The case of the disappearing bone: dogs' understanding of the physical properties of objects

Abstract

To assess dogs' memory for an occluded object, a gaze duration procedure was used similar to one often used with nonverbal infants. A bone shaped dog biscuit was placed behind a solid screen that then rotated in the depth plane through an arc front to back. Dogs were shown either of the two test events. In one event (the possible event), the screen rotated until it reached the point at which it would have reached the bone and then stopped (about 120°); in the other event (the impossible event), the screen rotated through a full 180° arc, as though it had passed through the bone. The dogs looked significantly longer at the impossible event. To control for the differential time it took for the screen to move, for a control group, a bone was placed behind the screen and the screen was rotated either 60° or 120° (both possible events). No difference in looking time was found. To control for the movement of the screen through 120° or 180° when both were possible, for a second control group, the bone was placed to the side of the screen rather than behind the screen and the screen was moved 120° or 180°. Again, no significant difference in looking time was found. Results suggest that much like young children, dogs understand the physical properties of an occluded object. That is they appear to understand that an object (such as a screen) should not be able to pass through another object (such as dog bone).