Extinction of a heterogeneous chain after several reinforcement schedules.
Discriminative stimuli between chain links slow extinction and preserve the response pattern.
01Research in Context
What this study did
The team built a two-step chain of different responses. Each step had its own light or sound cue. They compared extinction after this chain to extinction after a tandem schedule with no cues between steps. Pigeons served as subjects in a single-case lab setup.
What they found
The chain with cues held up longer during extinction. Birds kept following both steps and paid attention to the cues. The tandem group lost the pattern faster. Cues acted like glue for the response sequence.
How this fits with other research
Crossman et al. (1973) extends the idea: one big reinforcer at the end of a long interval also slows extinction. They added a terminal payoff; Jones (1969) added middle cues. Both tricks keep stimulus control alive.
Selekman (1973) adds a warning: if cues are weak, food deprivation can quadruple resurgence during extinction. Strong cues shield the chain; weak ones let deprivation break it.
Rees et al. (1967) did the groundwork. They showed that inserting external stimuli early in second-order DRL shifts response rate. Jones (1969) moved the same logic into extinction and found the chain survives longer.
Why it matters
When you fade reinforcement, leave discriminative stimuli in place. The stimuli keep the skill sequence intact and buy time for new reinforcement to take over. Use clear cues between steps of a task analysis, especially during extinction or thinning. Watch for resurgence if motivation is high and cues are weak.
Want CEUs on This Topic?
The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.
Join Free →Keep the picture cards, lights, or verbal cues between steps when you start extinction or reinforcement thinning.
02At a glance
03Original abstract
Rats were conditioned to emit the following two-member chain of responses on two different operanda always available: responses on a vertical bar produced a discriminative stimulus for food-reinforced responding on a horizontal bar. Responses on the vertical bar produced a discriminative stimulus on a variable-interval 1-min schedule, and the horizontal bar produced food on a variable ratio of 10 responses. Control conditions were included in which vertical bar responses were also food-reinforced simultaneous with the onset of the discriminative stimulus for the horizontal bar response and a tandem schedule which had the same response requirements but without different exteroceptive stimuli associated with the separate components of the response chain. The latter condition greatly retarded acquisition of the response chain compared to the other schedules studied here and compared to reports in the literature on homogeneous (single operandum) response chains. Intermittent reinforcement of the chain led to greater resistance to extinction of both members and the chain remained intact longer in the sense that stimulus control was maintained.
Journal of the experimental analysis of behavior, 1969 · doi:10.1901/jeab.1969.12-127