Human choice in concurrent ratio-interval schedules of reinforcement.

The team gave adults two buttons to press. One button paid off on a ratio schedule—work a set number of times, get a coin. The other paid on an interval schedule—wait a set time, get a coin.

They added a ticking clock light so people could see which schedule was ready. Then they counted how many presses went to each button.

Most people leaned toward the ratio side, but not all the way. Their split sat between strict matching and perfect maximizing.

Only one person shifted close to maximizing after the clock was added. The rest kept the same middle-of-the-road bias.

Wallander et al. (1983) ran the same VR-VI setup with animals and saw clear maximizing once extra downtime was trimmed. Humans in the new study did not copy that pattern—species matters.

Fovel et al. (1989) also used VR-VI schedules with mice and found longer reinforcers pushed the mice toward the interval key. The 1994 humans still preferred the ratio key, so the same schedule can push choice in opposite directions depending on who is pressing.

Older human work reviewed in Pierce et al. (1983) shows people usually match response rates to payoff rates. The current data partly agree—people moved toward the richer side—but they stopped short of full matching, showing a boundary condition.

When you build token boards or choice menus, do not assume clients will always pick the richest option or perfectly match past payoff rates. Expect a middle bias and be ready to add cues like clocks or lights; they can tip one client toward better maximizing, but most will still blend strategies.