Purchase this article with an account.
Zheng Ma, Michael McCloskey, Jonathan Flombaum; Differentiating between object-dependent and transient-dependent motion percepts through crowding. Journal of Vision 2014;14(10):278. doi: https://doi.org/10.1167/14.10.278.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Motion can be perceived by detecting changes in an object's location, or by detecting transient changes in sensory channels (e.g. luminance). The relative contribution of these mechanisms can be difficult to ascertain in many situations. We used spatial crowding to dissociate object- and transient-dependent mechanisms, reasoning that crowding should impair motion perception relying on object localization. Experiment1 investigated continuous motion, perception of which is typically thought to rely upon transient detection. A gray disc moved continuously through 0.64째, and participants reported the motion direction (left or right). Crowding of the moving disc by stationary discs more severely impaired judgment accuracy for slow (.08째/s) than fast (0.64째/s) motion. This result suggests that perception of slow continuous motion relies more on higher-level object systems, probably because slow motion produces transient signals that are too weak for summation. Experiment2 applied a crowding manipulation to the Ternus display, which is known to elicit percepts of element motion (one object moving with two stationary middle objects) or group motion (all three objects moving together). We hypothesized that element motion is possibly the output from transient detectors, since this percept violates object coherence. In contrast, group motion should be the output of an object-dependent system, as it arises from a construal of the least coincidental change to all objects. Participants viewed 0ms ISI displays, and reported whether they saw group or element motion. With stationary bars crowding the Ternus stimuli, participants reported predominantly element motion, whereas group motion was reported predominantly with uncrowded stimuli. These results suggest that Ternus percepts are the outputs of independent motion systems, differentially reliant upon low-level transients vs. object position representations. Together, these experiments evidence the utility of crowding to isolate motion systems, and suggest a taxonomy of motion perception deriving from its informational basis.
Meeting abstract presented at VSS 2014
This PDF is available to Subscribers Only