Some parameters of behavioral contrast and allocation of interim behavior in rats.

The team ran rats in two-key boxes. One key paid on a VI 10-s schedule. The other key never paid.

They swapped how long each side stayed available. Sessions had 30, 60, or 120-second components.

The goal was to see if time or pay rate drives contrast — the jump in responding when the rich side returns.

Contrast only showed up when the baseline was VI 10 s versus extinction. Longer or shorter components did not matter.

The results partly followed the matching law but clashed with theories that say time itself creates contrast.

Sturmey (1995) later saw the same pattern in pigeons and added a new twist: birds also shifted extra-key pecks toward the lean side.

Reiss et al. (1982) conceptually matched the rat data in pigeons. They proved that duration effects vanish when you unlink them from pay rate.

Rincover et al. (1975) looks like a contradiction. Free extra food lowered rats’ baseline rates yet raised pigeons’ rates. The studies differ by species and procedure, so both can be true.

If you run multiple-schedule interventions, set the rich side to about VI 10 s to see clear contrast. Do not count on simply stretching component length to boost effect size. The money is in the pay rate, not the clock.