ABA Fundamentals

Partial avoidance contingencies: Absolute omission and punishment probabilities.

Flye et al. (1979) · Journal of the experimental analysis of behavior 1979
★ The Verdict

Omission probability steers avoidance behavior in steady steps; punishment probability steers it in unpredictable jumps that differ for each learner.

✓ Read this if BCBAs who mix reinforcement with response-cost or mild punishment in skill-acquisition or reduction programs.
✗ Skip if Practitioners who rely solely on reinforcement without punishment or cost components.

01Research in Context

01

What this study did

Foster et al. (1979) worked with rats in a free-operant avoidance box. The rats could press a lever to delay shocks. The team changed two things across conditions: how often a press cancelled shock (omission probability) and how often a press brought immediate shock (punishment probability).

Each rat lived in the box for one long session. The researchers watched how lever pressing changed as the odds of omission or punishment rose or fell.

02

What they found

When omission probability went up, every rat pressed more in a smooth, step-by-step way. When punishment probability went up, the rats reacted very differently: some stopped pressing, others barely changed, and a few fell in between.

The takeaway: omission gives steady, graded control; punishment gives messy, individual control.

03

How this fits with other research

Schmidt et al. (1969) saw near-total avoidance shutdown when every response brought certain punishment. L et al. now show that dropping punishment to partial odds splits the group: some rats still stop, others keep going. The earlier paper set the extreme; the new paper maps the middle ground.

Vukelich et al. (1971) found that punishment suppression grows if the rat first tastes a punishment-free period. L et al. saw similar uneven effects, hinting that each rat’s past with shock helps decide how later odds hit.

Deluty (1976) showed that raising punishment on one lever can push the animal toward a second punished lever. L et al. add that even within one lever, higher punishment odds do not push every animal the same way—individual history again matters.

04

Why it matters

If you use response-cost or mild punishment in a treatment plan, do not assume a single dosage will work for every client. Omission-based rewards (extra minutes of iPad, tokens that cancel a chore) give reliable, graded shifts you can predict. Punishment-based reductions (losing tokens, brief restraint) can hit clients unevenly—watch each learner separately, start small, and measure individual response instead of waiting for a group norm.

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

Track each client’s response to punishment-based procedures separately; if one learner shows no suppression, lower the odds or switch to omission-based reward before raising intensity.

02At a glance

Intervention
other
Design
single case other
Population
neurotypical
Finding
mixed

03Original abstract

Avoidance contingencies were defined by the absolute probability of the conjunction of responding or not responding with shock or no shock. The "omission" probability (rho(00)) is the probability of no response and no shock. The "punishment" probability (rho(11)) is the probability of both a response and a shock. The traditional avoidance contingency never omits shock on nonresponse trials (rho(00)=0) and never presents shock on response trials (rho(11)=0). Rats were trained on a discrete-trial paradigm with no intertrial interval. The first lever response changed an auditory stimulus for the remainder of the trial. Shocks were delivered only at the end of each trial cycle. After initial training under the traditional avoidance contingency, one group of rats experienced changes in omission probability (rho(00)>0), holding punishment probability at zero. The second group of rats were studied under different punishment probability values (rho(11)>0), holding omission probability at zero. Data from subjects in the omission group looked similar, showing graded decrements in responding with increasing probability of omission. These subjects approximately "matched" their nonresponse frequencies to the programmed probability of shock omission on nonresponse trials, producing a very low and approximately constant conditional probability of shock given no response. Subjects in the punishment group showed different sensitivity to increasing absolute punishment probability. Some subjects decreased responding to low values as punishment probability increased, while others continued to respond at substantial levels even when shock was inevitable on all trials (noncontingent shock schedule). These results confirm an asymmetry between two dimensions of partial avoidance contingencies. When the consequences of not responding included occasional omission of shock, all subjects showed graded sensitivity to changes in omission frequency. When the consequences of responding included occasional shock delivery, some subjects showed graded sensitivity to punishment frequency while others showed control by overall shock frequency as well.

Journal of the experimental analysis of behavior, 1979 · doi:10.1901/jeab.1979.31-351