Some effects of response-independent reinforcement on auditory generalization gradients.

The team worked with rats in a small lab chamber.

They played tones of different pitches through a speaker.

Food came on a variable-time schedule no matter what the rat did.

The researchers tracked how often the rat pressed a lever during each tone.

They compared this to a second group that got no food at all.

Even without tying food to lever presses, the rats still pressed more to some tones.

Their response curve shifted away from the training tone, a pattern called peak shift.

The no-food group showed the same shift, only sooner and with a cleaner curve.

Variable-time food also created a small second bump in the curve.

Zeiler (1969) showed pigeons make steeper visual gradients when the schedule is FI or VI instead of FR.

C et al. now add that VT food, which never depends on a response, can still sculpt an auditory gradient.

Blue et al. (1971) saw no behavioral contrast under VT, yet here VT did shape stimulus control.

The difference is the measure: contrast looks between components, gradients look across stimuli.

Badia et al. (1972) found zero stimulus control with non-differential reinforcement, but their tones signaled location, not pitch, and they kept reinforcement frequent.

C et al. used extinction after training, letting peak shift emerge even when food had been non-contingent.

You now know that schedules you think are "neutral" can still teach.

If you run baseline sessions with free reinforcement, check afterward for accidental stimulus control.

Before probing generalization, switch to brief extinction; you will see a clearer peak shift faster.

This saves session time and sharpens your data.