Stereotypic behavior of mentally retarded adults adjunctive to a positive reinforcement schedule.

The team worked with adults who have profound intellectual disability. They paid the adults with tokens for simple adaptive tasks.

The catch: the tokens came on a fixed-interval schedule. The wait started at 15 seconds and grew to 180 seconds across phases.

As the wait for reinforcement got longer, stereotypic behavior jumped. Hand-flapping and body-rocking filled the extra time.

The behavior looked like "adjunctive" responding — extra actions that pop out while the next reinforcer is delayed.

Clarke et al. (1998) saw the opposite in a classroom. They used fixed-ratio token boards and stereotypy dropped. The key difference: ratio schedules keep the learner busy responding; interval schedules leave dead time.

Animal work foretold this. Clark et al. (1977) showed hamsters pace and drink more when food arrives on fixed-time, and Hamm et al. (1978) saw rats develop rigid response patterns under fixed-interval. The 1988 study extends the pattern to humans with ID.

Llinas et al. (2022) later confirmed schedule density matters: continuous matched stimuli cut stereotypy faster than lean fixed-time schedules. All three lines agree — tighter, response-linked schedules or richer non-contingent delivery reduce stereotypy.

If you run lean interval schedules for adaptive skills, watch for a boom in stereotypy. Fill the gap with embedded tasks, ratio requirements, or matched stimuli. Keep the schedule dense or response-based until the target behavior is strong.