Relationship between response rate and reinforcement frequency in variable-interval schedules: III. The effect of d-amphetamine.

The team gave lab rats tiny shots of d-amphetamine before their normal VI lever-press sessions. They used two doses: 1.6 and 3.2 micromoles per kilogram.

Each rat worked on five different VI schedules in one session. The schedules paid food pellets from 8 to 128 pellets per hour. The researchers tracked every press and pellet to map the full response-rate curve.

Both drug doses lowered lever pressing. Higher dose, fewer presses. The drop followed a clear dose-response line.

The drug also shifted the Herrnstein curve. Maximum possible response rate (k) fell and the half-max pellet rate (Rmax) moved left. In plain words, rats worked less and needed fewer pellets to stop trying.

Wilkie et al. (1981) part II tested smaller sucrose volumes on the same VI setup. Smaller drinks moved Rmax to the right, while k stayed put. Amphetamine now moves both k and Rmax, showing the drug hits the motor ceiling plus reward sensitivity.

Rose et al. (2000) later showed that thicker sucrose also changes k. Together the three papers prove k is not fixed; it slides with reinforcer size and with drugs.

Anger et al. (1976) saw haloperidol cut schedule-induced drinking. Both drugs drop behavior, but amphetamine does it through motor acceleration that crashes into the physics of the lever, not through sedation.

If you run VI preference assessments or teaching programs, remember that stimulant medication can flatten the response curve. You may see slower rates even though reinforcement stayed the same. Watch for left-shifted Rmax: the learner might reach satiation sooner. Consider shorter sessions or richer schedules on days when medication peaks.