An evaluation of resistance to change with unconditioned and conditioned reinforcers.
Tokens make behavior tougher against extinction and distraction, but food wins after the client has eaten.
01Research in Context
What this study did
The team compared token and food reinforcers head-to-head. Adults without disabilities pressed buttons for both items on equal schedules.
They then tested how long each behavior lasted during extinction, distraction, and pre-feeding.
What they found
Tokens kept the button pressing going longer when food was removed or when noise distracted the learner.
Surprise twist: after the adults ate first, food-reinforced pressing lasted longer than token-reinforced pressing.
How this fits with other research
Lejuez et al. (2001) showed the same resistance pattern in people with severe ID using computer tasks. The new study proves the effect also holds for neuro-typical adults using tokens.
Nevin et al. (2005) found faster conditioned-reinforcer delivery raised response rate but not resistance. K et al. now show the TYPE of reinforcer, not just its rate, can boost resistance.
KELLEHER (1961) warned that conditioned reinforcers can slow extinction. K et al. give the first clear data showing tokens do exactly that, except when the client is already full.
Why it matters
Use tokens when you need behavior to survive brief periods without food or when distractions pop up. Switch back to primary food if the client has just eaten and you want responding to stay high. Match the reinforcer to the likely disruption and you will get sturdier behavior.
Want CEUs on This Topic?
The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.
Join Free →Start your session with tokens; keep a few bites of food ready for post-snack work.
02At a glance
03Original abstract
Several reinforcer-related variables influence a response's resistance to change (Nevin, 1974). Reinforcer type (i.e., conditioned or unconditioned) is a reinforcer-related variable that has not been studied with humans but may have clinical implications. In Experiment 1, we identified unconditioned and conditioned reinforcers of equal preference. In Experiments 2, 3, and 4, we reinforced participants' behavior during a baseline phase using a multiple variable-interval (VI) 30-s VI 30-s schedule with either conditioned (i.e., token) or unconditioned (i.e., food; one type of reinforcement in each component) reinforcement. After equal reinforcement rates across components, we introduced a disruptor. Results of Experiments 2 and 3 showed that behaviors were more resistant to extinction and distraction, respectively, with conditioned than with unconditioned reinforcers. Results of Experiment 4, however, showed that when prefeeding disrupted responding, behaviors were more resistant to change with unconditioned reinforcers than with conditioned reinforcers.
Journal of applied behavior analysis, 2015 · doi:10.1002/jaba.226