Effects of size on judgments of collision: Theoretical and practical implications

To safely maneuver through the environment, people have to make accurate judgments of collision (Cutting, Vishton & Braren, 1995). Such judgments may have to be made when one or more objects are present in a scene, and when the objects are of different sizes (e.g., DeLucia, 2004). Prior research has overlooked two important types of collision judgments; judgments of whether a moving object would collide or stop moving before colliding with a target when the moving object is either small or large, and judgments of which object would collide with a target when two moving objects of different sizes are present in a scene. Lee (1976) showed mathematically that the information about the potentiality of a collision is specified veridically in the optic array by the optical parameter tau-dot, and hence observers can rely on tau-dot to make judgments of collision. This has been referred to as the tau-dot hypothesis (Kim, Effken & Carello, 1998). However, results from other studies have indicated that judgments of collision may be affected by biases in the perceived velocity of motion (Andersen, Cisneros, Atchley & Saidpour, 1999). Such a pattern of judgments has been explained with the constant deceleration model (Andersen et al., 1999). When motion is in the frontoparallel plane, the size of a moving object induces biases in its perceived velocity (Brown, 1931). Therefore, in three experiments the effects of size on two types of judgments of collision were examined; judgments of which moving object would collide with a stationary target and judgments of whether a moving object would collide with a stationary target. Results indicated that retinal size and tau-dot affected such judgments, suggesting that multiple sources of visual information could affect judgments of collision. Theoretical implications for the tau-dot hypothesis (Lee, 1976) and the constant deceleration model (Andersen et al., 1999) have been discussed. In addition, practical implications for the design of collision avoidance warning systems in vehicles, and for driver education and training have been discussed.