Abstract
A Grey parrot, Griffin, previously taught English labels for various colors (red, blue, green, yellow, orange, purple) and shapes (regular polygons labeled "1-", "2-", "3-", "4-", "6-", "8-corner"), was tested on modal and amodal completion. Stimuli were primarily laser-printed paper shapes. For amodal completion, portions of variously colored regular polygons were occluded by black circles (which Griffin could not label) or other black polygons. Occasionally, occlusion resulted in a potentially confusing figure (e.g., triangle occluded by square to form trapezoid). Controls were colored polygons missing circular pieces and black circles appropriately displaced. For modal completion, Kanizsa figures were constructed using black ‘pac-men’ to form regular polygons on colored paper. Controls involved placing additional circles or ‘pac-men’ near the Kanizsa figure so Griffin could not simply count black objects. The principal investigator (PI) placed stimuli 15 cm from one of Griffin’s eyes (distance determined as optimal in Pepperberg, Vicinay, Cavanagh, 2007), the other eye being focused on the PI; Griffin was then queried "What shape X?", where X was the color of the polygon in question. Griffin provided a vocal English shape label (using possible labels to 8). An undergraduate experimenter repeated Griffin’s answer; only if her response matched the correct response was the parrot’s label considered correct. Griffin received one trial per session, with sessions occurring 1-4x/wk, with breaks for student exam and intersession periods. Griffin’s responses were very accurate (25/33 correct on Kanizsa figures; 20/24 correct for amodal completion; chance was 0.20). Results demonstrate that despite a somewhat different brain structure and considerably different visual system from that of humans, a Grey parrot solves at least two of the same visual cognitive problems faced by humans. (Supported by NSF grants 0920878, 1026256 to Ken Nakayama and The Alex Foundation)
Meeting abstract presented at VSS 2012