Three versions of the additive theories of behavioral contrast.
Test each reinforcement schedule alone before you mix them; only then can you spot true behavioral contrast.
01Research in Context
What this study did
Wesp et al. (1981) looked at three ways to add up the causes of behavioral contrast.
They asked which version you can actually test with data.
The paper is pure theory: no kids, no pigeons, just logic and math.
What they found
Only the "weak" version can be proven right or wrong.
Past studies used setups that mix the pieces before measuring each one.
That makes the old numbers hard to read.
How this fits with other research
Innis (1978) ran pigeons on two schedules and saw negative contrast, but not positive.
Those data fit the weak-additive story the 1981 paper says is testable.
Lattal et al. (2024) go further: they treat contrast as resurgence under extinction.
They build on the same weak version, but stretch it to treatment relapse.
Boyle et al. (2018) later beg applied workers to study contrast; they cite the 1981 warning that we still lack clean tests.
Why it matters
If you want to predict how a client will react when one setting gets richer or poorer, you must first measure each schedule alone.
Then combine them and watch for swing-back.
That single baseline step turns a fuzzy idea into a checkable claim.
Want CEUs on This Topic?
The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.
Join Free →Map baseline rates in each setting separately before you change reinforcer size or rate in either one.
02At a glance
03Original abstract
The additive theories of behavioral contrast state that contrast will occur only when two types of responses interact during multiple schedules. Three more specific versions of the theories may be defined according to how they distinguish these two types of responses. A strong version physically distinguishes them. A second version distinguishes them according to the theoretical processes which control them. A weak version distinguishes them on the basis of the environmental relations which control them. Only the weak version of the theories is currently testable. The weak theory should be tested by establishing each of the two environmental relations independently and then combining them to assess their effect on behavior. Because this test is not usually performed, many of the results which have been taken to support or contradict the additive theories are actually ambiguous.
Journal of the experimental analysis of behavior, 1981 · doi:10.1901/jeab.1981.36-285