The functional independence of mands and tacts.

Rast et al. (1985) wrote a theory paper. They asked how mands and tacts link to the same toy, snack, or picture.

The authors said stop teaching the word first. Start with the item the child wants or sees. Then pile mand, tact, and little autoclitic frames onto that one referent.

They gave no new data. They drew a map for later experiments.

The paper found no numbers. It gave a plan. Teach around the object, not the response form.

If you do this, the same item can later cue asking, labeling, and commenting without extra drills.

Meier et al. (2012) and Lancioni et al. (2009) later tested kids with autism. They showed that teaching a mand often sprouts a brand-new tact for the same item. Their data extend the 1985 idea into real skill emergence.

Rojahn et al. (1987) went further. They mixed mand trials with tact trials on the same toy parts. Preschoolers learned the labels faster than with tact-only drills. The result lines up with the convergent-referent call.

Ribeiro et al. (2020) seems to push back. When they tried to teach a second tact to a picture the child already labeled, learning slowed down. The clash is only surface-deep: their task forced two different names for one picture, while the 1985 plan wants one name with many verbal functions.

You can save trials tomorrow. Pick the child’s favorite snack. Run a quick mand trial ("gimme cracker"), then a tact trial ("cracker") while the box is still in view. End with an autoclitic ("I see a big cracker"). One item, three operants, no extra stimuli. Watch for free emergence before you program more targets.