Evaluation of a discrimination training procedure for establishing praise as a reinforcer

Sainsbury et al. (2024) taught three autistic kids to treat neutral praise words like "good" as reinforcers. They used a quick discrimination drill: the child picked between two picture cards. One card paid off with praise plus candy. The other card paid off with candy only.

After several rounds, the team removed the candy. They checked if the praise alone could keep the kids choosing the praise-linked card.

The training worked. All three children kept picking the praise-linked card even when candy was gone. The praise had become a short-lived reinforcer.

The catch: once the candy stopped, the praise power faded within the same session. You have to keep the candy loop alive or the effect disappears.

Dudley et al. (2019) used a different route. They paired praise with four different reinforcers instead of one. Their praise stayed strong longer. Sainsbury’s single-reinforcer method is faster but weaker.

Cox et al. (2015) looked at reinforcer size. They found praise-only trials taught new skills just as fast as big edible reinforcers. Sainsbury adds the missing how: you can build that praise-only power with a short discrimination drill.

DeQuinzio et al. (2015) taught kids to spot which adult actions earned rewards. Both studies use discrimination training, but Sainsbury flips the target: the child learns to value the praise itself, not just copy the rewarded action.

You can turn plain words into reinforcers in under an hour. Use this when edible or toy reinforcers are impractical—at the grocery store, in line, or during transitions. Keep a tiny stash of backup reinforcers handy. Slip one in every few trials to keep the praise alive. Fade the backup slowly, not all at once.