Things that go bump in the light: on the optical specification of contact severity.

Psychologists are intrigued with the idea that optical variables can specify not only the time until an object impacts an observer but also the severity of the impact. However, the mapping between the optical variables (tau and .tau) and the kinematic variables (velocity, acceleration) has been misstated, erroneously implying that there exist critical values of the optical variables used for locomotion and control. In this commentary, the mathematical relationship between the optical and kinematic variables is reexamined and the erroneous assumptions that have led to the proposal of critical values are show. Also examined are the empirical data on deceleration to approach (particularly from active control paradigms) to assess whether the proposed optical variables are likely candidates for control strategies. Finally, problems associated with numerical approximations to dynamic systems, particularly when analytic solutions exist, are discussed.

The visually guided control of simulated altitude.

Simulated "flights" over three different ground textures were used to examine people's ability to extract optical information useful for active regulation of altitude. The textures were regularly spaced lines as follows: 1) orthogonal to the direction of flight (latitude texture); 2) parallel to the direction of flight (meridian texture); and 3) both parallel and orthogonal (square texture). Visual constant velocity forward flight simulations were displayed on a CRT screen, and subjects asked to maintain one of three initial altitudes using a rate control stick. This task was made difficult by the presence of lateral (irrelevant) and vertical (relevant) "wind gusts." The attitude never varied as winds, forward speed, and vertical rate control resulted in only translational movements. Adjusted root mean square errors (ARMSE) showed altitude regulation was more difficult at higher altitudes and when flying over meridian textures. Refined analysis of a single subject's data showed that this was due both to poorer regulation of the vertical wind disturbance and to a tendency to confuse the lateral wind disturbance for a vertical disturbance.