Abstract
Lateral facilitation for detection using luminance-modulated targets, unlike contour integration, has been suggested to be a purely monocular phenomenon (Huang, Hess & Dakin, Vis Res, 2006). However facilitation for detection of contrast-modulated targets in normal vision, does not occur in amblyopes (Wong, Levi & McGraw, Vis Res, 2005), suggesting that this facilitation requires normal binocular processing. We determined whether facilitation occurred dichoptically using luminance-modulated and contrast-modulated Gaussian stimuli to assess its neural locus.
Foveal detection thresholds for luminance-modulated and contrast-modulated Gaussian blobs in the presence of visible, laterally placed blobs (separations of 0–6 deg) were measured monocularly and dichoptically in observers with normal vision. Blobs were constructed by adding or multiplying random-dot dynamic noise with a Gaussian (σ=0.25 deg). Data were collected using a method of constant stimuli and temporal 2AFC paradigm. Psychometric function slopes were analysed in order to assess the role of uncertainty reduction in lateral interaction effects.
Monocular detection thresholds measured for luminance-modulated blobs and contrast-modulated blobs in the presence of visible flankers follow a standard pattern of lateral interaction. Masking occurs for overlapping blobs, followed by facilitation when they are completely separated (1–3 deg). Dichoptic viewing produces a similar pattern of results but with some differences. For both luminance-modulated and contrast-modulated blobs, there is more masking dichoptically for overlapping blobs. For luminance-defined blobs, threshold facilitation of 15–30% is demonstrated dichoptically. For contrast-defined blobs, more robust facilitation of 30–50% is demonstrated dichoptically. Psychometric function slopes were analysed and were not consistently shallower in the facilitation region. Thus facilitation is likely to reflect neural processing.
Lateral facilitation is not purely a monocular phenomenon and cannot be simply explained by uncertainty reduction. Facilitation for contrast-defined stimuli appears more robust dichoptically than for luminance-defined stimuli, which may suggest a more binocular locus for neural processing for these stimuli.
Acknowledgment: MIH holds a PhD Scholarship funded by the Government of Malaysia.