摘要

Many visual processes integrate information over protracted periods, a process known as temporal integration. One consequence of this is that objects that cast images that move across the retinal surfaces can generate blurred form signals, similar to the motion blur that can be captured in photographs taken with slow shutter speeds. Subjectively, retinal motion blur signals are suppressed from awareness, such that moving objects seem sharply defined. One suggestion has been that this subjective impression is due to humans not being able to distinguish between focussed and blurred moving objects. Contrary to this suggestion, here we report a novel illusion, and consequent experiments, that implicate a suppressive mechanism. We find that the apparent shape of circular moving objects can be distorted when their rear edges lag leading edges by similar to 60 ms. Moreover, we find that sensitivity for detecting blur, and for discriminating between blur intensities, is uniformly worse for physical blurs added behind moving objects, as opposed to in-front. Also, it was easier to differentiate between slight and slightly greater physical blurs than it was to differentiate between slight blur and the absence of blur, both behind and in-front of moving edges. These 'dipper' functions suggest that blur signals must reach a threshold intensity before they can be detected, and that the relevant threshold is effectively elevated for blur signals trailing behind moving contours. In combination, these data suggest moving objects look sharply defined, at least in part, because of a functional adaptation that actively suppresses motion blur signals from awareness.

  • 出版日期2013-8-9