Behavioral contrast as differential time allocation.
Behavioral contrast can show up in where subjects spend their time, not just how fast they respond — track time allocation when analyzing schedule effects.
01Research in Context
What this study did
White (1978) ran rats on a two-part schedule. One part gave food often. The other part gave food rarely.
The team tracked how long the rats stayed in each part. They wanted to see if contrast shows up in time, not just speed.
What they found
When the rich side became richer, rats spent even more time there. Their stay-time shifted before their lever-press speed did.
Positive contrast appeared in the clock, not just the counter.
How this fits with other research
Rilling et al. (1969) first said "choice is time matching." White (1978) adds "contrast is time matching too."
Burgio et al. (1986) later showed contrast can also come from two responses fighting for the same moment. The 1978 view and the 1986 view look opposite, but they measure different things: where time goes versus which response wins.
Herrnstein et al. (1979) soon proved time allocation is the steadier ruler. Their data back the 1978 decision to watch the clock, not just the click.
Why it matters
When a learner's rate jumps after you thicken reinforcement, ask: did they speed up, or just stay longer? Track session time per area with a stopwatch or app. If the shift is in time, adjust reinforcement richness or add alternate tasks before you chase response topography.
Want CEUs on This Topic?
The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.
Join Free →Time five minutes of your next multielement session; record how long the learner stays in each area to see if contrast lives in location, not just speed.
02At a glance
03Original abstract
In Experiment I, hooded rats were exposed to multiple variable-interval schedules of reinforcement in which manipulanda and reinforcement magazines at opposite ends of the experimental chamber were associated with the different components. Time allocated to each component was measured by recording the time spent by the subject in the appropriate half of the chamber. Positive behavioral contrast was observed for the comparison between multiple variable-interval 30-second variable-interval 30-second and multiple variable-interval 30-second variable-interval 90-second conditions for both response frequency and time allocation measures, but not for mean local response rate (response frequency per time allocated to a component). In Experiment II, rats were exposed to multiple variable-time schedules in which reinforcement was response independent. Time allocated to each component was measured for two conditions, multiple variable-time 30-second variable-time 30-second and multiple variable-time 30-second variable-time 90-second. Positive behavioral contrast of time allocation was exhibited. The results indicated that time allocation was differentially sensitive to changes in reinforcement probability, and that behavioral contrast may result from the differential allocation of time to the different components of the multiple schedule.
Journal of the experimental analysis of behavior, 1978 · doi:10.1901/jeab.1978.29-151