Comparing the reinforcing efficacy of tokens and primary reinforcers

Two adults worked for tokens or for snacks on a progressive-ratio schedule. The schedule asked for more button presses each time to earn the next reward.

First the team paired new plastic tokens with favorite snacks. Then they compared how hard each person worked for the tokens versus the snacks themselves.

Both people pressed more for the snacks than for the new tokens. The gap stayed large across three tests.

In plain words, the just-trained tokens were weaker reinforcers than the food they were paired with.

Krentz et al. (2016) saw the opposite in the real world. Adults with ID walked three times more laps when staff paid with tokens. The difference is setting: Krentz used months of practice in a day program, while Bonfonte tested brand-new tokens in a lab.

Skrtic et al. (1982) also found tokens added no extra punch. Stuttering clients kept fluent speech after tokens were removed. Like Bonfonte, the study says tokens may not be needed once behavior is strong.

Thomas et al. (1988) warns that mixing reinforcers can back-fire. They showed social praise alone beat praise-plus-candy, and Bonfonte’s data fit the same theme: simpler, high-quality reinforcers often win.

Before you swap snacks for tokens, run a quick preference check. Ask the client to do a small task for each option and see which one keeps the response high. If tokens lose, spend more time pairing them with strong backup items or keep the primary reinforcer in the mix until the token value builds.