Are preference and resistance to change convergent expressions of stimulus value?
Brief extinction probes give a false read on reinforcer value; use resistance-to-disruption tests instead.
01Research in Context
What this study did
The team worked with pigeons in a lab. They wanted to know if two common measures of reinforcer value tell the same story.
First they trained two keys on different reinforcement schedules. Then they tested each key two ways: a short extinction probe and a longer resistance-to-disruption test.
What they found
The birds pecked more on the key that paid off more often. That part looked normal.
But when the researchers added free food elsewhere, the high-rate key stayed strong while the brief extinction probe did not. The two value signals stayed opposite and never lined up.
How this fits with other research
LeBlanc et al. (2003) showed that richer schedules make both response rate and discrimination accuracy harder to disrupt. The 2013 study keeps that rule but shows brief extinction probes miss it.
Bai et al. (2016) later found that adding extra reinforcement for another response can backfire, making the target behavior stickier. Both papers warn that extra reinforcement can hide, not help, true persistence.
Gilroy et al. (2021) use elasticity tests to find which items really drive work. Their method agrees with the 2013 advice: skip short preference checks and use a test that pushes the behavior.
Why it matters
If you want to know how powerful a reinforcer is, do not trust a quick extinction probe alone. Run a short disruption test instead: keep the task but give free goodies elsewhere. The item that keeps the learner working is the one with true strength. Use that item for tough teaching steps and for guarding against relapse.
Get CEUs on This Topic — Free
The ABA Clubhouse has 60+ on-demand CEUs including ethics, supervision, and clinical topics like this one. Plus a new live CEU every Wednesday.
Pick your top two reinforcers, then run a 3-minute free-access distraction while the task stays available; the one that keeps responding wins.
02At a glance
03Original abstract
Behavioral momentum theory asserts that preference and relative resistance to disruption depend on reinforcement rates and provide converging expressions of the conditioned value of discriminative stimuli. However, preference and resistance to disruption diverge when assessing preference during brief extinction probes. We expanded upon this opposing relation by arranging target stimuli signaling equal variable-interval schedules across components of a multiple schedule. We paired one target stimulus with a richer reinforced alternative and the other with a leaner alternative. Furthermore, we varied reinforcement rates for the paired alternatives to assess the effects of manipulating relative conditioned value on preference and resistance to disruption by presession feeding, intercomponent food, and extinction. We replicated the opposing relation between preference and resistance to disruption but varying reinforcement rates for the paired alternatives did not systematically affect preference or resistance to disruption beyond levels observed in our initial condition. Importantly, we found that only preference between the target stimuli was related to relative baseline response rates in the presence of those stimuli. These findings suggest that preference during extinction probes might reveal more about baseline response rates between concurrently available alternatives than relative conditioned value. Resistance to disruption, conversely, appears to better reflect conditioned value because it is less confounded with baseline response rates and is a function of all sources of reinforcement obtained in the presence of a stimulus context.
Journal of the experimental analysis of behavior, 2013 · doi:10.1002/jeab.33