The effects of establishing operations on preference assessment outcomes.

The team ran paired-stimulus preference tests with four children who had developmental delays. They compared three conditions: withhold the items for 24-48 hours, let the kids eat or play with them for 10 minutes first, or give no special instructions.

Each child saw two items at a time and touched the one they wanted. The order of pairs changed every session so the data would be clear.

After deprivation, the kids picked the withheld items more often. After satiation, the same items lost appeal. The control condition sat in the middle.

The pattern held for every child, showing that a short break or a quick preload can swing the whole hierarchy.

McAdam et al. (2005) repeated the idea with toys instead of snacks and got the same lift-and-drop effect, so the finding crosses edible and tangible items.

Smith et al. (1997) looks like a clash at first—they saw food crush leisure items in rankings. But they never withheld or pre-fed the items; they just mixed them together. Once you test the displaced leisure items as actual reinforcers, they still work. The studies differ in method, not in truth.

Carter et al. (2020) push the story further. They show that even a clear preference list can mislead you when you move to real work tasks. A top-ranked item does not guarantee the strongest reinforcer—you still need a quick reinforcer check before treatment.

You can clean up messy preference data in one evening. Withhold the likely winners for a day, or let the client snack on them for 10 minutes, then test. The clearer hierarchy saves you from picking weak reinforcers and speeds up skill programs. Just remember to run a brief reinforcer assessment afterward, because preference and potency are not always the same thing.