The bisection of a brightness interval by pigeons.

Researchers taught pigeons to split a brightness range into two equal parts. The birds first learned to peck one key when the light was dim and another key when the light was bright.

After training, the team slipped in surprise test lights. The pigeons had to pick the side that felt like the middle brightness. This probe method kept the birds from learning new rules during testing.

The pigeons kept choosing the same middle brightness again and again. Their picks lined up with the general mean theorem, a simple rule that predicts the midpoint of any range.

The probe trials did not change the birds' trained choices, showing the method stayed clean and stable.

Walker (1968) ran the same split-the-interval task one year earlier, but used left-right space instead of brightness. Both studies prove the successive-discrimination trick works across senses.

Raslear (1975) later moved the idea to loudness with rats. G found that spacing the test sounds could shift the midpoint, while Nevin (1969) saw no shift with light. The difference is the stimulus set, not the species: brightness gave a stable middle, sound did not.

Hodos et al. (1976) showed pigeons see best at moderate luminance. Nevin (1969) picked that sweet spot for the bisection lights, so the birds' steady choices may owe partly to their visual peak.

You can borrow the probe-trial trick any time you need a pure read of stimulus control. Train two clear anchors, then sneak in test values without extra rewards. The method keeps the learner's original rule intact while you map the midpoint or any other landmark. It works with pigeons, rats, and likely humans—just match the stimulus range to the client's sensory sweet spot.