**4. Alzheimer's disease can present with additional changes in visual cortex**

AD is characterized by progressive cognitive deficits including disturbances in memory, language, executive function, and vision [57, 84]. Somewhat surprisingly, it is not uncommon for visual deficits to be reported as one of the first symptoms of AD. However, despite many descriptions of visual symptoms in AD, only a very few studies have begun to examine the extent of changes in the organization, functionality, and connectivity of visual cortex that underlie these visual deficits.

#### **4.1. Patterns of neurodegeneration in the visual cortex of AD patients**

AD can present with a variety of visual symptoms across subjects, from lower-level deficits such as changes in visual acuity, contrast sensitivity, color discrimination, visual-spatial perception, and visual-processing speed [40, 73, 85, 132, 153, 154] to higher-level deficits such as problems in visual-spatial attention and in feature recognition of complex objects such as faces [40, 61, 63, 65, 69, 76, 78, 155, 156]. The neuropathology of AD results in gray matter lesions of varying density within regions of visual cortex [48, 57, 76, 84], and the visual symptoms could be attributed in part to a random pattern of neurodegeneration across regions of visual cortex [132]. However, there is also evidence for a more precise distribution of neurodegen‐ eration in the AD visual pathways [48, 84], with some studies showing neurofibrillary tangles and neuritic senile plaques increasing steadily from primary to associative visual cortex and degenerative changes in both the retina ganglion cells and optic nerves [58, 59, 64, 66–68, 70]. The role that such changes may play in the visual symptoms of AD is discussed in the following sections.

Although AD is primarily a disorder of cortical gray matter, some studies have also shown a decreased density of the connections through the splenium of the corpus callosum, the region of the major interhemispheric white matter pathway that connects left and right visual cortex [47, 79, 157]. The changes in the white matter tracts in dementia may result from Wallerian degeneration following retinal and cortical lesions or may be the product of a primary neuropathological process within the white matter itself. In the first case, we expect such white matter changes to reflect the behavioral deficits associated with the related gray matter or retinal lesion. In the latter, the direct loss of white matter connectivity may itself further contribute to these visual symptoms. Future studies tying functional MRI measurements of VFM organization and function together with diffusion tensor imaging (DTI) MR measure‐ ments of white matter tracts in the same subjects should help to clarify the presence and extent of each option in dementia with visual symptoms.

In addition, the broadening of pRF sizes in the aging foveae of these four early VFMs is also consistent with the other behavioral deficits discussed in the sections above. Loss of resolution through these increased foveal pRF sizes likely underlies the decrease in spatial contrast sensitivity (**Figure 11**). The increased pRF sizes of aging V1 and V3 may similarly play a role in temporal-contrast-sensitivity and motion-discrimination deficits (**Figure 11A** and **C**). With respect to issues with color discrimination [27, 29, 30, 35], we have measured significant differences in pRF sizes in V2 and hV4 out to 5° and 6°, respectively, as well as foveal changes in V1 from 0° to 3° (**Figure 11A, B,** and **D**). It is possible that these greater regions of expanded pRFs in aging subjects are associated with aging changes specific to a ventral visual color and form pathway involving V1, V2, and hV4. Finally, V2 and hV4 both showed increases in pRF sizes across larger foveal and parafoveal regions (**Figure 11B** and **D**). These larger pRF sizes could reflect deficits in the proper tuning of visual-spatial attention and less ability to attend

**4. Alzheimer's disease can present with additional changes in visual cortex**

AD is characterized by progressive cognitive deficits including disturbances in memory, language, executive function, and vision [57, 84]. Somewhat surprisingly, it is not uncommon for visual deficits to be reported as one of the first symptoms of AD. However, despite many descriptions of visual symptoms in AD, only a very few studies have begun to examine the extent of changes in the organization, functionality, and connectivity of visual cortex that

