Behavioral and dimensional contrast in rats.
Stopping rewards for some cues boosts responding to the remaining cues, with an extra jump right next to the cutoff.
01Research in Context
What this study did
The researchers worked with rats in an operant chamber. They lined up tones from low to high pitch. One half of the line earned food for lever presses. The other half stopped giving food.
The team watched what happened when the rats hit the border between pay and no-pay tones.
What they found
When food stopped for half the tones, lever pressing jumped for the tones that still paid. This is classic behavioral contrast.
Extra jumps showed up right next to the cutoff. Rats worked hardest for tones that sat beside the new no-food zone. The paper calls this extra lift dimensional contrast.
How this fits with other research
Ginsburg et al. (1971) already showed that longer extinction spells make later contrast bigger. Blough (1980) adds that the edge itself, not just time, can give an added push.
Richards (1974) saw the same border bump in pigeons, but the lift faded with more training. Blough (1980) keeps the lift steady, showing the effect can last under the right setup.
Ring et al. (2023) tried the idea with adults in a fake office. Only one person showed clear contrast. The rat data warn us that contrast may vanish when we move from lab to life.
Why it matters
If you run multiple-schedule programs, watch the line between reinforced and extinguished tasks. Responses can spike at that edge, giving false hope of progress or masking true mastery. Track each stimulus or task separately and be ready to thin reinforcement near borders first.
Want CEUs on This Topic?
The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.
Join Free →Graph each stimulus or task in your multielement plan and watch for surprise spikes at the border where reinforcement ends.
02At a glance
03Original abstract
Rats pressed a nose key for brain stimulation reinforcement presented on a fixed-interval schedule. Stimuli were drawn at random from a continuum of 12 white noise intensities in the range 62-95 decibels, spaced in 3 decibel steps. Experiment 1 varied the number of stimuli and the reinforcement contingencies associated with them. In Condition I (baseline) all stimuli signaled reinforcement; in Conditions II and III stimuli from one half of the continuum signaled reinforcement and those from the other half, extinction. However, in Condition II the 6 stimuli from the middle of the continuum were omitted. Experiment 2 held constant the number of stimuli and varied their spacing. In Condition I, each of 6 sounds signaled reinforcement. In Conditions II and II, three stimuli from one half of the continuum signaled reinforcement and three from the other half, extinction. However, in Condition II the stimuli were near the extremes of the continuum (Stimuli 1, 3, 4, 9, 10, 12). Condition III replaced Stimulus 3 with Stimulus 6 and Stimulus 10 with Stimulus 7. Behavioral contrast was seen in an increase over baseline in response rate to the stimuli associated with the constant schedule component when the variable component was changed to extinction. Dimensional contrast was seen in a further elevation of rate to intermediate positive stimulus values when stimuli were added to the border region between positive and negative values.
Journal of the experimental analysis of behavior, 1980 · doi:10.1901/jeab.1980.33-345