Renewal

Renewal is an increase in previously reduced responding when testing occurs outside the context in which that response was reduced.

It has been studied primarily after extinction, a procedure in which a learned response declines when the relation that once supported it is no longer maintained. Related renewal effects have also been demonstrated after instrumental and operant responses have been reduced.

Renewal is a form of behavioural return. It is not personal reinvention, restored motivation or beginning again.

Why Context Matters

Learning occurs within a surrounding configuration of conditions. In experiments, these conditions may include room lighting, background displays, virtual environments or task settings.

Researchers label these arranged contexts A, B and C. The labels make the sequence easier to follow, but they should not be mistaken for absolute psychological boundaries. People may notice some contextual features more than others or generalise learning between settings.

Suppose a cue repeatedly predicts an outcome and begins to produce a learned response. Later, the cue is presented without that outcome, and responding declines. Extinction can establish new learning that competes with what was acquired earlier: under these conditions, the cue no longer predicts the previous outcome.

The expression of that newer learning can be sensitive to context. When testing occurs elsewhere, it may exert less influence, allowing the earlier response to increase again.

Renewal is this observed increase in responding. Competing learning and context-dependent retrieval are explanations of why the effect may occur; they are not renewal itself.

The effect does not prove that the original learning remained entirely unchanged. Nor does it show that extinction only suppressed it. It demonstrates that the reduction in responding did not transfer equally across all tested conditions.

The Experimental Pattern

A response appearing in a different environment is not enough to establish renewal.

The target response must first be acquired and then demonstrably reduced. Reduction means that responding decreases relative to acquisition or an appropriate baseline; it does not have to reach zero.

Testing must then occur outside the context in which the response was reduced. Responding must increase relative to the end of reduction and an appropriate comparison.

Imagine that participants learn, in one visual environment, that a symbol predicts an unpleasant noise. The symbol begins to produce expectation or avoidance.

In another visual environment, the symbol repeatedly appears without the noise. Responding declines.

The symbol is then tested outside that second environment. If responding increases, the result may support renewal. Researchers commonly strengthen that conclusion by comparing responding outside the extinction context with responding inside it or with a same-context control.

These comparisons matter because a context change can itself affect attention, expectation or arousal. Without adequate controls, increased responding might reflect contextual novelty, incomplete extinction, ordinary variability or another effect rather than renewal.

Renewal may also be partial or brief. It can be strongest in the first test trials and then decline. Behaviour, expectancy, physiological responding and self-report may not all change together because they capture different aspects of the response.

Renewal therefore does not necessarily restore the complete original pattern.

ABA, ABC and AAB Renewal

The three-letter notation records the contexts used during acquisition, extinction and testing.

ABA renewal occurs when a response is acquired in context A, reduced in context B and tested back in context A.

This is the most intuitive arrangement because responding returns in the environment where it was initially learned.

But returning to the acquisition context is not necessary.

ABC renewal occurs when acquisition takes place in A, extinction in B and testing in a third context, C.

AAB renewal occurs when acquisition and extinction both take place in A, followed by testing in B.

All three arrangements involve testing outside the extinction context. Findings from ABC and AAB arrangements show that increased responding can occur without returning to the acquisition context, although the effect depends on the procedure and comparison conditions.

The arrangements are recognised experimentally, but they are not equally strong or equally well documented across human populations, tasks and measures. Human evidence is especially extensive for ABA renewal. Including all three forms does not imply equal evidentiary depth.

Renewal Is Not Every Return

Renewal belongs to a wider family of return effects.

In spontaneous recovery, responding returns after a delay. Its defining change is temporal rather than movement outside an extinction context.

In reinstatement, responding returns after later exposure to a relevant event or outcome.

In resurgence, an earlier operant response returns when conditions supporting a newer alternative worsen.

In reacquisition, the earlier response is learned again after its previous contingency is restored.

These effects can occur alongside contextual changes, but they are not interchangeable. A behaviour returning after time has passed is not automatically renewal. Neither is an old action returning because a newer alternative is no longer reinforced.

Renewal should be used only when prior response reduction and testing outside the reduction context are relevant to the observed return.

What Renewal Can Explain

Renewal demonstrates that the generalisation of response reduction can be limited.

A response can decline reliably in one configuration of conditions without the newer learning governing behaviour equally well elsewhere. This does not make the reduction false. It shows that its expression was more context-sensitive than performance in one setting revealed.

An everyday routine can illustrate the principle. Someone establishes a newer routine under one recurring set of conditions. When those conditions change, an earlier routine becomes more likely again.

That sequence may resemble renewal, but it cannot usually establish the technical effect. An everyday case rarely reveals precisely whether the earlier response had been reduced, which contextual features mattered or whether another return process better explains the recurrence. Travel, moving home or changing schedules can alter cues, opportunities, consequences and social arrangements simultaneously.

Laboratory renewal can be identified because acquisition, response reduction, context and testing are controlled. Outside the laboratory, contextual recurrence should therefore be treated as possible evidence of context sensitivity, not automatically diagnosed as renewal.

Renewal does not show that the earlier response is the person's real response. It does not show that nothing changed.

It shows that change can be genuine and still be context-sensitive.