Harrison S J, Keeble D R T, 2003, "Texture orientation contrast across a subjective contour aids detection, but not discrimination of boundary orientation" Perception 32 ECVP Abstract Supplement
Texture orientation contrast across a subjective contour aids detection, but not discrimination of boundary orientation
S J Harrison, D R T Keeble
Neuronal activity in V1 is modulated by nearby orientation contrast, either a single pop-out element or a boundary defined by many local orientation contrasts. In comparison, subjective contours defined by aligned line-terminations cause modulation of activity in V2 but only very weak modulation in V1. These modulations in orientation-tuned cells are caused by the integrated contour, not just a single line-termination. Combinations of mapping techniques suggest that modulation by subjective contours may be due to feedback from higher cortical areas rather than originating in V1 or V2. In the present study, we used a speed - accuracy trade-off paradigm to investigate the temporal dynamics of texture border processing. The texture stimulus was composed of randomly placed, uniformly oriented line segments. A discontinuity was defined by either a boundary of aligned line terminations (subjective contour), or with an additional change in orientation across the boundary (orientation contrast). In a detection task, subjects indicated whether the boundary appeared to the left or right of fixation, and in an orientation-discrimination task whether the boundary tilted clockwise or anticlockwise. Detection of contours with orientation contrast was maximal at all response times and eccentricities. Detection of subjective contours was slower and asymptoted at a lower accuracy as eccentricity was increased. After correction for lower detection rates, we found that discrimination performance for subjective contours was not significantly different from that for contours with orientation contrast. We conclude that while orientation contrast is a powerful cue for detection, boundary integration occurs at a later processing stage in common with that for subjective contours.
[Supported by EPSRC studentship.]
These web-based abstracts are provided for ease of seaching and access, but certain aspects (such as as mathematics) may not appear in their optimum form. For the final published version of this abstract, please see
ECVP 2003 Abstract Supplement (complete) size: 1712 Kb