Operative mechanisms of noncontingent reinforcement at varying magnitudes and schedules.

Cullinan et al. (2001) asked a simple question. If you give free reinforcers, does size or timing matter more? They ran adults in a lab. The team varied how big each free snack was and how often it arrived.

All reinforcers were noncontingent. No response was needed. The researchers watched how much time people spent eating versus working.

Big reinforcers on dense schedules cut work the most. The reason was simple: people spent more minutes eating. Less time left to press buttons.

The drop in work was large and predictable. Higher density meant more mouth time and fewer responses.

Heavey et al. (2000) looked at the same issue one year earlier. They also used dense NCR. But they saw a problem: free snacks blocked children from learning to ask for things. The fix was to thin the free snacks first, then teach mands.

The two studies seem to clash. A et al. say dense NCR is great for fast response reduction. L et al. say dense NCR can stop new skills. The gap is the goal. If you only want to stop a behavior fast, go dense. If you also need to teach, thin first.

Wilkinson et al. (1998) add another layer. They showed people pick the side that gives more total reinforcers per minute. A et al. prove the same rule works when the reinforcers are free.

Use dense, large NCR when safety calls for a quick stop. Then thin the schedule as soon as you can. Pair the thinning with DRA so the client gains a new skill while the old one fades. Track mouth time; it tells you why the behavior dropped.