The effect of multiple S periods on responding on a fixed-interval schedule: IV. Effect of continuous S with only short S probes.

Pigeons pecked a key on a fixed-interval 500-second schedule. Most of the time the key stayed dark, meaning pecks had no effect. Every 30 or 60 seconds the key lit green for 3 seconds; during these short probes food could set up.

The dark periods acted like mini-extinction sessions. Researchers wanted to know if the birds still showed the classic FI "scallop" when almost the whole interval blocked reinforcement.

The birds still paused right after food, then gradually sped up. Response rates were lowest early and highest just before the next food chance. The familiar curved pattern stayed intact even though a large share of the interval was extinction time.

Short green probes were enough to keep the scallop alive. The birds didn’t need steady access to reinforcement to show schedule control.

Pierce et al. (1983) later zoomed in and found the scallop is really three flat steps—pause, interim, terminal—not smooth acceleration. Their micro-view adds detail to the same curve Dews (1966) first defended against "mediating behavior" ideas.

McKearney (1970) swapped food for electric shock and still got the scallop. Together the three studies show the pattern is stubborn: it survives brief extinction, it survives shock, and it’s made of discrete stages.

Landa et al. (2016) used a similar on-off design in PECS training. They kept the picture card available only during reinforcement periods, paralleling the brief green probes here. Both papers show stimulus control can be tightened by limiting access to the "go" signal.

You can thin reinforcement heavily and still keep schedule-shaped behavior. In practice, give clients short, clear windows to earn reinforcement while the rest of the interval is extinction or DRO time. The scallop will hold, so use brief S-delta probes to stretch inter-reinforcement times without losing the response pattern.