ABA Fundamentals

Reinforcement and substitution in humans: a multiple-response analysis.

Bernstein et al. (1978) · Journal of the experimental analysis of behavior 1978
★ The Verdict

Making fun activities contingent on work tasks reliably increases work behavior through both reinforcement and substitution effects.

✓ Read this if BCBAs working with clients who avoid non-preferred tasks
✗ Skip if BCBAs focused solely on behavior reduction without skill building

01Research in Context

01

What this study did

Researchers asked adults to work on two boring tasks. One task earned points for a fun activity later. The other task earned nothing.

They tracked how much time people spent on each task. They wanted to see if making the fun activity contingent on the boring task would increase work on the boring task.

02

What they found

When the fun activity was contingent on the boring task, people spent 3-4 times more time on that task.

But here's the twist: people didn't just switch from the boring task to the fun one. They also started doing other random activities more often. This shows both reinforcement and behavioral substitution at work.

03

How this fits with other research

This builds on PREMACK et al. (1963) who first showed substitution in rats. When rats couldn't run on their wheel, they ate more food. The 1978 study proves the same principle works in humans with reinforcement added.

Zerger et al. (2016) extended this to preschoolers. They found that making adult attention contingent on physical activity increased kids' exercise. Same mechanism, different age group.

Lipschultz et al. (2017) seems to contradict this at first. They found that giving preferred items for free didn't increase compliance. But this actually supports the 1978 finding - contingency is key. Free items don't work, contingent access does.

04

Why it matters

You can use this today. Pick a highly preferred activity your client loves. Make access to it contingent on completing a less-preferred task first. The data shows this reliably increases engagement in the less-preferred activity. Just remember - the effect works because of the contingency, not just having the preferred item available.

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

Pick one non-preferred task and one highly preferred activity. Make the preferred activity available only after 5 minutes of the non-preferred task. Track the time spent on each.

02At a glance

Intervention
differential reinforcement
Design
single case other
Sample size
3
Population
neurotypical
Finding
positive
Magnitude
large

03Original abstract

Three adult human subjects engaged in activities such as reading, sewing, artwork, and candlemaking while living alone in a laboratory apartment 24 hours per day for several weeks. After a baseline period in which the activities were fully available, access to a particular activity (contingent response) was made dependent on engaging in another less-preferred activity (instrumental response). The contingencies produced substantial increases in instrumental responding, and responding decreased toward baseline levels when the dependency was removed. Under the contingent conditions, time earned for the concurrent activity was always less than the baseline level. To determine the contribution of this reduction to the instrumental increase, access to the contingent activity was restricted in the absence of any dependency. The results indicated that increases among responses that filled the newly available time could be selective, e.g., artwork increased when reading was restricted but candlemaking did not. It was concluded that the reductions in the contingent response that accompany contingencies usually do not exclusively determine instrumental increases, but selective increases can contribute to the increase in time devoted to the instrumental response.

Journal of the experimental analysis of behavior, 1978 · doi:10.1901/jeab.1978.30-243