The pigeon's perception of saturation.
Pigeons see color saturation like we do—dullest in green-yellow—so pick hues away from that band when you need clear stimulus control.
01Research in Context
What this study did
The team asked pigeons to peck one key when color looked bright and another when it looked dull.
They tested many wavelengths to map how "grayish" each color looked to the birds.
What they found
Green-yellow light (550-600 nm) looked least saturated to the pigeons.
Colors on each side of that band looked richer, just like they do to people.
How this fits with other research
Gentry et al. (1980) and Rapport et al. (1982) later showed that changing payoff rates shifts bias but not how well pigeons see the difference. Thompson (1975) gives the baseline: once you know the bird’s saturation curve, you can be sure the color difference is big enough before you test payoff effects.
Rand (1977) and Fields (1978) tracked response rate and bias during simple discriminations. Their payoff tricks work best when the stimulus difference is above the saturation dip shown in Thompson (1975); otherwise the bird may not see the cue at all.
Kendrick et al. (1981) used delayed matching and found context controls memory. Picking colors near the low-saturation band from Thompson (1975) would make their task harder, so their strong effects likely used clearer hues.
Why it matters
If you run pigeon labs or teach stimulus control, check your color wheels. Avoid green-yellow when you need a sharp difference; use those wavelengths when you want a subtle one. Set saturation first, then play with reinforcement rates or memory delays. The bird’s eyes already give you the roadmap.
Want CEUs on This Topic?
The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.
Join Free →Check your color stimuli: swap any green-yellow cues for richer blue or red if the bird’s accuracy is low.
02At a glance
03Original abstract
Three experiments used similar methods to investigate the pigeon's perception of saturation of monochromatic lights. This trial-wise procedure consisted of brief presentations of positive and negative stimuli in random sequence. Pecks to the positive stimuli were occasionally reinforced on a low fixed-ratio schedule. The first study determined absolute thresholds for "white" and monochromatic lights by establishing a discrimination between lights of various radiances and a dark key. Experiment II investigated generalization from a white light to various monochromatic lights under conditions that minimized the use of luminance as a cue. The third experiment examined discrimination of various monochromatic lights along a colorimetric purity continuum; responses to white light were reinforced, while responses to lights that combined white and monochromatic lights in various proportions were not. The results indicated that lights of different wavelength differ in saturation, but that all are discriminable from white. Wavelengths between 550 and 600 nm are least saturated for the pigeon, and saturation increases markedly as wavelength decreases below this region of the spectrum.
Journal of the experimental analysis of behavior, 1975 · doi:10.1901/jeab.1975.24-135