ABA Fundamentals

Some effects of combining reinforcers in operant training with mentally handicapped persons.

Thomas et al. (1988) · Behavior modification 1988
★ The Verdict

Adding tangible rewards to social praise offers no learning gain and can make the behavior stop sooner when rewards end.

✓ Read this if BCBAs teaching daily living skills to adults with intellectual disabilities.
✗ Skip if Clinicians already using high-quality token economies with built-in backup reinforcers.

01Research in Context

01

What this study did

The team worked with six adults who had intellectual disabilities. Each person learned a simple assembly task.

The trainers compared three reward types: social praise alone, small candy alone, or praise plus candy together.

02

What they found

Adding candy to praise gave zero extra learning speed. All three reward types taught the skill at the same rate.

When rewards stopped, praise-only learners kept working longest. The candy-plus-praise group quit fastest.

03

How this fits with other research

Bonfonte et al. (2020) saw the same pattern: one strong reinforcer beats a mixed pair. Their edibles beat new tokens, just like praise here beat praise-plus-candy.

Weinsztok et al. (2022) updated the story. They showed that bigger or better quality rewards can protect a treatment when staff slip up. The 1988 study used small candy; bigger rewards might have changed the outcome.

Krentz et al. (2016) extended the idea to natural settings. Their tokens tripled walking laps for adults with ID, showing single reward systems can still shine in real life.

04

Why it matters

You can skip the extra reinforcer. Pick the strongest, simplest reward and stick with it. Praise is cheap, easy, and keeps working when you fade it out. Save your edible budget for times you truly need a boost.

Free CEUs

Want CEUs on This Topic?

The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.

Join Free →
→ Action — try this Monday

Run a quick praise-only condition next session and track how long the client keeps working after you stop delivering praise.

02At a glance

Intervention
other
Design
quasi experimental
Population
intellectual disability
Finding
negative

03Original abstract

Two experiments investigated the conventional practice of combining social and tangible rewards as a compound reinforcer in operant training procedures for mentally retarded persons. In the first experiment, there were no significant differences in acquisition of a color discrimination between subjects receiving combined social and food reinforcement and those receiving only one of the component reinforcers. In subsequent extinction, subjects who had received social reinforcement on its own responded more than combined reward subjects, who did not differ in extinction responding from food reward alone subjects. In the second study, higher rates of a simple manipulative response were established during training with food reinforcement alone and with combined reinforcement than with social reinforcement alone. Nevertheless, resistance to extinction was again superior after social reinforcement alone than after the food or combined reinforcement. This was the case despite generalization decrements from acquisition contingencies to extinction being kept to a minimum in this second study by the use of noncontingent reinforcement as the extinction procedure. Neither experiment, therefore, found any beneficial effects of combining the rewards, either in acquisition or in subsequent extinction.

Behavior modification, 1988 · doi:10.1177/01454455880124003