Using fMRI to Study Adult Human Visual Cortex Reorganization Following Retinal Injury

Stelios Smirnakis, M.D., Ph.D.

Massachusetts General Hospital, Boston, MA

Grant Program:

David Mahoney Neuroimaging Program

Funded in:

June 2006, for 3 years

Funding Amount:


Lay Summary

Exploring Contradictions About the Nature and Extent of Cortical Reorganization After Retinal Injury

This study will use fMRI imaging to examine contradictory findings about the nature and extent of reorganization of the visual cortex following retinal injury.

Research has not demonstrated conclusive answers to several fundamental questions concerning brain plasticity that occurs following retinal injury.  Among the questions: How long is needed for substantial visual cortex reorganization?  Which visual areas reorganize? Is injury to the periphery less conducive to reorganization than injury to the fovea, which is located in the center of the retina, free of blood vessels, and containing a high concentration of color and light-sensitive cells)?  Is training necessary to promote plasticity?  What conditions enhance cortical capacity for reorganization?

These researchers hypothesize that: the adult visual cortex is capable of reorganizing after retinal injury; higher visual areas exhibit more extensive reorganization compared to earlier visual areas; foveal injuries result in more extensive reorganization than peripheral retinal injuries; and the degree of reorganization increases with rehabilitative training.  They will use fMRI and BOLD imaging to test these hypotheses by mapping the topography of early visual areas and monitoring how it changes in 32 adults with one of five retinal conditions: retinal detachment, retinal artery occlusion, macular degeneration, or ischemic optic neuropathy.

All study participants will have a dense visual field of diminished sight (“scotoma”), which is expected to remain stable throughout the study period. This stability will assure that observed changes in cortical topography are the result of cortical reorganization rather than changes occurring in the retina.

Significance:  This carefully controlled, systematic study is anticipated to address several areas of scientific controversy concerning cortical reorganization following retinal injury.  The findings are anticipated to help guide vision rehabilitation strategies.


Using fMRI to Study Adult Human Visual Cortex Reorganization Following Retinal Injury

The ability of human cortex to reorganize after injury has been demonstrated by fMRI, chiefly in the somatosensory, motor, and language systems. In the visual system, extensive reorganization has also been demonstrated by positron emission tomography (PET) in subjects blind from an early age whose primary visual cortex was activated by reading Braille. Little is known, however, about the ability of the visual cortex to reorganize in adult human subjects with lesions of the visual system. In 1999, Baseler et al. used fMRI to show that adult human extrastriate cortex reorganizes after losing its input from area V1. More recently, Baker et al. described two subjects with long standing macular degeneration involving the fovea whose early visual areas displayed a remarkable degree of reorganization. This contrasts with the results of Sunness and Yantis, who studied a patient with extrafoveal macular degeneration of relatively recent onset (two years) and found no cortical reorganization. Our own results in the macaque suggest that, without training, macaque primary visual cortex shows limited reorganization in the months following peripheral homonymous retinal lesions. Reports from other labs also suggest that, in the absence of specific training, cortical reorganization is limited.

Significant discordance exists among the findings of these reports to warrant a systematic approach in trying to understand the conditions that promote visual cortex reorganization after retinal lesions in adult human patients. Several unproven hypotheses have been suggested to explain the variability in the published reports: 1) reorganization may favor patients with foveal versus peripheral injuries, because foveal injuries force the subject to use their periphery, whereas peripheral injuries can be compensated by changing the pattern of eye movements, 2) reorganization may depend on whether the scotoma is homonymous or not, or 3) reorganization may depend strongly on rehabilitative training. Here I propose to initiate a systematic study of visual cortical plasticity in a small cohort of adult human patients with carefully selected retinal pathology, designed to resolve some of the existing controversy and to identify the conditions that maximize the cortical capacity for reorganization.



Diseases that afflict the brain, such as stroke, are associated with high morbidity for patients and incur a tremendous burden to individuals, their families, and society. Understanding brain repair processes is an important step in the effort to design treatments aimed at enhancing the ability of the nervous system to recover after injury. In attempting to achieve this, it is important to study in detail how the adult human brain adapts or reorganizes after injury. This proposal’s main aim is to define what conditions promote reorganization in adult human visual cortex after retinal injury. This will likely have implications for designing effective rehabilitative strategies, not only for patients with visual field scotomas, but potentially for stroke victims with a variety of other brain lesions.

It is hypothesized that adult human visual cortex may be capable of reorganizing after retinal injury. We propose to study the extent of this reorganization across visual areas in selected cohorts of human patients with retinal lesions. We hypothesize that higher visual areas will exhibit more extensive reorganization than early visual areas (such as primary visual cortex). We plan to test whether the degree of reorganization depends on the visual field location of the retinal lesion, since it has been hypothesized that foveal lesions may result in more extensive reorganization than peripheral lesions. Importantly, we will test whether the degree of reorganization increases with visual rehabilitative training, and with time after injury.

Functional Magnetic Resonance Imaging is ideal for monitoring cortical reorganization in vivo as it allows global coverage of the brain with high spatial resolution. We will use standard phase-encoding stimuli to obtain high resolution retinotopic maps of human visual areas. Cortical reorganization after retinal injury should manifest as a change in the topography of the retinotopic maps of cortical visual areas over time. We will focus on retinal lesions that occur in adulthood and produce dense visual field scotomas, with well defined, stable borders. Carefully selected homogeneous patient cohorts will be compared with age-matched healthy controls. For patients with monocular lesions retinotopic maps obtained through the intact eye will provide additional controls.

Selected Publications

Smirnakis S.M., Brewer A.A., Schmid M.C., Tolias A.S.,  Schuz A., Augath M., Inhoffen W., Wandell B.A., and Logothetis N.K.  Lack of long-term cortical reorganization after macaque retinal lesions.  Nature. 2005 May 19;435(7040):300-7 .

Tolias A.S., Keliris G., Smirnakis S.M., and Logothetis N.K.  Neurons in macaque area V4 acquire directional tuning after adaptation to motion stimuli.   Nat Neurosci. 2005 May;8(5):591-3 .

Tolias A.S., Smirnakis S.M., Augath M., Trinath T., and Logothetis N.   Motion processing in the macaque: revisited with functional magnetic resonance imaging.   J Neurosci. 2001 Nov 1;21(21):8594-601 .

Smirnakis S.M., Berry M.J., Warland D.K., Bialek W., and Meister M.  Adaptation of retinal processing to image contrast and spatial scale.   Nature. 1997 Mar 6;386(6620):69-73 .