We’ll start this one off with a bit of a test. Have a look at the images below – can you guess what they are? You might find a judgment like this a bit difficult, and tasks like this reveal that visual perception is much more than simply registering an image. To do this accurately, we need to be able to recognize and interpret the information that we get, processes that rely on the brain.

While we can normally recognize objects effortlessly, the examples above might give you an appreciation of what it’s like to really struggle to identify something. Recognizing objects can be particularly challenging for people with degenerative visual disorders that affect the brain, as seen in certain forms of Alzheimer’s disease.

With a population that is living longer, the prevalence of Alzheimer’s disease is rapidly growing, affecting one in 13 people between the ages of 65 and 84 and one-third of people over age 85. Many of these individuals experience some form of visual impairment. This is particularly severe in posterior cortical atrophy (PCA), where Alzheimer’s attacks the visual system (a condition sometimes known as “visual variant Alzheimer’s”). Patients with PCA often have difficulty visually recognizing objects; they may struggle when it comes to “putting the pieces together” to perceive a single form, especially an object is presented under an unusual viewpoints or partially covered by another object. You can imagine how challenging this must be on a daily basis, as tasks like shopping or choosing food in a cafeteria would become quite difficult.

But how can we distinguish this type of impairment from common eye diseases? For example, how would you tell if somebody has glaucoma or cataract or another eye condition instead of PCA? In a recent paper in the Psychonomic Society journal Behavior Research Methods, authors Keir Yong, John Greenwood (pictured below), and colleagues set out to develop a test for posterior cortical atrophy.

Developing a good assessment is important, and a matter of some urgency, as patients may not realize that the problem is a neurological one. Imagine you start to have vision problems where objects become difficult to identify. Naturally, your first instinct would be to go to an optometrist, right? According to Yong,

“People with posterior cortical atrophy typically present initially to eye clinics, with multiple appointments, changes in glasses, or even eye surgery before being referred to neurology or dementia services. These diagnostic challenges often delay treatment for years.”

However, developing a good test to detect cortical visual loss is not so easy. There are many things to consider – first, it needs to correctly identify those who have that impairment. Second, it’s important to avoid falsely identifying individuals who have more common eye-related conditions, some of which are increasingly common as people age. In other words, the task should be difficult for those with detect cortical vision loss, but also doable by individuals with other forms visual impairments (e.g., cataracts) that are common in older populations.

With all of these considerations in mind, the authors developed a new digital test, known as the Graded Incomplete Letters Test (GILT) for cortical vision loss. The test works as follows: patients are shown a set of letters made up of small fragments, one at a time, like the letter shown below and asked to identify the letter out of the options underneath. As the participant completes the test, the letters become increasingly degraded.

The GILT builds on previous tests in important ways, by providing a granular assessment of participants’ performance in identifying degraded letters.  While prior tests have simply measured accuracy at a single level of degradation, including multiple levels would allow clinicians to catch more subtle impairments, and allow them to track patients’ visual function at a finer scale. For example, a person with a milder impairment may be able to still perform well when only 30% of the letter is intact, but may not do so well once it’s reduced to 20% visibility.

The authors tested the GILT on a large sample of adults in the UK from the UK Biobank, as well as a group of participants with a PCA diagnosis. The figure below shows the lowest level of completeness at which each of the groups could reliably identify the letters.

As shown above, the test could readily differentiate those with PCA from the remaining participants. Those with PCA struggled to identify the letters when a relatively small proportion of them was removed. Importantly, individuals with other forms of visual impairments, like glaucoma and cataracts, were still able to perform well on the test, particularly with the large letter sizes used. In other words, the test was sensitive enough to pick up on visual impairments that were seen in the PCA patients, without affecting the performance of those with more common eye conditions.

Together, these results offer a promising approach to detecting cortical vision loss early, before it becomes more severe. According to Yong,

“The GILT offers an opportunity for earlier detection of dementia-related visual loss in eye clinics to enable timely diagnosis and treatment by specialist services.”

As different forms of dementia become increasingly common, detecting any associated vision loss is crucial, and assessments like the GILT provide a rapid way to detect these impairments.

Psychonomic Society article featured in this post

Yong, K. X. X., Petzold, A., Foster, P., Young, A., Bell, S., Bai, Y., … & Greenwood, J. A. (2024). The Graded Incomplete Letters Test (GILT): A rapid test to detect cortical visual loss, with UK Biobank implementation. Behavior Research Methods, 1-13. https://doi.org/10.3758/s13428-024-02448-7