Effects of response rate, reinforcement frequency, and the duration of a stimulus preceding response-independent food.
A quick pre-reinforcement cue can spike responding, but keep it too long and fast responders slow down.
01Research in Context
What this study did
Pigeons pecked a key for food on a regular schedule. Sometimes a light came on right before extra food dropped, no peck needed. The team varied how long that light stayed on and how fast the birds pecked before it appeared.
They wanted to know if a short 'food-is-coming' cue would speed up or slow down pecking, and whether the bird's own baseline rate mattered.
What they found
Quick flashes (a few seconds) boosted pecking for both slow and fast birds. Longer lights helped slow birds a little but put the brakes on fast, steady peckers.
So the same cue can excite or suppress responses, depending on its length and how busy the bird already was.
How this fits with other research
Henton et al. (1970) first showed an 80-second pre-reward light speeds up lever presses. Kendall (1974) now adds the twist: push the duration past a few seconds and high-rate pecking drops.
Harper (1996) later found extra free food slows responding, calling it a 'behavioral momentum' hit. The 1974 study foreshadows this: longer pre-food stimuli act like extra food, dampening already-fast rates.
Edwards et al. (1970) saw only tiny slowdowns on fixed-ratio schedules unless they doubled the free-food rate. Kendall (1974) shows the slowdown can appear faster if you simply lengthen the cue, no extra food needed.
Why it matters
When you use response-independent reinforcement or warning stimuli, watch the clock. A brief 'good news' cue can momentarily energize a learner, but let it linger and you may suppress high-rate appropriate behavior. Start short, measure baseline speed, and adjust duration to keep the response you want.
Want CEUs on This Topic?
The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.
Join Free →If you flash a 'token coming' card before delivering free tokens, cap the display at 2-3 s and track whether the client's response rate rises or falls.
02At a glance
03Original abstract
Food-reinforced key pecking in the pigeon was maintained under a four-component multiple schedule. In two components, responding was maintained at high rates under a random-ratio schedule. In the other two components, responding was maintained at low rates under a schedule that specified a minimum interresponse time. For both high and low response rates, one of the schedule components was associated with a high reinforcement frequency and the other components with a lower reinforcement frequency. During performance under these schedules, a stimulus terminated by access to response-independent food was periodically presented. The duration of this pre-food stimulus was 5, 30, 60, or 120 sec. Changes in rate of key pecking during the pre-food stimulus were systematically related to baseline response rate and the duration of the stimulus. Both high and low response rates were increased during the 5-sec stimulus. At longer stimulus durations, low response rates were unaffected and high response rates were decreased during the stimulus. For two of three pigeons, high response rates maintained under a lower frequency of reinforcement tended to be decreased more than high response rates maintained under a higher reinforcement frequency. In general, the magnitude of decrease in high response rates was inversely related to the duration of the pre-food stimulus.
Journal of the experimental analysis of behavior, 1974 · doi:10.1901/jeab.1974.21-215