The role of elicited responding in behavioral contrast.
Behavioral contrast includes reflex-like pecks, so control or measure them before you call a change learning.
01Research in Context
What this study did
Rodewald (1974) worked with pigeons on a two-key setup.
One key stayed the same color all the time. The other key changed color when food was coming.
The team counted every peck to see which ones were driven by the color change itself.
What they found
When food odds rose, birds pecked the color-change key more. They also pecked the steady key less.
Many extra pecks were pure stimulus reactions, not true operant responses.
How this fits with other research
Mulvaney et al. (1974) ran the same birds and saw the same jump in pecking. They just called it "autopecking" instead of "elicited." The two papers are direct replications with different labels.
Terrace (1974) adds a twist. He showed that birds first freeze on the steady key before the big burst. That freeze predicts how large the later burst will be, extending the dual-response idea to include active non-responding.
Crossman et al. (1973) set the stage. Their long-interval schedule kept stimulus control strong across tests, the same schedule Rodewald (1974) used to keep contrast clear.
Why it matters
If you see sudden rate jumps during schedule changes, ask: are these true skill gains or just stimulus-bound reactions? Watch for extra responses or brief freezes. Block or measure them so your contrast data stay clean.
Want CEUs on This Topic?
The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.
Join Free →Add a brief stimulus-only trial with no reinforcement to count elicited responses before your next contrast probe.
02At a glance
03Original abstract
An attempt was made to separate operant and elicited pecks occurring in multiple schedules of food reinforcement by moving the component stimuli to a second key, upon which pecks had no effect. The operant key stimulus was constant, regardless of the reinforcement schedule in effect. Experiments included two- and three-component multiple schedules and a comparison of the single-key and the two-key procedures. In general, conditions that typically produce positive contrast in single-key procedures reduced responding to the constant-stimulus key (induction) and increased responding to the component-stimulus key (contrast) in the two-key procedure. The results were interpreted as supporting the contention that two response classes, operant and elicited, are present in standard multiple schedules. In addition, elicited responses were strongly implicated in contrast phenomena.
Journal of the experimental analysis of behavior, 1974 · doi:10.1901/jeab.1974.21-249