The McCollough effect in pigeons: tests of persistence and spatial-frequency specificity.

Eight pigeons were trained on a conditional discrimination using red and green saturated and desaturated fields, and red and green saturated and desaturated vertical and horizontal black-on-colour gratings. The pigeons learned to discriminate the stimuli on the basis of colour, to a high level of accuracy, regardless of saturation or the presence of gratings. The pigeons were then repeatedly exposed to stimuli in which colour and grating orientation were correlated, following which they were tested for the presence of the McCollough orientation-contingent colour after-effect, using black-on-white vertical and horizontal gratings. Six of the birds showed convincing evidence of the presence of the McCollough effect, and the effect was significant across all birds. These findings support those of Roberts (1984), using a substantially different methodology. The six birds showing the McCollough effect were then tested for the persistence of the effect at delays of 24 to 96 h. Four of the birds showed evidence of the McCollough effect at least 24 h after the induction procedure. Three of these birds were also tested to investigate the spatial frequency selectivity of the effect. The results suggest a narrow tuning of the McCollough effect in pigeons of less than 0.36 log units.