AD can present with a variety of visual symptoms across subjects, from lower-level deficits such as changes in visual acuity, contrast sensitivity, color discrimination, visual-spatial perception, and visual-processing speed [40, 73, 85, 132, 153, 154] to higher-level deficits such as problems in visual-spatial attention and in feature recognition of complex objects such as faces [40, 61, 63, 65, 69, 76, 78, 155, 156]. The neuropathology of AD results in gray matter lesions of varying density within regions of visual cortex [48, 57, 76, 84], and the visual symptoms could be attributed in part to a random pattern of neurodegeneration across regions of visual cortex [132]. However, there is also evidence for a more precise distribution of neurodegen‐ eration in the AD visual pathways [48, 84], with some studies showing neurofibrillary tangles and neuritic senile plaques increasing steadily from primary to associative visual cortex and degenerative changes in both the retina ganglion cells and optic nerves [58, 59, 64, 66–68, 70]. The role that such changes may play in the visual symptoms of AD is discussed in the following

Although AD is primarily a disorder of cortical gray matter, some studies have also shown a decreased density of the connections through the splenium of the corpus callosum, the region of the major interhemispheric white matter pathway that connects left and right visual cortex [47, 79, 157]. The changes in the white matter tracts in dementia may result from Wallerian

**4.1. Patterns of neurodegeneration in the visual cortex of AD patients**

across the entire visual field.

294 Update on Dementia

underlie these visual deficits.

sections.

#### **4.2. Visual deficits may arise in AD from both random and precise degenerative changes in cortex**

fMRI measurements of cortical gray matter in AD subjects point to a combination of patterns of neurodegeneration, with some specific changes within cortical representations like VFMs seen consistently across subjects (e.g., **Figures 8**–**11**) in addition to variable alterations in gross

**Figure 12. Visual field map measurements in mild Alzheimer's disease subjects. (A, B)** Examples of VFMs V1, V2, V3, and hV4 are shown for the left **(A)** and right **(B)** hemispheres of a single subject with mild AD (AD-S1). For clarity, the visual responses are only shown for the VFMs of interest—V1, V2, V3, and hV4—and only voxels with a powerful response at a coherence ≥0.20 are colored. Note the visibly smaller size of these VFMs in this subject compared to those shown for young and healthy aging subjects in **Figure 6**. While the polar angle gradients still contain the expected rep‐ resentations of contralateral visual space with orderly reversals between VFMs, the eccentricity measurements, drawn from the same fMRI scans using the moving bar stimulus, are more disorganized. For cortical surfaces, dark gray rep‐ resents sulci, and light gray represents gyri. "**\***" denotes the approximate location of the occipital pole. CaS: calcarine sulcus. **(C**, **D)** Images depict the full 3D cortical renderings, approximate anatomical orientations for each hemisphere, the color legends for the respective measurements, and the 1 cm scale bar. Note that all flattened hemispheres have been adjusted to the same scale; scale bar is duplicated for ease of comparison. **(E**, **F)** A second set of examples is shown for the left **(E)** and right **(F)** hemispheres from a second subject with mild AD (AD-S2). This AD subject displays more normal VFM sizes and foveal eccentricity representations, but also has visible changes in the peripheral eccen‐ tricity representations. Other details are as described in **Figure 6**. Data were collected from [37].

cortical organization (e.g., **Figure 12**). Measurements of both aspects of distributed neurode‐ generation in the visual pathways can be useful in the diagnosis of AD in a specific individual and for understanding the progression this disease across cortex generally.

To date, the only neuroimaging study of VFM changes in AD patients that we are aware of is our Brewer and Barton (2014) study [36, 37], which used fMRI and pRF modeling to measure VFMs in a small number of patients with mild-to-moderate AD. Our results did not demon‐ strate simply a worsening of the deficits we measured in healthy aging subjects; rather, we observed both the visual deficits we found in healthy aging and additional changes in extrastriate VFMs (V2, V3, and hV4) unique to our AD subjects. Differences among our measurements of the hemispheres of each AD subject were likely primarily due to individual variations in the pattern and progression of neurodegeneration in each subject, revealed by detailed individual-subject data analysis. In addition, there were consistent patterns of changes across the cortical hemispheres that also likely reflected more uniform effects of AD on the visual pathways.

These measurements both demonstrated the feasibility of examining VFM changes in patients with dementia—despite the potentially difficult requirements of maintaining fixation and visual-spatial attention for several consecutive minutes—and emphasized the need for such detailed analyses in individual subjects for these types of investigations. Cortical changes seen consistently across AD patients may underlie the visual symptoms seen early in the disease [67] and may prove to be a useful tool for early and accurate diagnosis of AD. We review here some of the basic trends of VFM changes in AD patients, and, as above for healthy aging subjects, we suggest how these cortical changes may relate to specific deficits in visual behavior.

