Functional Imaging of Perceptual Completion in Artificial Scotomas

Susana Martinez-Conde, Ph.D.

Barrow Neurological Institute

Funded in June, 2005: $100000 for 4 years
LAY SUMMARY . ABSTRACT . HYPOTHESIS . SELECTED PUBLICATIONS .

LAY SUMMARY

back to top

Determining How the Brain “Fills-In” Visual Blind Spots

Using fMRI BOLD imaging in normal adults, the researchers will determine how our brains “fill-in” normal blind spots that occur in our vision.  This study will provide baseline information that can be used to explore how filling-in occurs following diseases that can affect vision, such as strokes and other brain traumas, as well as diseases that directly affect the eye’s retina, such as macular degeneration and diabetic retinopathy.

Everyone has visual blind spots, called scotomas, which are localized areas of blindness in the visual field.  The brain fills-in these blind spots with perceptual texturing so people are unaware of them. The number of blind spots can increase exponentially, however, following stroke or other brain traumas, and in retinal diseases such as macular degeneration that occurs primarily in older people, and diabetic retinopathy, a complication of diabetes.  The researchers hypothesize that filling-in occurs in the higher visual brain areas.  The investigators will test this hypothesis by using fMRI BOLD to conduct physiological measurements of filling-in, while simultaneously obtaining perceptual reports from study participants of whether or not they experience blind spots under experimental conditions.

Significance:  This project on how the brain fills-in blind spots will provide baseline information for future studies in patients with visual loss.  The approach should provide a method for determining the limits of brain plasticity and functional reorganization following brain trauma or resulting from eye diseases, and the findings could lead to development of new therapies to improve visual function in these patients.

ABSTRACT

back to top

Functional Imaging of Perceptual Completion in Artificial Scotomas

Some diseases of the visual system result in a paradoxical situation: patients are blind in several spots in their visual field ("scotomas"), yet they are unaware of this problem, due to a visual process called "perceptual completion", or "filling-in". Filling-in also takes place near the center of vision in normal healthy retinas, hiding the naturally occurring scotoma that occurs due to the optic nerve head (the only part of the retina that contains no photoreceptors). The visual system moreover acts to fill-in artificial scotomas. For instance, fields of gray embedded within visual dynamic noise disappear when the subject's gaze is fixated. These facts suggest that filling-in is a brain process of great importance to normal vision as well as in the clinic.

This proposal seeks to determine the brain level at which filling-in occurs in the visual hierarchy. We propose to study the fMRI correlates of filling-in during the presentation of over 1,000 small artificial scotomas embedded within dynamic noise. Whole brain scanning of BOLD signal will be conducted in human subjects with a 3T MRI scanner. Previous single-neuron data from the primate visual cortex suggest that filling-in is first achieved in visual areas V2 and V3, but not all studies agree. We hypothesize that filling-in correlated BOLD modulation will occur in areas V2 and higher, matching previous electrophysiological data from the awake primate. This will be the first study to investigate texture filling-in in the human brain, through imaging techniques. Moreover, this will be the first study to conduct simultaneous measurements of physiological and perceptual filling-in in any species.

 

HYPOTHESIS

back to top

Hypothesis:
We propose to study the fMRI correlates of perceptual completion (i.e. filling-in) during the presentation of artificial scotomas. Previous single-neuron data suggest that filling-in is first achieved in visual areas V2 and V3, but not all studies agree. We will investigate the whole-brain functional imaging correlates of perceptual filling-in of artificial scotomas in human subjects. This will be the first study to investigate texture filling-in in the human brain, with imaging techniques. Moreover, no studies to date have carried out physiological measurements of texture filling-in with simultaneous perceptual reports. We expect to find that filling-in of artificial scotomas will correlate with increased BOLD signal in higher visual areas, where filling-in is thought to take place (i.e. areas V2v/V2d, V3/V3a, V4v/VP).

Goals:
To determine the brain area(s) in which filling-in of artificial scotomas takes place. The results of this research may help develop better diagnostic tools for evaluating loss of vision in scotoma patients, as well as therapies to improve visual function following trauma, stroke or disease.

Methods:
The neural correlates of filling-in in the whole human brain have not been investigated to date. We will conduct whole-brain functional imaging while normal human subjects view dynamic noise in which multiple artificial scotomas have been inserted. Subjects will simultaneously indicate the perceptual state of the scotomas: they will press a button when the scotomas appear to be filling-in, and they will release the button when the scotomas appear to be intensifying. We will correlate the time course of the BOLD signal in all the different visual areas with the perceptual report of the subjects. The results of this study will determine the relative contributions of the different visual areas to texture filling-in in the normal brain.

SELECTED PUBLICATIONS

back to top

Troncoso X.G., Tse P.U., Macknik S.L., Caplovitz G.P., Hsieh P.J., Schlegel A.A., Otero-Millan J., and Martinez-Conde S.  BOLD activation varies parametrically with corner angle throughout human retinotopic cortex. Perception. 2007;36(6):808-20.

Tse P.U., Martinez-Conde S., Schlegel A.A., and Macknik S.L.  Visibility, visual awareness, and visual masking of simple unattended targets are confined to areas in the occipital cortex beyond human V1/V2. Proc Natl Acad Sci USA. 2005 Nov 22;102(47):17178-83.