Auditory generalization gradients for response latency in the monkey.

Scientists trained monkeys to release a key when they heard a certain tone.

Releasing fast after that tone earned a food pellet.

Other tones nearby in pitch never paid off.

The team then tested many tones to see how fast the monkey let go.

The monkey let go fastest at the reinforced pitch.

Latencies grew evenly as tones moved away, forming a smooth gradient.

The shape stayed stable across tests, showing clear stimulus control.

RISLEY (1964) ran a similar two-tone task and also saw a latency gradient, but it had two peaks instead of one.

The difference is small: both studies show that timing, not just rate, can map stimulus control.

AZRIN et al. (1963) later repeated the idea with blind children who had severe ID.

Their gradients were lopsided, proving the same rules work across species and abilities.

SIDMAELLIOTT et al. (1962) looked at the flip side—inhibitory control—and found flat, weak gradients.

Together the papers say: excitatory control sharpens timing, while inhibitory control stays fuzzy.

You can use latency as a live measure of stimulus control.

If a learner responds faster to one word, color, or sound, you know that cue is winning.

Track release times during discrimination drills to spot when generalization is too wide.

Then tighten training by reinforcing only the target and thinning rewards for close, non-target items.