Polydipsia induced in rats by second-order schedules of reinforcement.
Adjunctive polydipsia peaks after the real food pellet, not the brief cue, so the final reinforcer controls adjunctive behavior.
01Research in Context
What this study did
Corfield-Sumner et al. (1977) watched rats drink water while they earned food pellets.
The rats worked under second-order schedules. A brief light flashed before each pellet.
Some rats got fixed-interval food. Others got fixed-time food with no response needed.
What they found
Most licking happened right after the food pellet arrived.
The brief light alone caused only a small sip.
When food came on fixed-time, the light stopped triggering any drinks at all.
How this fits with other research
Bacon-Prue et al. (1980) extends this picture. They showed the pellet itself acts like a signal. Rats drink more after the pellet than after the light, proving the pellet is the real cue.
Ramer et al. (1977) conceptually replicates the effect. They found food pellets always create polydipsia, but brain stimulation never does. The drinking is tied to food, not to any reward.
Kodera et al. (1976) looks similar but differs. Their brief stimuli controlled lever pressing only when a gap sat between stimulus and food. Here, the same brief flash controls licking only when food follows right away.
Why it matters
Second-order schedules are common in clinics. You may pair praise with tokens, then tokens with candy. This study warns that the final reinforcer, not the token, will drive most adjunctive behavior. If you want to reduce odd post-reinforcement habits, deliver the main reward less often or change the cue that comes just before it.
Want CEUs on This Topic?
The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.
Join Free →Watch for post-candy drinking or pacing; try inserting a longer delay between token and food to weaken the adjunctive response.
02At a glance
03Original abstract
Three rats were exposed to a second-order schedule in which fixed-interval components ended either with food or with a brief stimulus that was never paired with food. Food and the brief stimulus occurred in a random sequence (variable-ratio 2 overall schedule). Another three rats were exposed to a similar second-order schedule, the only difference being that the food or the stimulus was presented independently of operant behavior (fixed-time components). The three rats exposed to the fixed-interval components licked at a water spout after each food presentation. These rats also licked in the intervals after the brief stimulus. Although the discriminative properties of food and of the brief stimulus were identical in relation to subsequent reinforcement, licking after the stimulus was less than after food. The three rats exposed to the second-order schedules with fixed-time components also licked at the water spout after food, but these rats did not lick consistently after brief stimulus presentations.
Journal of the experimental analysis of behavior, 1977 · doi:10.1901/jeab.1977.27-265