Studies on responding under fixed-interval schedules of reinforcement: the effects on the pattern of responding of changes in requirements at reinforcement.

The team used a fixed-interval schedule with lab animals.

They changed the local rule for earning the reinforcer right before delivery.

The goal was to see if the classic scallop pattern would break.

The scallop stayed intact even when the last-second rule shifted.

Small wiggles appeared, but the overall curve looked the same.

Long-term history with the FI schedule mattered more than the final response.

Feinstein et al. (1988) later inserted extra reinforcers into FI runs.

They showed each reinforcer blocks prior responses from future effect, sharpening the idea that FI curves reflect extended history, not single events.

Rapport et al. (1982) found that simple response count carries little control across intervals, seeming to clash with the 1969 claim that cumulative history drives the scallop.

The gap is methodological: D et al. tried to strengthen count control with added cues and still saw null results, while Schwartz (1969) altered only the terminal contingency and kept the curve.

Rosenberg (1986) extends the story by proving that higher response totals within an interval raise reinforcement probability, giving a mechanism for how cumulative responding can sculpt the scallop without contradicting D et al.'s null finding.

When you run FI schedules in practice, stop tinkering with last-second rules to fix pausing or burst shapes.

Focus on the learner's long reinforcement history instead.

If the curve is flat, stretch the interval slowly and let the history build; quick patch contingencies at the end won't redraw the pattern.