Preference between variable-ratio and fixed-ratio schedules: local and extended relations.

Sturmey et al. (1996) let pigeons choose between two keys. One key always asked for the same number of pecks (fixed-ratio). The other key asked for a changing number (variable-ratio).

The team kept the average ratio the same. They only moved the smallest ratio in the VR bag. In one test the smallest was 1. In another test the smallest was 5.

Birds voted with their beaks. More pecks on a key meant "I like this schedule."

When the smallest VR component was 1, birds strongly picked the variable key.

When the smallest component rose to 5, preference flipped. Birds now favored the steady fixed-ratio.

The tiny bottom number, not the average, steered choice.

Soreth et al. (2009) ran the same test with random-interval schedules. They also saw that a higher chance of short waits boosted preference. Together the two papers show: quick wins drive choice on both ratio and interval setups.

Fine et al. (2005) asked pigeons to pick fixed-interval or random-interval. Birds only weakly preferred the fixed side. Their data extend the 1996 rule into the interval world: variability itself can be attractive if the shortest delay stays tasty.

Davison et al. (1995) flipped the question. They kept delays fixed and varied the amount of food. Starlings liked variable delays but disliked variable amounts. The two studies seem to clash until you see the difference: 1996 varied response requirement; 1995 varied pellet count. The animal likes surprise in work, not in payoff.

When you build a VR schedule, set the lowest ratio to 1 if you want the client to stay with that task. A VR 20 that can jump straight to 1 feels better than a VR 20 that always starts at 5. Check your program sheets: the minimum component, not the posted average, controls choice.