People with a mental eye deficiency known as aphantasia tend to have trouble recognizing faces. Recent research published in journals cortex These people provide evidence that they have specific difficulties in recognizing and remembering faces, but also in visually matching them. The findings help explain the underlying mechanism of the facial blindness that many people with aphantasia experience.
Aphantasia is a psychological condition in which the ability to form visual images in the mind is absent or severely reduced. When people with typical visual imagery think of an apple, they can think of its shape and color. People with aphantasia are unable to create this mental image.
This lack of visual imagery suggests possible effects on other cognitive functions. Previous studies have shown that many patients with aphantasia self-report having difficulty recognizing faces. Some people meet the criteria for prosopagnosia. Prosopagnosia is a neurological condition commonly known as facial blindness.
Recognizing faces is not a single, simple action. It relies on a combination of different cognitive processes working together. Analyzing these individual steps can help scientists understand where cognitive impairment occurs.
Mirta Stantic, a lecturer at Royal Holloway, University of London and co-author of the study, explained the motivation behind the study. “We’ve been looking at the profile of facial recognition abilities in different populations for years, and we’re already looking at populations of people who are face-blind, autistic, or who are very good at remembering faces,” Stantić said. “Here we were interested in understanding how aphantasia, the inability to form mental images, affects different stages of facial recognition.”
The authors wanted to isolate certain subprocesses. “We wanted to understand whether some steps of this complex mental process are influenced more than others, and whether this varies from person to person,” Stantic said.
Researchers recruited 34 adults with congenital aphantasia, a condition they were born with. We also recruited a control group of 41 adults with typical visual imagery abilities. Participants were closely matched in age, with an average age of approximately 40–42 years.
To ascertain group ranking, participants completed a questionnaire regarding the vividness of their visual images. This assessment asks people to rate the vividness of the mental images they are able to produce. Those with very low scores were placed in the aphantasia group, and those with high scores were placed in the typical imagery group.
To test face matching and face recognition, participants completed the Oxford Face Matching Test. During this task, participants saw 200 pairs of faces displayed on a computer screen for just over a second and a half each. Half of the pairs were shown two different pictures of the same person, and the other pairs were shown pictures of two different people.
Participants first had to decide whether the two faces belonged to the same person or to different people. They then rated how perceptually similar the two faces were on a scale of 0 to 100. These assessments helped separate basic decision-making from visual perception.
To purely measure facial perception, the scientists used a new analytical strategy. They compared human participants’ similarity ratings to similarity scores generated by three advanced facial recognition computer algorithms. If a participant’s rating deviates significantly from the computer algorithm’s objective rating, it indicates poor facial recognition ability.
Next, the scientists tested facial memory using the Cambridge Facial Memory Test. In this assessment, participants were able to recognize six new faces. They then had to identify these target faces among distracting factors over 72 trials. As the images changed, the tests became progressively more difficult.
The memory test is divided into three stages. The first stage displays the face with the same lighting and perspective. Later stages introduce dramatic changes in facial illumination and orientation, forcing the brain to rely on deep memories of facial structure rather than simple snapshots.
The researchers also asked participants to fill out a self-report survey about their facial recognition abilities in everyday life. This included a 20-item survey asking participants to rate statements about their struggles to recognize classmates and friends. Finally, to ensure that underlying visual acuity differences did not influence the results, we conducted an acuity task in which participants read lines of reduced letters.
The data showed that people with aphantasia reported much higher levels of difficulty recognizing faces in daily life. Approximately 26 percent of the aphantasia group scored above the strict criteria for facial blindness on a self-report survey. None of the common imagers met this threshold.
Regarding visual acuity tasks, there was no difference in basic visual acuity between the two groups. In addition, participants completed a survey designed to measure autistic traits in adults. The researchers found no statistically significant differences in autistic traits between the aphantasia and control groups, although the aphantasia group did tend to have slightly higher autistic traits.
