Orientation bandwidths of cyclopean channels measured with the use of notched noise

P B Hibbard

Research output: Contribution to journalAbstractpeer-review


The spatial frequency bandwidths of cyclopean channels, responding to depth variations defined by binocular disparity, were measured by masking experiments and estimated to lie between 1 - 3 octaves (Cobo-Lewis and Yeh, 1994 Vision Research 34 607 - 620). Here, a notched noise procedure was used to measure the orientation bandwidth of these channels. Thresholds were measured for the detection of sinusoidal depth modulations of a random-dot stereogram, either in isolation or in the presence of a disparity mask. Thresholds were measured for both horizontal and vertical modulations. The mask was formed from two samples, each the sum of a number of sinusoidal depth variations covering a 10° range of orientations. The orientation ranges of the two samples were positioned symmetrically around the signal, leaving a gap (notch) around the signal of between ±10° and ±80°. Signal detection thresholds were measured for each notch size, while the magnitude of the noise was held constant. Threshold elevation relative to the unmasked condition decreased as the size of the notch increased. Threshold elevation as a function of mask amplitude was also measured for masks centred on the signal orientation, allowing the orientation bandwidth of the channels to be calculated. The bandwidth (half-width at half-height) was estimated to be around 35°, for channels tuned to both horizontal and vertical orientations. These results are considered in relation to the suggestion that different areas of spatial summation underlie the detection of different orientations of cyclopean stimuli (Tyler and Kontsevich, 2001 Vision Research 41 2235 - 2243).
Original languageEnglish
Pages (from-to)75-76
Number of pages2
Issue numberECVP Abstract Supplement
Publication statusPublished - 2003


Dive into the research topics of 'Orientation bandwidths of cyclopean channels measured with the use of notched noise'. Together they form a unique fingerprint.

Cite this