#### *4.2.1. Declines in visual acuity and contrast sensitivity*

Psychophysical studies have observed decreases in both visual acuity and contrast sensitivity in AD subjects beyond that expected for age-matched controls [71, 75]. In particular, psycho‐ physical measures in AD patients showed a decrease in spatial contrast sensitivity for lower spatial frequencies than measured in the healthy aging population [40, 63]. Our measurements of significantly decreased BOLD coherence in regions of V1 and V2, with a marginally significant decrease in coherence in V3, could underlie these deficits (**Figure 8A–C**) [37]. Similarly, we found that the AD subjects significantly differed from healthy aging subjects in terms of total surface area of V3, with a general trend for decreased total surface area across V1–3 (**Figure 9**). While there were no significant differences in the surface area of V1 between healthy aging and AD subjects, the higher variability of these measurements highlights individual differences in neurodegenerative patterns in primary visual cortex as well as the need for additional studies of VFMs in a large number of AD subjects. On average, our AD subjects had no further decline in the surface-area-percent distribution of the foveal represen‐ tation from 0° to 3° than that seen in the healthy aging subjects compared to youth (**Fig‐ ure 10**). However, there was a trend for shifts from foveal to peripheral representations, which could point to a variable but important loss of central visual processing.

A striking feature of our AD measurements was the much-reduced total surface areas of V1– hV4 in one subject (AD-S1; **Figure 12A** and **B**). The shrunken VFMs additionally displayed very disorganized eccentricity representations with little foveal representation, likely due to an idiosyncratic pattern of neurodegeneration around the occipital pole. It is important to note that the polar angle representation, drawn from the same scan as the eccentricity representa‐ tion, remained normal; thus the disorganization seen in the eccentricity representations cannot be simply attributed to a problem with that particular scan, but rather likely reflects alterations in visual function in this individual. Such a dramatic change in early VFM sizes would be expected to result in at least a significant decline in visual acuity and likely reflects changes in multiple aspects of visual processing; even so, the clinical examination of visual function in this subject reported no issues.

#### *4.2.2. Deficiencies in color and form processing*

cortical organization (e.g., **Figure 12**). Measurements of both aspects of distributed neurode‐ generation in the visual pathways can be useful in the diagnosis of AD in a specific individual

To date, the only neuroimaging study of VFM changes in AD patients that we are aware of is our Brewer and Barton (2014) study [36, 37], which used fMRI and pRF modeling to measure VFMs in a small number of patients with mild-to-moderate AD. Our results did not demon‐ strate simply a worsening of the deficits we measured in healthy aging subjects; rather, we observed both the visual deficits we found in healthy aging and additional changes in extrastriate VFMs (V2, V3, and hV4) unique to our AD subjects. Differences among our measurements of the hemispheres of each AD subject were likely primarily due to individual variations in the pattern and progression of neurodegeneration in each subject, revealed by detailed individual-subject data analysis. In addition, there were consistent patterns of changes across the cortical hemispheres that also likely reflected more uniform effects of AD on the

These measurements both demonstrated the feasibility of examining VFM changes in patients with dementia—despite the potentially difficult requirements of maintaining fixation and visual-spatial attention for several consecutive minutes—and emphasized the need for such detailed analyses in individual subjects for these types of investigations. Cortical changes seen consistently across AD patients may underlie the visual symptoms seen early in the disease [67] and may prove to be a useful tool for early and accurate diagnosis of AD. We review here some of the basic trends of VFM changes in AD patients, and, as above for healthy aging subjects, we suggest how these cortical changes may relate to specific deficits in visual

