalence training let the mand function jump to new stimuli.

Barnes et al. (1990) showed children need at least a two-year verbal age to form equivalence classes.

Carol’s study extends that finding to kids with autism who meet that level.

Jennings et al. (2017) later showed adults can build equivalence through tact and intraverbal training alone.

Carol’s procedure still used matching-to-sample, so the field has since moved toward more verbal-only methods.

Valentino et al. (2019) reviewed 45 mand studies and found most skip checks for equivalence or emergent control.

Carol’s paper is one of the few that actually tests and confirms emergent mand transfer.

You can expand a child’s request repertoire without teaching every single item.

Set up equivalence classes around high-interest mands, then probe for untrained requests.

This saves teaching time and builds a flexible verbal system.

Before you start, check the learner’s verbal age to be sure equivalence is possible.