The aphantasia group showed significant impairment in face perception. Similarity ratings in the paired face task deviated significantly from objective computer algorithm scores. This suggests that their basic perceptual processing of complex facial features is not very accurate.
This finding was somewhat unexpected based on previous literature. “Although some previous studies suggested that vision is largely intact in aphantasia, we found that participants with aphantasia had lower accuracy on face perception tasks,” Stantić told Cypost. She suggested that this may be because faces are highly complex visual stimuli.
“Furthermore, the task was designed to be challenging, requiring participants to make decisions in a relatively short amount of time,” she added.
When it came to basic face matching, the aphantasia group performed exactly like a typical imager. Both groups were equally accurate in determining whether two side-by-side photos showed the same person.
“Perhaps one would expect, given their difficulties with face perception and face memory, that people with aphantasia would have trouble performing face-matching tasks accurately,” Stantić says. “But our results suggest otherwise.”
“Interestingly, given their perceptual difficulties, their ability to distinguish between two photographs (one with the same person or one with a different person) remains intact,” she noted.
To separate these intertwined skills, the researchers used a statistical regression model. They were able to isolate memory performance by mathematically removing the influence of a person’s matching and perceptual scores. This allowed them to confirm that memory for faces made a truly independent contribution.
The aphantasia group had significantly lower scores on memory tests. Even after subtracting perceptual and matching abilities, aphantasia patients still had worse memory for faces than controls.
“Aphantasia doesn’t seem to cause a general inability to process faces, but as you might expect, people with aphantasia have more difficulty processing faces,” Stantić says.
“More broadly, this highlights the importance of understanding that facial recognition is not a special skill,” she added. “People may struggle with (or excel at) different aspects of this ability.”
One potential misinterpretation of this study is the assumption that all people with aphantasia suffer from severe clinical facial blindness. The data show that the majority of participants with aphantasia did not score low enough on objective memory tests to be formally classified as having prosopagnosia. Compared to typical imagers, their performance was simply lower on average.
“Sometimes people think that people with aphantasia have the same difficulties as people with prosopagnosia (often referred to as ‘face inability’), but this is not the case,” Stantić says. “There are clear differences between these two populations and they face unique challenges.”
“Our data support this: Only about 10 percent of this sample with aphantasia scored clinically low on standard tests for prosopagnosia, and these samples are clearly differentiated,” Stantic explained. “So the important thing to remember is that there are subtle but crucial differences between the experiences of people with aphantasia and prosopagnosia.”
This research also raises questions about how exactly visual images relate to perception. The authors point out that mentally constructing a 3D representation of a face from a 2D photo may require some degree of mental imagery. It remains unclear whether visual imagery is strictly required for face memory or whether the two functions simply rely on shared brain networks.
Future research may investigate whether different subtypes of aphantasia exist. The researchers also hope to expand the study to other age groups. “In the future, we hope to use this knowledge to understand how these abilities develop or don’t develop, particularly in typically and atypically developing children,” Stantić said.
Using a new computer algorithm approach, scientists can now more accurately measure facial processing skills. “This allows us to be more targeted and precise in our attempts to help people who have difficulty remembering faces,” she added. “In the long term, we want to develop programs that help train people with challenges.”
Collecting personal stories is an important part of the next step. “We are very keen to hear from people who have difficulty recognizing and remembering faces in everyday life, and from parents who think their children may be struggling with this,” Stantić said.
“First-hand accounts can help us understand a variety of experiences that may be missed by standard clinical examinations,” she explained. “These stories not only directly inform the questions we are investigating, but also allow us to provide targeted interventions.”
“If you would like to get in touch with our research or potentially participate (you can visit or test remotely!), please contact us at mirtastantic.com.”
The study, “Patients with aphantasia show independent impairments in face perception and face memory, but not in face matching,” was authored by Zoe Pounder, Mirta Stantic, Reshan Rieder, Caroline Catmoor, and Jeffrey Bird.
