Discriminative stimulus

Discriminative stimulus refers to the internal state produced by a drug — a pattern of physiological and sensory experience — that a trained animal learns to use as a behavioural cue.

In a standard drug-discrimination study, an animal (usually a rat or pigeon) is trained to associate one drug state with one response — pressing a particular lever — and a vehicle or different drug state with another. Once that discrimination is stable, a test compound is given and the animal's choice reveals how closely the new compound's internal effects match those of the training drug.

The technique sits within behavioural pharmacology and is used almost entirely in preclinical research. It answers a narrow, well-defined question: does this compound produce an internal state that a trained subject treats as equivalent to a known one?

What this design can establish

A well-conducted drug-discrimination study can establish pharmacological similarity at the level of subjective cue. Full substitution — the animal choosing the training-drug lever at the same rate as when given the training drug itself — is evidence that the two compounds produce overlapping internal signals.

Partial substitution indicates partial overlap; failure to substitute suggests the internal profiles are distinct. The method is also used to probe receptor involvement: if blocking a specific receptor prevents substitution, that receptor is likely carrying the trained signal.

Because the design uses a real behavioural outcome rather than a binding assay, it captures functional similarity — how a drug's aggregate effects register as a cue — rather than structural or receptor-binding similarity alone.

What it cannot

Full substitution does not establish a shared mechanism. Two compounds can produce indistinguishable discriminative stimuli through very different pharmacological routes — a releaser and a reuptake inhibitor may both fully substitute for each other if their downstream effects converge on the same trained cue.

It does not establish that the resemblance extends to human experience. Drug discrimination studies are conducted under controlled laboratory conditions in non-human animals. The subjective texture of a person's response — shaped by expectation, context, and individual variation — may diverge substantially from what lever-pressing implies.

The result also says nothing about frequency, severity, or generalisability. A full substitution finding is a statement about a trained cue in a specific animal population; it does not generalise to population-level human experience.

The most common inferential error is reading full substitution as confirmation that a novel compound works like a reference drug in a clinical or therapeutic sense. That conclusion requires evidence the method alone cannot provide.