Suppression pathways saturate with contrast for parallel surrounds but not for superimposed cross-oriented masks

Timothy S. Meese, K.L. Challinor, Robert J. Summers, Daniel H. Baker

Research output: Contribution to journalArticle

Abstract

Contrast masking from parallel grating surrounds (doughnuts) and superimposed orthogonal masks have different characteristics. However, it is not known whether the saturation of the underlying suppression that has been found for parallel doughnut masks depends on (i) relative mask and target orientation, (ii) stimulus eccentricity or (iii) surround suppression. We measured contrast-masking functions for target patches of grating in the fovea and in the periphery for cross-oriented superimposed and doughnut masks and parallel doughnut masks. When suppression was evident, the factor that determined whether it accelerated or saturated was whether the mask stimulus was crossed or parallel. There are at least two interpretations of the asymptotic behaviour of the parallel surround mask. (1) Suppression arises from pathways that saturate with (mask) contrast. (2) The target is processed by a mechanism that is subject to surround suppression at low target contrasts, but a less sensitive mechanism that is immune from surround suppression ‘breaks through’ at higher target contrasts. If the mask can be made less potent, then masking functions should shift downwards, and sideways for the two accounts, respectively. We manipulated the potency of the mask by varying the size of the hole in a parallel doughnut mask. The results provided strong evidence for the first account but not the second. On the view that response compression becomes more severe progressing up the visual pathway, our results suggest that superimposed cross-orientation suppression precedes orientation tuned surround suppression. These results also reveal a previously unrecognized similarity between surround suppression and crowding (Pelli, Palomares, & Majaj, 2004).
Original languageEnglish
Pages (from-to)2927-2935
Number of pages9
JournalVision Research
Volume49
Issue number24
DOIs
Publication statusPublished - 10 Dec 2009

Fingerprint

Masks
Visual Pathways

Bibliographical note

Copyright of Elsevier

Keywords

  • human vision
  • contrast gain control
  • psychophysics
  • surround suppression
  • crowding
  • XOM
  • XOS
  • detection

Cite this

Meese, Timothy S. ; Challinor, K.L. ; Summers, Robert J. ; Baker, Daniel H. / Suppression pathways saturate with contrast for parallel surrounds but not for superimposed cross-oriented masks. In: Vision Research. 2009 ; Vol. 49, No. 24. pp. 2927-2935.
@article{c5d8d08908f44e15b83c795efe69f5ad,
title = "Suppression pathways saturate with contrast for parallel surrounds but not for superimposed cross-oriented masks",
abstract = "Contrast masking from parallel grating surrounds (doughnuts) and superimposed orthogonal masks have different characteristics. However, it is not known whether the saturation of the underlying suppression that has been found for parallel doughnut masks depends on (i) relative mask and target orientation, (ii) stimulus eccentricity or (iii) surround suppression. We measured contrast-masking functions for target patches of grating in the fovea and in the periphery for cross-oriented superimposed and doughnut masks and parallel doughnut masks. When suppression was evident, the factor that determined whether it accelerated or saturated was whether the mask stimulus was crossed or parallel. There are at least two interpretations of the asymptotic behaviour of the parallel surround mask. (1) Suppression arises from pathways that saturate with (mask) contrast. (2) The target is processed by a mechanism that is subject to surround suppression at low target contrasts, but a less sensitive mechanism that is immune from surround suppression ‘breaks through’ at higher target contrasts. If the mask can be made less potent, then masking functions should shift downwards, and sideways for the two accounts, respectively. We manipulated the potency of the mask by varying the size of the hole in a parallel doughnut mask. The results provided strong evidence for the first account but not the second. On the view that response compression becomes more severe progressing up the visual pathway, our results suggest that superimposed cross-orientation suppression precedes orientation tuned surround suppression. These results also reveal a previously unrecognized similarity between surround suppression and crowding (Pelli, Palomares, & Majaj, 2004).",
keywords = "human vision, contrast gain control, psychophysics, surround suppression, crowding, XOM, XOS, detection",
author = "Meese, {Timothy S.} and K.L. Challinor and Summers, {Robert J.} and Baker, {Daniel H.}",
note = "Copyright of Elsevier",
year = "2009",
month = "12",
day = "10",
doi = "10.1016/j.visres.2009.09.006",
language = "English",
volume = "49",
pages = "2927--2935",
journal = "Vision Research",
issn = "0042-6989",
publisher = "Elsevier",
number = "24",

}

Suppression pathways saturate with contrast for parallel surrounds but not for superimposed cross-oriented masks. / Meese, Timothy S.; Challinor, K.L.; Summers, Robert J.; Baker, Daniel H.

