Concurrent performances: stimulus-control gradients during schedules of signalled and unsignalled concurrent reinforcement.
Concurrent food signals flatten stimulus-control gradients, so extra reinforcers can widen the range of cues a learner responds to.
01Research in Context
What this study did
Pigeons pecked a key for food. The key showed lines tilted at different angles.
Blue or red lights sometimes signaled that food was available on another key. A white light meant no food there.
The team tracked how line angle controlled pecking when these extra signals were present.
What they found
The blue and red signals flattened the birds’ angle-response curve. The white signal kept it steep.
In plain words: concurrent food cues made the birds less picky about line tilt.
How this fits with other research
Staddon (1972) showed that fancier visual patterns push higher response rates. C et al. now add that any concurrent food signal, plain or fancy, widens the range of angles the bird will accept.
CATANIDINSMOOR (1962) proved that a short changeover delay keeps concurrent VI schedules from bleeding into each other. The 1974 study used that same delay, so the gradient flattening is not just schedule spill-over.
Bloomfield (1967) found quick behavioral contrast when reinforcement shifted. C et al. show the shift can be in stimulus control, not only response rate.
Why it matters
When you run two reinforcers at once, the stimuli that signal them can blur stimulus control. If you want crisp discrimination, keep the extra reinforcer unsignaled or separate in time. Check your client’s “error” pattern—it may be a normal by-product of concurrent rewards, not faulty teaching.
Want CEUs on This Topic?
The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.
Join Free →Before adding a second reward, decide if you need tight stimulus control; if yes, keep the new reward unsignaled or in a different location.
02At a glance
03Original abstract
On one key, pigeons' pecks were reinforced according to a variable-interval schedule in the presence of vertical lines, and were not reinforced in the presence of oblique lines. On a second key, pecks were reinforced according to a variable-interval schedule in the presence of blue, according to a signalled variable-interval schedule in the presence of red, and were not reinforced in the presence of white. Subsequently, during extinction, stimulus-control gradients were obtained by presenting eight different line orientations on the first key concurrent with each of the three colors on the second key. On the first key, line-orientation gradients tended to be lower, narrower, and less shifted in peak or area when the second-key stimulus was blue or red, the stimuli respectively correlated with unsignalled and signalled reinforcement, than when it was white, the stimulus correlated with extinction. Thus, the effect on first-key line-orientation gradients depended on second-key stimuli correlated with concurrent reinforcement, whether or not these stimuli were also correlated with concurrent responding. As a function of first-key line orientation, an inverted gradient was obtained on the second key during blue; during both red and white, rates of pecking on the second key were near zero.
Journal of the experimental analysis of behavior, 1974 · doi:10.1901/jeab.1974.21-99