Some effects of relative reinforcement rate and changeover delay in response-independent concurrent schedules of reinforcement.

Pigeons pecked two keys. Food arrived on each key at different rates, no matter how they pecked. The researchers added a changeover delay (COD) that made the birds wait before switching keys. They asked: do birds still match their time to the food rate when food is free? And does a longer wait cut switching?

Time on each key still matched the food rate, even when pecking did not matter. Longer waits cut switching. Bigger food-rate gaps also cut switching. The matching law holds even when responses do not earn food.

Blue et al. (1971) built on this by raising the wait in tiny 0.5-s steps. Tiny steps kept real food rates closer to the plan. Fahmie et al. (2013) showed the wait itself becomes a cue: birds later pick the key that once had the longer wait. PLISKOFF (1963) first showed strict matching with response-dependent food; Neuringer et al. (1968) proved it still works when food is response-independent. Quilitch et al. (1973) held food rate steady and changed delay instead; together the two papers show choice follows both rate and delay rules.

When you run concurrent schedules, remember that clients will match time even if responses do not produce reinforcers. Add a brief changeover delay to cut rapid switching. Raise the delay slowly, 0.5 s at a time, to keep programmed and obtained rates in line. This keeps your data clean and your learners calm.