In: Vision Research, Vol. 49, No. 24, 10.12.2009, p. 2927-2935.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Suppression pathways saturate with contrast for parallel surrounds but not for superimposed cross-oriented masks

AU - Meese, Timothy S.

AU - Challinor, K.L.

AU - Summers, Robert J.

AU - Baker, Daniel H.

N1 - Copyright of Elsevier

PY - 2009/12/10

Y1 - 2009/12/10

N2 - Contrast masking from parallel grating surrounds (doughnuts) and superimposed orthogonal masks have different characteristics. However, it is not known whether the saturation of the underlying suppression that has been found for parallel doughnut masks depends on (i) relative mask and target orientation, (ii) stimulus eccentricity or (iii) surround suppression. We measured contrast-masking functions for target patches of grating in the fovea and in the periphery for cross-oriented superimposed and doughnut masks and parallel doughnut masks. When suppression was evident, the factor that determined whether it accelerated or saturated was whether the mask stimulus was crossed or parallel. There are at least two interpretations of the asymptotic behaviour of the parallel surround mask. (1) Suppression arises from pathways that saturate with (mask) contrast. (2) The target is processed by a mechanism that is subject to surround suppression at low target contrasts, but a less sensitive mechanism that is immune from surround suppression ‘breaks through’ at higher target contrasts. If the mask can be made less potent, then masking functions should shift downwards, and sideways for the two accounts, respectively. We manipulated the potency of the mask by varying the size of the hole in a parallel doughnut mask. The results provided strong evidence for the first account but not the second. On the view that response compression becomes more severe progressing up the visual pathway, our results suggest that superimposed cross-orientation suppression precedes orientation tuned surround suppression. These results also reveal a previously unrecognized similarity between surround suppression and crowding (Pelli, Palomares, & Majaj, 2004).

AB - Contrast masking from parallel grating surrounds (doughnuts) and superimposed orthogonal masks have different characteristics. However, it is not known whether the saturation of the underlying suppression that has been found for parallel doughnut masks depends on (i) relative mask and target orientation, (ii) stimulus eccentricity or (iii) surround suppression. We measured contrast-masking functions for target patches of grating in the fovea and in the periphery for cross-oriented superimposed and doughnut masks and parallel doughnut masks. When suppression was evident, the factor that determined whether it accelerated or saturated was whether the mask stimulus was crossed or parallel. There are at least two interpretations of the asymptotic behaviour of the parallel surround mask. (1) Suppression arises from pathways that saturate with (mask) contrast. (2) The target is processed by a mechanism that is subject to surround suppression at low target contrasts, but a less sensitive mechanism that is immune from surround suppression ‘breaks through’ at higher target contrasts. If the mask can be made less potent, then masking functions should shift downwards, and sideways for the two accounts, respectively. We manipulated the potency of the mask by varying the size of the hole in a parallel doughnut mask. The results provided strong evidence for the first account but not the second. On the view that response compression becomes more severe progressing up the visual pathway, our results suggest that superimposed cross-orientation suppression precedes orientation tuned surround suppression. These results also reveal a previously unrecognized similarity between surround suppression and crowding (Pelli, Palomares, & Majaj, 2004).

KW - human vision

KW - contrast gain control

KW - psychophysics

KW - surround suppression

KW - crowding

KW - XOM

KW - XOS

KW - detection

UR - http://www.scopus.com/inward/record.url?scp=70449529402&partnerID=8YFLogxK

U2 - 10.1016/j.visres.2009.09.006

DO - 10.1016/j.visres.2009.09.006

M3 - Article

VL - 49

SP - 2927

EP - 2935

JO - Vision Research

JF - Vision Research

SN - 0042-6989

IS - 24

ER -