Fixed versus variable sequences of food and stimulus presentation in second-order schedules.

Downing et al. (1976) worked with pigeons on second-order schedules. The birds pecked for food under fixed-interval parts. Each part ended with either food or a brief light.

The team compared two ways to order the food and the light. One group got a fixed A-B-A-B pattern. The other group got the same events in a random shuffle.

Random sequencing made the brief light act more like real food. Birds responded in a smoother fixed-interval curve when the order was mixed up.

Fixed alternation weakened stimulus control. The light lost power when it always came second in a set pattern.

Pomerleau et al. (1973) showed that any food-paired brief stimulus can speed up early responding. K et al. added the rule: shuffle the pairings to keep that power strong.

Matson et al. (2013) tested the same fixed-versus-random question with children in a multielement functional analysis. They got the opposite result: fixed order gave clearer data. The difference is the goal. K wanted strong stimulus control; L wanted fast differentiation. Same tool, different job.

Gardner et al. (1977) kept the random idea and swapped food for cocaine. The fixed-interval pattern still held, proving the rule works across reinforcer types.

If you use tokens, praise, or lights as conditioned reinforcers, mix when they occur. A fixed praise-every-third-trial routine can water the stimulus down. Shuffle the timing to keep your signals strong and your learner’s responding steady.