ABA Fundamentals

Factors affecting choice of signaled or unsignaled food schedules.

Harsh et al. (1983) · Journal of the experimental analysis of behavior 1983
★ The Verdict

Signal length and commitment rules decide if warnings help or hurt—preference isn’t fixed.

✓ Read this if BCBAs shaping token boards, timers, or advance notice for kids who escape tasks.
✗ Skip if Clinicians only running simple FR schedules with no timing cues.

01Research in Context

01

What this study did

Funderburk et al. (1983) worked with pigeons in a small lab chamber.

The birds pecked two keys. One key gave food right away with no warning light. The other key gave food after a light blink.

The team changed two things: how long the light stayed on and whether the bird could switch keys after picking.

02

What they found

Short blinks made birds hate the signaled key. Long blinks made them love it.

When the birds had to stay with their first pick, the short-blink birds now wanted the signal. The long-blink birds now skipped it.

The results flip-flopped. Signal length and commitment rules controlled choice, not some fixed taste for warnings.

03

How this fits with other research

Meyer et al. (1987) later saw the same flip in people and pigeons. Early commitment made low-probability rewards look better. The 1983 study shows the same rule works for signaled food.

Catania et al. (1982) found pigeons also act "superstitious" when colored lights come and go. Together these papers prove that tiny stimulus changes can drive big, stable preferences.

Norris et al. (2024) looked at kids who hit themselves for different reinforcers. They also saw strong, personal rankings. The bird data remind us that preference is always conditional, even in clinical work.

04

Why it matters

Your client may hate warnings today and love them tomorrow. Check signal length and whether the learner can "change their mind." A short countdown plus easy escape can kill motivation. A long preview plus locked-in choice can boost it. Test, don’t assume.

Free CEUs

Want CEUs on This Topic?

The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.

Join Free →
→ Action — try this Monday

Try a 10-second sand timer before work; if the client bolts, let them leave once, then require them to stay next cycle—note which rule set boosts work.

02At a glance

Intervention
not applicable
Design
single case other
Population
not specified
Finding
mixed

03Original abstract

Choice between signaled and unsignaled food schedules was assessed in two experiments using a shuttlebox. Experiment 1 examined preference with both a five-second signal and a thirty-second signal. Subjects preferred the unsignaled side with a five-second signal but preferred the signaled side with a thirty-second signal. Experiment 2 assessed preference using either a commitment procedure (a shuttle response resulted in exposure to the schedule for a fixed time) or a noncommitment procedure (each shuttle response changed the schedule in effect). Subjects preferred the signaled condition with the commitment procedure but the unsignaled condition with the noncommitment procedure. These results indicate that the discrepant findings of earlier studies are due to procedural differences involving signal duration and choice commitment. The data are consistent with a conditioned reinforcement interpretation of choice.

Journal of the experimental analysis of behavior, 1983 · doi:10.1901/jeab.1983.40-265