Attention and "visual field dependency" in the pigeon.
Reinforcement can flip an animal’s default cue from what looks right to what feels level.
01Research in Context
What this study did
The team worked with three pigeons in a small lab chamber. The floor of the chamber could tilt 20° left or right.
Two keys lit up with white lines. One line stayed truly vertical to gravity. The other line stayed vertical to the bird’s eyes even when the cage tilted.
Pecks on the gravity-true key paid food. Pecks on the eye-true key paid nothing. The tilt changed every few minutes.
What they found
At first the birds always pecked the eye-true key. After a few sessions they switched and pecked the gravity-true key almost every time.
The birds learned to ignore how the line looked and used gravity instead. Reinforcement overrode their built-in visual habit.
How this fits with other research
Cicerone (1976) showed that pigeons can follow either the overall rate of food or quick signal changes. Wilkie (1973) adds a third cue—body tilt—proving that any reliable payoff can grab control.
Dews (1978) found that pairing a keylight with food later makes that key glow like a reward. Wilkie (1973) goes further: the payoff can even flip what "up" means to the bird.
Grant (1989) showed birds prefer pecking over treadling when both ends pay the same. Wilkie (1973) shows the same peck can be steered toward gravity when the schedule picks that form.
Why it matters
Your client may cling to a built-in cue—eye-catching toys, loud voices, or room tilt. This study says you can swap that control to any cue you reinforce. Put the reinforcer on the quiet, calm, or level choice and the child will shift just like the pigeons did.
Want CEUs on This Topic?
The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.
Join Free →Pick one context where the client reaches for the "eye-catch" item; reinforce reaches for the stable, gravity-true alternative instead.
02At a glance
03Original abstract
Three pigeons were trained in an upright conditioning chamber to peck a key transilluminated by a vertical line. This training was followed by a line orientation generalization test. During the test, the chamber was tilted laterally 22.5 degrees from upright. The chamber floor remained horizontal with respect to gravity. Under these conditions, the subjects responded more often in the presence of a visually vertical (parallel to chamber walls) line orientation than in the presence of a gravitationally vertical line orientation. Subsequent reinforcement of pecking in the presence of a line that was always gravitationally vertical but not always visually vertical temporarily abolished this "visual field dependency" and resulted in generalization gradients with peak responding in the presence of the gravitationally vertical line orientation. The results are discussed in terms of selective attention to the gravitational and visual components of line orientation.
Journal of the experimental analysis of behavior, 1973 · doi:10.1901/jeab.1973.20-7