Schedule-induced polydipsia in rats living in an operant environment.
Longer fixed-interval schedules reliably trigger intense, pellet-linked drinking in rats, a warning that timing alone can spawn powerful adjunctive behavior.
01Research in Context
What this study did
Nelson et al. (1978) watched rats that lived inside an operant box.
Food pellets arrived on fixed-interval schedules of 15 s or 60 s.
The team recorded when the rats drank water and how much.
What they found
Stretching the interval from 15 s to 60 s doubled or tripled total water intake.
Drinking shifted from a meal-time pattern to short bursts right after each pellet.
The longer schedule created classic schedule-induced polydipsia.
How this fits with other research
Falk (1966) showed the same linear climb in drinking up to 300 s, so the 1978 data tighten the curve inside that range.
Blackman (1970) proved the key variable is time between pellets, not accidental reinforcement; R et al. echo that with pure FI schedules.
O'Leary et al. (1979) replaced rats and water with humans and walking, yet the fixed-interval still produced extra behavior right after reinforcement, confirming the effect crosses species and responses.
Why it matters
The study reminds you that schedule timing can create collateral behaviors you did not plan.
If you space reinforcers farther apart, watch for sudden bursts of drinking, pacing, or other adjunctive actions that might interfere with your target skill.
Adjusting interval length or offering competing stimuli can keep these side behaviors low and keep learning on track.
Want CEUs on This Topic?
The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.
Join Free →Check for sudden post-reinforcer behaviors when you lengthen FI timing; shorten the interval or add a brief toy task if side behaviors surge.
02At a glance
03Original abstract
The effect of variations in interreinforcement interval on the temporal and distributional relation between feeding and drinking was continuously monitored. Rats were housed continuously in an operant chamber in which water was freely available, but lever pressing was required to obtain food (45-mg pellets). Initially, pellets were delivered on a fixed-ratio 1 schedule of reinforcement, which was followed by testing on response-initiated fixed-interval 15-, 30-, and 60-second schedules. The total number of discrete, daily meals (a period in which several pellets were earned in succession) was slightly higher during the fixed-interval schedules than during the fixed-ratio 1, but there was no systematic effect of fixed-interval length on meal frequency. Total water consumption, in contrast, increased dramatically as the interval was lengthened: both subjects consumed two to three times as much water on the fixed-interval 60-second schedule as on the fixed-ratio 1. The increased water consumption was the result of an alteration in the distribution of drinking relative to eating. During the fixed-ratio 1 condition, drinking occurred infrequently following individual food pellets and represented the smallest percentage of total drinking; drinking occurred predominantly just before or after a meal. As the fixed interval was lengthened, however, the frequency of postpellet drinking gradually increased and eventually comprised the largest proportion of daily drinking.
Journal of the experimental analysis of behavior, 1978 · doi:10.1901/jeab.1978.29-493