On the relation between reinforcer efficacy and preference.

Doughty et al. (2010) asked a simple question. Does liking an item mean it will work as a reinforcer?

They tested two people. First they ranked items by preference. Then they checked if those items really made the person work harder.

For one person, top picks matched strong reinforcers perfectly. For the other, only some high-preference items worked.

Weak reinforcers were the trouble spot. They looked good on paper but did nothing in the session.

Fritz et al. (2020) saw the same gap. Their MSWO picked the best edible for only three out of four clients. Both papers warn that even standard assessments can miss the mark.

Kodak et al. (2009) add another twist. Different preference formats gave different winners. When formats disagree, reinforcer tests become even more important.

Allan et al. (1991) set the stage. They showed systematic beats caregiver guesswork. Doughty et al. (2010) refine that message: check each item, not just the list.

You already run preference assessments. Keep doing that, but add a one-minute reinforcer probe before treatment. Present the top two items in a quick task. If one fails to boost responding, drop it. This tiny step saves you from weak reinforcers and keeps your program humming.