Absolute and relative measures of dimensional contrast.
Contrast effects come from how far apart your stimuli are, not how strong each one is alone.
01Research in Context
What this study did
Hinson (1988) ran two experiments with pigeons on multiple schedules. The birds pecked a key for food while the color changed along a single dimension.
The team asked: does contrast come from an absolute jump in pecks, or from sharper discrimination between colors?
What they found
Contrast shoulders showed up on the generalization curve, but the height could not be guessed from baseline rates. The birds were not just pecking more; they were telling colors apart better.
In plain words, the effect is relative, not absolute.
How this fits with other research
Malouff et al. (1985) had already shown that sequential details—like what happened on the last trial—do not change the contrast curve. Hinson (1988) keeps that rule and adds why it holds: the curve is driven by the relation between stimuli, not by local ups and downs.
Varley et al. (1980) found a similar story in brightness. Pigeons cared about the contrast between line and background, not the raw light level. Both papers point to the same lesson—control relative differences, not absolute values.
Schwartz (1975) showed that moving the key light can wipe out positive contrast. Together with Hinson (1988), this suggests two levers: keep stimuli on the key and make sure the color steps are clearly spaced.
Why it matters
When you build multiple-schedule programs, think relative, not absolute. Widen the color gap between components instead of just cranking up reinforcement rate. Check that your SD stays on the same key location. These small layout choices give you cleaner contrast and faster discrimination.
Want CEUs on This Topic?
The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.
Join Free →Space your color cues one extra step apart on the gradient and watch the contrast curve sharpen.
02At a glance
03Original abstract
Two sets of experiments examined the discriminative performance of pigeons on a visual flicker-rate continuum using a maintained generalization procedure. In the first experiment, responses during the intermediate stimulus value were not reinforced, whereas responses during all other stimuli were reinforced periodically. In the second experiment, training was similar to the first, with the exception of one condition in which stimuli adjacent to the negative stimulus border were eliminated from the discrimination set. Results from both experiments show that positive dimensional contrast seems to represent a relative enhancement of discrimination gradient form, rather than an absolute increase in responding with respect to prior baseline. Further, the form and magnitude of positive dimensional contrast are not predictable from the form and magnitude of baseline response rates. Results from the second experiment indicate that eliminating border stimuli increased response-rate differences between positive and negative stimuli, but did not necessarily diminish the magnitude of positive dimensional contrast.
Journal of the experimental analysis of behavior, 1988 · doi:10.1901/jeab.1988.50-249