Related Experiment Videos
Are judgements of circularity local or global?
R F Hess1, Y Z Wang, S C Dakin
1McGill Vision Research, Department of Ophthalmology, McGill University, Montreal, Québec, Canada. rhess@bradman.vision.mcgill.ca
Vision Research
|May 2, 2000
Summary
Human visual perception of shape relies on a global mechanism, not just local details. This system processes overall form, using coarse filters tuned to orientation, to detect deviations from circularity.
Area of Science:
- Visual perception
- Computational neuroscience
- Psychophysics
Background:
- The human visual system processes shape information through various mechanisms.
- Understanding whether shape perception relies on local feature detection or global form analysis is crucial.
Purpose of the Study:
- To investigate whether the visual mechanisms for detecting smooth deviations from circularity are spatially localized to individual perturbations or compute global shape properties.
- To determine the characteristics of the visual filters involved in global shape detection.
Main Methods:
- Subjects performed a task requiring detection of smooth deviations from circularity.
- Experimental manipulations included varying the phase, number of modulation cycles, and spatial arrangement of perturbations.
- Performance metrics were analyzed to distinguish between local and global processing mechanisms.
Main Results:
- While both local perturbations and global shape aspects could be detected, overall performance was limited by a global shape detection mechanism.
- This global mechanism was found to receive input from spatially coarse filters.
- These filters exhibit cross-orientation tuning, with peak orientation sensitivity dependent on the overall shape being detected.
Conclusions:
- Visual shape perception, specifically for deviations from circularity, is primarily constrained by a global shape analysis mechanism.
- This global mechanism utilizes spatially coarse, orientationally tuned filters to process overall form information.
- The findings shed light on the hierarchical processing of visual shape information in the human brain.