Schedule-induced drinking as functions of interpellet interval and draught size in the Java macaque.
Schedule-induced drinking peaks at middle-length intervals because monkeys start fewer bouts at very long waits, not because they lick slower.
01Research in Context
What this study did
The team studied Java macaques in a lab cage. Food pellets dropped on fixed schedules. Water was always there. The only thing that changed was the time between pellets and the size of each water sip.
They watched every lick. They asked: does longer wait change how much the monkeys drink? Does sip size matter?
What they found
Drinking rose, then fell as the gap between pellets grew. Best drinking happened in the middle, not at the shortest or longest waits.
Long waits made each drinking bout bigger, but the monkeys started fewer bouts. Sip size had only small effects.
How this fits with other research
Anger et al. (1976) saw the same hump-shaped curve in rhesus monkeys. The new paper adds that bout size, not lick speed, drives the change.
Reid et al. (1987) later found the same hump in rats when they changed meal size instead of time. Again, bout initiation explained the dip.
Castilla et al. (2013) mixed food deprivation with short intervals. They saw more licking with both hunger and fast pellets. The 1978 paper kept hunger steady, so it shows the pure time effect.
Why it matters
If you run fixed-time or FI schedules, expect more adjunctive drinking at moderate intervals. Stretch the interval too far and the behavior drops even though each bout is longer. Watch bout starts, not lick speed, to spot the change. This keeps your data clean when studying collateral behavior during reinforcement schedules.
Want CEUs on This Topic?
The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.
Join Free →Count bout starts, not licks, when you track collateral drinking in your next FI session.
02At a glance
03Original abstract
Three Java monkeys received food pellets that were assigned by both ascending and descending series of fixed-time schedules whose values varied between 8 and 256 seconds. The draught size dispensed by a concurrently available water-delivery tube was systematically varied between 1.0 and 0.3 milliliter per lick at various fixed-time values during the second and third series determinations. Session water intake was bitonically related to the interpellet interval and was determined by the interaction of (1) the probability of initiating a drinking bout, which fell off at the highest interpellet intervals and, (2) the size of the bout, which increased directly with increases in interpellet interval. Variations in draught size had little effect on total session intakes, but reduced bout size at draught sizes of 0.5 milliliter and below. Thus, a volume-regulation process of schedule-induced drinking operated generally at the session-intake level, but was limited to higher draught sizes at the bout level.
Journal of the experimental analysis of behavior, 1978 · doi:10.1901/jeab.1978.30-139