Purchase this article with an account.
D. Goren, J. G. Flanagan; Resolution of Flicker Defined Form. Invest. Ophthalmol. Vis. Sci. 2007;48(13):5889.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
To investigate resolution thresholds for flicker defined form (FDF), a magnocellularly driven illusion. FDF is defined by high temporal frequencies that create an illusory boundary between counterphase-flickered elements. Spatial changes such as dot density and organization, target area and continuous contour have been shown to affect stimulus detection.
FDF is created by flickering random dots of 0.34° at 17Hz square wave flicker against a mean luminance background (50cdm-2). Dots within the 5° stimulus were flickered in counterphase to background dots. Stimuli consisted of different shapes (circle, square, triangle) and circle permutations (ring, arc). Subjects performed a temporal, 2 forced-choice procedure, in which they were asked to identify the interval containing the circle. Method of constant stimuli was used to asses contrast thresholds at 4 eccentricities (0°, 4°, 13° and 21°). Results were recorded in log Michelson contrast (MC%).
: Circular FDF stimuli could be discriminated from other shapes, rings and arcs with great accuracy. Resolution and detection thresholds were similar at all eccentricities other than fixation. Resolution thresholds for triangles compared to circles were lower than for other stimuli, ranging from 0.6 to 0.9 log MC% (across eccentricities), as compared to detection thresholds between 0.8 and 1.8 log MC% (with highest thresholds at fixation). For stimuli which were hard to detect (such as broken and thin rings), resolution thresholds were lower than detection thresholds, particularly at fixation (e.g. differences between 0.3 and 0.8 MC%). At further eccentricities, the differences between detection and resolution were minimal.
Differences between detection and resolution were most evident for difficult-to-detect stimuli at fixation. In these cases, resolution had lower thresholds than corresponding detection thresholds. Minimal differences were found for stimuli which were easy to detect. The implications for these differences will be discussed.
This PDF is available to Subscribers Only