Uninstructed human responding: sensitivity to ratio and interval contingencies.
Humans act like the schedule says they should—VR beats VI—when you skip instructions and make the prize feel real.
01Research in Context
What this study did
Researchers asked adults to press a key for points. No instructions. Just the schedule.
They compared two setups. One paid on a variable-ratio (VR) schedule. The other on a matched variable-interval (VI) schedule. To get points, participants had to lift a cup and drink. That extra step made the reinforcer feel real.
What they found
People pressed faster under VR than under VI. The classic pattern showed up even without instructions.
The cup-and-drink step mattered. When reinforcers feel earned, humans follow the schedule logic.
How this fits with other research
Attwood et al. (1988) later bunched all reinforcers at the end of the session. VI rates dropped, VR rates stayed high. Together the two studies show: immediacy drives the VR > VI gap.
Bacon-Prue et al. (1980) moved the idea into a classroom. Kids worked harder only when the schedule blocked access to the fun activity first. Same rule: schedules work when they create real deprivation.
Henton (1972) found shock can either maintain or suppress key pressing, depending on the schedule. The pair warns us: consequences alone don’t tell the story; the schedule pattern does.
Why it matters
Before you tell a client “do your work, then get iPad,” check the schedule. If access is too easy, the contingency may fail. Add a small response cost or wait time so the reinforcer feels earned. Watch response rates rise just like in the lab.
Want CEUs on This Topic?
The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.
Join Free →Add a brief consummatory step before delivery—have the client stand up, open the box, or tear off the token—to sharpen schedule control.
02At a glance
03Original abstract
College students' presses on a telegraph key were occasionally reinforced by light onsets in the presence of which button presses (consummatory responses) produced points later exchangeable for money. One student's key presses were reinforced according to a variable-ratio schedule; key presses of another student in a separate room were reinforced according to a variable-interval schedule yoked to the interreinforcement intervals produced by the first student. Instructions described the operation of the reinforcement button, but did not mention the telegraph key; instead, key pressing was established by shaping. Performances were comparable to those of infrahuman organisms: variable-ratio key-pressing rates were higher than yoked variable-interval rates. With some yoked pairs, schedule effects occurred so rapidly that rate reversals produced by schedule reversals were demonstrable within one session. But sensitivity to these contingencies was not reliably obtained with other pairs for whom an experimenter demonstrated key pressing or for whom the reinforcer included automatic point deliveries instead of points produced by button presses. A second experiment with uninstructed responding demonstrated sensitivity to fixed-interval contingencies. These findings clarify prior failures to demonstrate human sensitivity to schedule contingencies: human responding is maximally sensitive to these contingencies when instructions are minimized and the reinforcer requires a consummatory response.
Journal of the experimental analysis of behavior, 1977 · doi:10.1901/jeab.1977.27-453