Optical specification of time-to-contact for a spectator

In this article, we provide an account of the informational basis underlying judgments of time-to-contact (TTC) under the conditions examined in Kim, Effken, and Carello (1998), in which observers watched collisions between 2 objects from various vantage points, over varying collision paths, and with additional observer movement. First, we examine the informational candidates proposed by Bootsma and his colleagues (e.g., Bootsma & Oudejans, 1993) and by Tresilian (1990). Simulation results showed that these models do not: accurately describe TTC for such varying conditions. We present an alternative model, emphasizing the advantages of body scaling. Our model applied this scaling to the vertical gap between edges of the opposing surfaces of each object. Because all model terms are expressed in eyeheight units, distortions that occurred in the other models due to polar projection are eliminated. The proposed model provided veridical estimates of TTC for straight-line collision cases over a variety of viewing angles and with the addition of observer movement, but not for collisions along circular paths, in which the model only approximated TTC values.