Reorganization of Visual Processing in Macular Degeneration Is Not Specific to the “Preferred Retinal Locus”

Daniel D. Dilks, Chris I. Baker, Eli Peli, Nancy Kanwisher

Abstract

Recent work has shown that foveal cortex, deprived of its normal bottom-up input as a result of macular degeneration (MD), begins responding to stimuli presented to a peripheral retinal location. However, these studies have only presented stimuli to the "preferred retinal location," or PRL, a spared part of the peripheral retina used by individuals with MD for fixating, face recognition, reading, and other visual tasks. Thus, previous research has not yet answered a question critical for understanding the mechanisms underlying this reorganization: Does formerly foveal cortex respond only to stimuli presented at the PRL, or does it also respond to other peripheral locations of similar eccentricity? If foveal cortex responds to stimuli at PRL because it is the long-term habitual use of this region as a functional fovea that drives the formerly foveal cortex to respond to stimuli presented at the PRL (the "use-dependent reorganization" hypothesis), then foveal cortex will not respond to stimuli presented at other locations. Alternatively, it may be that foveal cortex responds to any peripheral retinal input, independent of whether input at that retinal location has been chronically attended for months or years (the "use-independent reorganization" hypothesis). Using fMRI, we found clear activation of formerly foveal cortex to stimuli presented at either the PRL or an isoeccentric non-PRL location in two individuals with MD, supporting the use-independent reorganization hypothesis. This finding suggests that reorganization is driven by passive, not use-dependent mechanisms.

Introduction

When macular degeneration (MD) destroys central vision, the foveal cortex—the region of the primary visual cortex normally dedicated to high-resolution central input—is deprived of its primary sensory source. Previous studies established that this "deprived" cortex can reorganize to process information from the peripheral retina, specifically the Preferred Retinal Locus (PRL). The PRL is the specific peripheral spot that patients naturally adopt for tasks like reading or facial recognition.

A fundamental question in neuroscience is whether this reorganization is use-dependent (driven by the active, habitual use of the PRL as a "new fovea") or use-independent (a passive response where the cortex takes up any available peripheral signals). This study investigates these competing hypotheses by comparing brain responses to stimuli at the PRL versus an equally eccentric but non-habitual peripheral location.

Methods

The researchers conducted functional Magnetic Resonance Imaging (fMRI) on two individuals with long-term bilateral MD:

  • MD8: A 50-year-old female with Stargardt’s disease (26 years of vision loss).
  • MD1: A 57-year-old male with cone–rod dystrophy (20 years of vision loss). Two age-matched individuals with normal vision served as controls.

Microperimetry was used to map each patient's scotoma and identify their PRL. During fMRI, participants viewed grayscale object images at three locations: the PRL, an isoeccentric non-PRL location, and the former foveal location (the blind spot). The study focused on a "Region of Interest" (ROI) at the occipital pole, corresponding to the formerly foveal cortex.

Results

The study found robust evidence for the use-independent reorganization hypothesis. In both MD patients, the foveal cortex responded strongly to stimuli presented at both the PRL and the non-PRL location ($p < 0.01$).

Figure 3: Brain activation maps for PRL and non-PRL stimuli, showing similar patterns of reorganization in the foveal cortex. The subtraction map (PRL-nonPRL) shows no significant difference between the two conditions.

Crucially, there was no significant difference in activation magnitude between the habitual PRL and the unused non-PRL location ($p > 0.40$). This suggests that the foveal cortex's "rewiring" is not restricted to the part of the retina that has been actively trained.

Figure 2: Time-course of fMRI signal change in the foveal cortex. The overlapping high-amplitude curves represent responses to peripheral stimuli (PRL and non-PRL), while the lower curve represents the lack of response to foveal stimuli.

In contrast, healthy controls showed no activation in the foveal cortex for peripheral stimuli; instead, the signal often dropped below baseline, indicating active suppression of peripheral noise in the foveal region. Furthermore, stimuli presented to the patients' former foveal location (the scotoma) produced no response, confirming that the observed activity originated from peripheral reorganization rather than residual foveal function.

Discussion and Key Insights

The findings suggest that the adult human visual system possesses a degree of passive plasticity. Rather than requiring years of active training or "use," the foveal cortex appears to broaden its receptive fields to sample from a wider area of the peripheral visual field once central input is lost. This may occur through the "unmasking" of existing horizontal connections between neurons that are normally inhibited in a healthy brain.

Limitations

While the results are consistent across the two patients, the small sample size is a limitation. Additionally, while "habitual use" is not required, the role of immediate attention during the task remains a variable to explore. Future research should also address conflicting findings in other studies, which may be due to differences in disease progression, scotoma size, or experimental tasks.

Summary updated

Abstract

Recent work has shown that foveal cortex, deprived of its normal bottom-up input as a result of macular degeneration (MD), begins responding to stimuli presented to a peripheral retinal location. However, these studies have only presented stimuli to the "preferred retinal location," or PRL, a spared part of the peripheral retina used by individuals with MD for fixating, face recognition, reading, and other visual tasks. Thus, previous research has not yet answered a question critical for understanding the mechanisms underlying this reorganization: Does formerly foveal cortex respond only to stimuli presented at the PRL, or does it also respond to other peripheral locations of similar eccentricity? If foveal cortex responds to stimuli at PRL because it is the long-term habitual use of this region as a functional fovea that drives the formerly foveal cortex to respond to stimuli presented at the PRL (the "use-dependent reorganization" hypothesis), then foveal cortex will not respond to stimuli presented at other locations. Alternatively, it may be that foveal cortex responds to any peripheral retinal input, independent of whether input at that retinal location has been chronically attended for months or years (the "use-independent reorganization" hypothesis). Using fMRI, we found clear activation of formerly foveal cortex to stimuli presented at either the PRL or an isoeccentric non-PRL location in two individuals with MD, supporting the use-independent reorganization hypothesis. This finding suggests that reorganization is driven by passive, not use-dependent mechanisms.
BibTeX
@article{dilks2009reorganization,
  author = {Daniel D. Dilks and Chris I. Baker and Eli Peli and Nancy Kanwisher},
  title = {{Reorganization of Visual Processing in Macular Degeneration Is Not Specific to the “Preferred Retinal Locus”}},
  journal = {Journal of Neuroscience},
  year = {2009},
  doi = {10.1523/jneurosci.5258-08.2009},
}
Figure