Variable-time reinforcement schedules in the treatment of socially maintained problem behavior.

Two participants with socially maintained problem behavior got two kinds of noncontingent reinforcement. One schedule delivered the reinforcer every few minutes on the dot. The other delivered it at random times. The team flipped the schedules back-to-back to see which cut problem behavior more.

No extra demands or rewards were tied to the behavior. The reinforcers just showed up on a timer.

Both the fixed-time and the variable-time schedules knocked problem behavior down by a lot. The random schedule worked just as well as the clockwork one.

The results say you can pick either timing style and still get the same drop in behavior.

Perez et al. (2015) pooled every NCR paper through 2014 and found a very large average effect. The 2000 VT study is inside that pool, so the new data add one more vote that NCR is a heavy hitter.

Dall et al. (1997) ran a yoked NCR arm inside an FCT study. They also saw NCR alone cut self-injury, but FCT gave the extra win of building a new communicative response. The 2000 paper keeps the spotlight on NCR's power while staying quiet about replacement skills.

Verriden et al. (2019) tried NCR plus DRA for automatically reinforced behavior and saw no change until they layered on a punisher. That looks like a clash, but the difference is the function: the 2000 study targeted social reinforcement; the 2019 study targeted sensory reinforcement. NCR still rules for social functions, but sensory cases may need more tools.

If you have a client whose problem behavior is fed by adult attention or escape, you can run NCR on a variable timer and get the same calm as a fixed timer. That frees you from watching the clock and makes thin-out easier. Just pick the schedule that fits the setting best, keep the functional reinforcer, and watch the behavior fall.