Combining S-cone and luminance signals adversely affects discrimination of objects within backgrounds

Benjamin J. Jennings, Konstantinos Tsattalios, Ramakrishna Chakravarthi, Jasna Martinovic

Research output: Contribution to journalArticlepeer-review

2 Citations (Scopus)
8 Downloads (Pure)


The visual system processes objects embedded in complex scenes that vary in both luminance and colour. In such scenes, colour contributes to the segmentation of objects from backgrounds, but does it also affect perceptual organisation of object contours which are already defined by luminance signals, or are these processes unaffected by colour's presence? We investigated if luminance and chromatic signals comparably sustain processing of objects embedded in backgrounds, by varying contrast along the luminance dimension and along the two cone-opponent colour directions. In the first experiment thresholds for object/non-object discrimination of Gaborised shapes were obtained in the presence and absence of background clutter. Contrast of the component Gabors was modulated along single colour/luminance dimensions or co-modulated along multiple dimensions simultaneously. Background clutter elevated discrimination thresholds only for combined S-(L + M) and L + M signals. The second experiment replicated and extended this finding by demonstrating that the effect was dependent on the presence of relatively high S-(L + M) contrast. These results indicate that S-(L + M) signals impair spatial vision when combined with luminance. Since S-(L + M) signals are characterised by relatively large receptive fields, this is likely to be due to an increase in the size of the integration field over which contour-defining information is summed.

Original languageEnglish
Article number20504
Pages (from-to)1-10
Number of pages10
JournalScientific Reports
Publication statusPublished - 9 Feb 2016


Dive into the research topics of 'Combining S-cone and luminance signals adversely affects discrimination of objects within backgrounds'. Together they form a unique fingerprint.

Cite this