Abstract
To intercept a moving object, the brain must compute a reliable estimate of time-to-contact (TTC), which can be inferred from measurements of speed, elapsed time and position. Since these variables are physically interrelated, it is challenging to tease apart the sensory cues that guide this behavior. Previous studies have shown that humans rely on speed estimation and distance to compute TTC. However, given the brain's capacity to measure interval timing independent of speed, it is conceivable that humans utilize an estimate of elapsed time to further improve their performance. To test this possibility, we asked subjects to press a button when an initially visible moving object would arrive at a designated target location behind an occluder. We found that their TTC estimation had the lowest mean squared error when the visible and occluded regions had the same length and duration, suggesting that subjects might additionally use a measure of elapsed time to compute TTC. To test the hypothesis more directly, we modified the object interception task to dissociate speed and timing information. The speed cue was provided by a brief visible region at the beginning of the trial, whereas the timing cue was provided by a flash bisecting the actual TTC. When the cues were presented separately, a Bayesian model integrating prior distribution with relevant sensory measurement accounted for performance. When both cues were available, performance improved beyond what could be attributed to either time or speed measurement alone, suggesting that subjects combined both speed (from the visible region) and elapsed time (from the flash) cues to estimate TTC. An additional cue conflict paradigm further substantiated the evidence that subjects integrate both speed and elapsed time to optimize their performance. These findings suggest that the brain integrates signals associated with timing and speed to control motor responses during object interception.
Meeting abstract presented at VSS 2016