Robust representation of shape in a Grey parrot (Psittacus erithacus)

A Grey parrot, Griffin (Psittacus erithacus), previously taught English labels for various colors and shapes with respect to three-dimensional (3D) stimuli, was tested on his ability to transfer to very different two-dimensional (2D) images consisting of modal and amodal completion stimuli. For modal completion (aka subjective contours), Kanizsa figures were constructed using black 'pac-men' to form regular polygons on colored paper. For amodal completion, portions of variously colored regular 2D polygons were occluded by black circles or other black figures. For each task, Griffin provided a vocal English shape label for five possible shapes designated by their vertices (one, two, three, four, six). His accuracy was high for both amodal completed figures, including probe stimuli (28/38 correct) and modally completed figures (29/38 correct), with chance=0.20. The modally completed case (i.e., Kanizsa subjective figures) is of particular importance as there are no shared image parts between training and testing stimuli. We draw several conclusions from these results. First, a surface level completion process is fully operative insofar as Griffin was able to correctly identify shapes that differed considerably from training images. Second, because parrots can generalize from shapes of real objects to drawings where original image contours were clearly absent, the data provide a compelling example of shape invariance, indicating that visual shapes are processed far beyond that of their image description. Third, parrots with a repertoire of multiple vocal responses can be rigorously tested for visual competencies, an option as yet to be tried in other experimental animals.