Isolating the Error in Clocking Process from the Error in
Time-to-contact Based Performance

Jiantao Pei

Centre for Applied Psychology, University of Canberra,
Canberra, Australia


There has already been a great deal of research on the mechanism of Time-to-contact (TTC) estimation with a variety of TTC based tasks, but so far no agreement has been achieved. While high accuracy was exposed on TTC based interceptive tasks (e.g., hit a baseball), TTC studies based on the dominant method, the Prediction Motion (PM) task, widely report large variability and a tendency to underestimate TTC (i.e., to anticipate the moment of collision), which is unusual and embarrassing. This study is predicated on the argument that the PM task used by previous studies is deficient (with the contamination of the Cognitive Motion Process [CME], DeLucia & Lindell, 1998). The present paper will be concerned with the accuracy with which we estimate time-to-contact (TTC) under conditions where we can assume that this is based on direct perception of visual flow.

On a Silicon Graphics (SGI) workstation, we manipulated the properties of background visual flow without changing the actual expansion rate of the target, and thereby presented simulated in-depth looming targets with constant local tau values but with varied display properties (e.g., background optic flow, approaching velocity, and rotating velocity). To isolate the error in clocking process from the error in TTC based performance, a novel procedure of timing temporal intervals was introduced.

Our study demonstrates that (1) the underestimation error of TTC estimates in the PM task is largely due to the error from temporal interval estimates; (2) variability of TTC estimates is mostly due to the inherent variability in the temporal interval estimates (clocking process); and (3) observers perceived and distinguished the TTC optic cue (e.g., Tau) in estimating TTC. Optical variations such as background optic flow, target size, approaching velocity had no effect on TTC estimates, refuting the suggestions that CME procedures are involved in cognitive operations in the PM task, and also favouring suggestions that direct perception of Tau is the basis for estimating TTC. It is hoped that this study will stimulate further investigation in this field.