Preference for unreliable reinforcement in children with mental retardation: the role of conditioned reinforcement.

The team ran three lab games with children who have intellectual disability.

Each child picked between two colored boxes on a screen. One box gave a toy every time. The other gave a toy only half the time but flashed a fun picture before each payoff.

The researchers watched which box the child kept choosing across many turns.

Most kids stayed with the 50% box when the flash came with it. The blinking picture acted like extra candy, even though the sure box paid off more often.

Conditioned reinforcement beat the math of the schedule.

Lalli et al. (1995) first saw this in pigeons. The 2000 study shows the same rule holds for kids with ID, moving the idea from birds to humans.

Shahan et al. (2021) extends the point to the clinic: when we thin reinforcement too fast, the old problem behavior flashes back, just like the kids returned to the risky box.

Ward-Horner et al. (2017) review says some learners actually like leaner schedules if bigger or signaled payoffs come with them. The new data match that review and give one reason why—conditioned reinforcers sweeten the deal.

Your client may stick with an unsafe response that only works sometimes if lights, sounds, or adult attention come with it. Before thinning a schedule, check what stimuli are tagging along. Add those same sights or sounds to the new, safer response so the child follows the cue, not the odds.