Psychophysical studies have observed decreases in both visual acuity and contrast sensitivity in AD subjects beyond that expected for age-matched controls [71, 75]. In particular, psycho‐ physical measures in AD patients showed a decrease in spatial contrast sensitivity for lower spatial frequencies than measured in the healthy aging population [40, 63]. Our measurements of significantly decreased BOLD coherence in regions of V1 and V2, with a marginally significant decrease in coherence in V3, could underlie these deficits (**Figure 8A–C**) [37]. Similarly, we found that the AD subjects significantly differed from healthy aging subjects in terms of total surface area of V3, with a general trend for decreased total surface area across V1–3 (**Figure 9**). While there were no significant differences in the surface area of V1 between healthy aging and AD subjects, the higher variability of these measurements highlights individual differences in neurodegenerative patterns in primary visual cortex as well as the need for additional studies of VFMs in a large number of AD subjects. On average, our AD subjects had no further decline in the surface-area-percent distribution of the foveal represen‐ tation from 0° to 3° than that seen in the healthy aging subjects compared to youth (**Fig‐ ure 10**). However, there was a trend for shifts from foveal to peripheral representations, which

could point to a variable but important loss of central visual processing.

and for understanding the progression this disease across cortex generally.

visual pathways.

296 Update on Dementia

behavior.

*4.2.1. Declines in visual acuity and contrast sensitivity*

Our measurements of VFM changes in AD were also consistent with the deficiencies in color and form processing frequently described in AD [73, 85, 153]. As suggested by Chan et al. [153], the degeneration of excitatory neurons in AD with the relative sparing of inhibitory inter‐ neurons in V1 may result in the color vision disorders reported by a subset of AD patients. This process could also drive the significant decrease we observed in pRF sizes in more peripheral hV4, a key region for color vision processing (**Figure 11D**). Patients with idiosyn‐ cratic foveal loss—as seen in our AD subject with greatly reduced VFM surface areas—might also present with these deficiencies in color and form processing.

#### *4.2.3. Problems in the visual-spatial attentional network*

Finally, the commonly reported changes in visual-spatial attention in AD patients may be related to the coherence changes we observed in V1 and V2 (**Figure 8**) [40, 63]. The marginally significant decreases in pRF sizes in the periphery of V2 and V3 and the significant pRF size decreases in the periphery of hV4 may also be involved in the deficiencies in the visual-spatial attentional network in AD and could contribute to the shrinkage of useful visual field that is often be even worse in AD patients than in healthy aging (**Figure 11B–D**) [72]. These cortical changes again may reflect degenerative disease in the retina and optic nerves or variations in feedback from higher-order VFMs [37]; some studies have shown that these regions contain more lesions in mild-to-moderate AD than V1 [48, 132]. In addition, V1, V2, and V4 both have been shown to play major roles in the visual-spatial attentional network, as described above [125, 149–151]. Future studies will be needed to examine whether similar changes in VFMs can be measured in the higher-order visual-spatial attention regions of parietal and frontal cortex (e.g., [158–161]).

#### **4.3. These visual field map measurements may be able to improve the early diagnosis of specific types of dementia**

Investigations into the early diagnosis of AD include such a wide range of methods as biochemical markers, cognitive testing, and structural and functional neuroimaging [47, 162, 163]. The ability to identify changes in cortical structure or function very early in the devel‐ opment of AD would increase the efficacy of treatments that stop the progression of the neurodegeneration before a significant amount of cortex is lost [62]. Neuroimaging measure‐ ments of VFM changes in patients with mild AD may provide an avenue for such early diagnosis, as these measurements can reveal subtle and highly detailed cortical changes using non-invasive fMRI [37]. In addition, these measurements in individual subjects also provide the opportunity to follow neurodegenerative changes in specific individuals over the course of their dementia progression (e.g., [9]). Further research into VFM characteristics in AD should include not only a larger population of AD patients, but also should examine the potential onset of visual symptoms in patients with mild cognitive impairment (MCI), which may provide an even earlier diagnostic tool [37, 55, 154, 164].

Disagreement also persists regarding the categorization of neurodegenerative symptoms into specific types of dementia. We do not yet have a complete understanding of how the different types of dementia—e.g., AD, PCA, DLB, etc.—vary with respect to the start of their associated neurodegenerative changes. Criteria have been outlined to differentiate AD from other dementias, but there still remains significant overlap across the symptoms associated with each dementia (e.g., [56, 78, 85, 86, 165]). The ability to distinguish a patient's particular type of dementia at an early time point in the disease may be vital for the identifying the correct treatment. Such comprehensive measurements of alterations in VFM characteristics as discussed here may assist in this early identification and diagnosis, as the onset and severity of changes in visual cortex are expected to follow patterns specific to each particular dementia [37].
