Interaction of procedural factors in human performance on yoked schedules.

The team tested how small setup choices change the way people respond on VR and VI schedules.

Adults without disabilities worked on yoked pairs: one schedule paid after a number of responses, the other after time passed.

The researchers tried three tweaks: cut the instructions, shaped the tap response, and let participants drop a coin in a bank after each payoff.

The clearest schedule difference showed up only when all three tweaks were in place.

With full instructions and no coin drop, VR and VI lines looked almost the same.

VR schedules also held their shape better when later instructions tried to trick the learner.

Clark et al. (1977) first saw the VR jump over VI in pigeons; Hagopian et al. (2000) now shows the same jump in people, but only if you run the session like an animal lab.

Cullinan et al. (2001) found VR beats VI for rate yet loses faster when disruption hits. The 2000 paper adds that you may miss the basic rate gap entirely if you talk too much at the start.

Schroeder et al. (1969) and McKearney (1970) punished VR and VI responding and saw VR collapse first. Hagopian et al. (2000) shows VR also resists rule-based sabotage better, hinting that ratio schedules tie more tightly to the contingency, not to the words about it.

If you want to see true schedule control in humans, talk less, shape the response, and add a concrete payoff action. This keeps the contingency, not the instructions, in charge. Next time you run a lab demo or train staff, strip the briefing, build the target move step-by-step, and let the learner do something physical with the reinforcer. You will get cleaner data and clearer teaching examples.