Skip to main content

Could 'famous faces test' be used to spot dementia?

Tuesday 13 August 2013

“Testing whether patients recognise Princess Diana and Elvis could help diagnose dementia," says The Daily Telegraph, with just one of the many misleading headlines based on new psychological research.

The research in question investigated a specific test that asks people to name and recognise images of famous faces of the twentieth century including Albert Einstein and Oprah Winfrey.

This test was given to 27 healthy adults and 30 people with a rare neurological condition called primary progressive aphasia (PPA) and comparisons were made between the groups. Primary progressive aphasia (PPA) causes problems with communication, especially spoken language, but other brain functions are normally unaffected. It is thought that PPA is one of the rarest types of dementia.

In this test 30 people with PPA were asked to name and recognise images of 20 famous people. Their results were compared to a control group of 27 healthy adults. As would be expected, the people with PPA had significantly more difficulties in naming and recognising the famous people’s faces.

The researchers then tested whether the findings from the test were associated with changes to the brain structure. Using MRI scans they found that people with PPA had more atrophy (wasting) in areas of the brain involved with visual perception and language.

Despite some early positive findings, this is a small study that only looked at test performance in people already diagnosed with one rare type of early dementia (PPA). The study has not examined whether this test could be used to accurately diagnose people in the initial diagnosis of PPA and certainly not the more common types of dementia seen in older age such as Alzheimer’s disease.

Where did the story come from?

The study was carried out by researchers from Northwestern University Feinberg School of Medicine in Chicago, US. It was funded by various US organisations, including the National Institute on Deafness and Other Communication Disorders, the National Institute on Aging, National Center for Research Resources and other institutions. The study was published in the peer-reviewed journal, Neurology.

Once past the misleading headlines, the study itself was reported reasonably accurately by the UK media.

However, despite media speculation, it is unproven whether this type of test would be either accurate or particularly useful in diagnosing PPA or other more common forms of dementia seen in older age, such as Alzheimer’s. 

What kind of research was this?

This was a cross-sectional study that used a famous faces test designed by Northwestern University (NUFFACE Test) to compare face naming and face recognition in people with a rare neurological disorder called primary progressive aphasia (PPA), compared with a group of healthy controls. The researchers then also looked at whether test performance was associated with the changes to brain structure typically seen in PPA.

This study compared test performance and brain changes in the group of people with the condition of interest (PPA) compared with people without the condition.

However, such a study cannot tell us whether this test could accurately be used in the identification and diagnosis of people who are in the early stage of development of PPA. It also can’t tell us whether the test could be used to diagnose other forms of dementia.

What did the research involve?

The researchers recruited 30 people with diagnosed primary progressive aphasia or PPA (cases). They also recruited a comparison group of 27 healthy people of similar age and education who did not have PPA (controls). Both groups were recruited from the Language in Primary Progressive Aphasia Research program in the US. The participants had an average age of 62 years. The researchers report that cases included people with different subtypes of PPA.

The participants underwent a series of assessments of their general cognition, language function and facial identification. They were then tested with the Northwestern University Famous Faces (NUFFACE) Test which involved showing 20 black and white printed images of famous faces downloaded from the internet. The images were selected by the researchers based on the following criteria:

  • popularity and celebrity status of the famous person in the visual media and press
  • race and sex
  • the period of time in which the person was famous (the NUFFACE Test is reported to use images relevant for individuals younger than 65 years)

All of the famous people selected were considered cultural icons including entertainers, politicians or internationally acknowledged leaders. The researchers say that although other face recognition tests exist, most have become outdated and are not appropriate to use in younger people affected by rare and specific forms of dementia such as PPA. The images of famous people included:

  • Albert Einstein
  • George W Bush
  • Elvis Presley
  • Princess Diana
  • Oprah Winfrey
  • Humphrey Bogart
  • Muhammad Ali
  • Barbara Streisand
  • Pope John Paul II

Before showing the images to the participants, the selected images were shown to a different group of 30 healthy people to ensure the images had the appropriate level of difficulty. Following this preliminary testing, no changes to selected faces were made by the researchers.

There were two parts to the NUFFACE Test – the first was the accuracy of naming the image (reporting the whole or part of the name – “Albert Einstein”) and the second was accuracy of recognition (the person could provide details about the person if they could not name them – for example in the case of Einstein, one answer was “I don’t know...scientist...E=MC2”).

Each image was shown to the participants and points were awarded dependent on how accurate they were. Statistical methods were then used to compare the results of the NUFFACE test between the cases and controls.

The researchers then performed brain scans (magnetic resonance imaging, MRI) on the 27 people with PPA (cases) and an additional 35 healthy volunteers recruited solely for this part of the research (controls). These results were used by the researchers to look at how abnormal changes in the brain structure were associated with the results of the NUFFACE Test. 

What were the basic results?

Unsurprisingly, participants with PPA (cases) performed significantly worse than healthy participants (controls) on both the naming and recognition parts of the NUFFACE Test:

  • the control group had an accuracy of 93.4% compared to 46.4% accuracy of cases on the naming part of the NUFFACE Test
  • the control group had an accuracy of 96.9% compared to 78.5% accuracy of cases in the recognition part of the NUFFACE Test

Participants with PPA were found to have widespread atrophy (wasting) identified on MRI. Face-naming difficulties were associated with the degree of atrophy in the anterior temporal lobe (involved with visual perception and language) on only the left side of the brain (the left side of the brain is the dominant side involved with language in most people). However, face recognition difficulties were associated with atrophy of the anterior temporal lobes on both sides of the brain.

How did the researchers interpret the results?

The researchers conclude that the research had three main findings.

The NUFFACE Test is a convenient tool for assessing face naming and recognition in people aged 40 to 65 years – a period during which early-onset dementias are often diagnosed.

People with primary progressive aphasia can have face identification difficulties that reflect naming or recognition impairments.

The findings ‘shed additional light’ on the changes to brain structure associated with face naming and face recognition.

The lead researcher, Tamar Gefen from Northwestern University’s Feinberg School of Medicine, is reported as saying: “it would be useful to add the test to the others that doctors use to spot early dementia.”

“In addition to its practical value in helping us identify people with early dementia, this test also may help us understand how the brain works to remember and retrieve its knowledge of words and objects.”


This study provides some preliminary findings of use of the NUFFACE Test in people with primary progressive aphasia – a rare and specific form of early dementia.

One of the main limitations of this study was that it was very small, including only 30 people with PPA. A small sample size decreases the reliability of the study findings; if another sample of people with PPA were examined, the results could be different.

Another limitation is that it only included people with PPA who were relatively young and in the early stages of disease. Findings may not be applicable to other people in later stages or severities of PPA who tend to have more wide ranging symptoms. Most importantly, the findings may not be applicable to people with other forms of early-onset dementia.

Further studies in larger populations of people with PPA and different types of early-onset dementia are required to draw further conclusions about use of the NUFFACE Test as a useful tool in clinical practice.

Certainly no assumptions should be made about the applicability of these findings to people with the more common types of dementia seen in older age, such as Alzheimer’s. 

Analysis by Bazian
Edited by NHS Website

Links to the headlines

Testing whether patients recognise Princess Diana and Elvis could help diagnose dementia

The Daily Telegraph, 12 August 2013

Test of famous faces 'helps to spot early dementia'

BBC News, 13 August 2013

Dementia: Failure to recognise famous faces could be a clue

Daily Express, 13 August 2013

Links to the science

Gefen T, Wieneke C, Martersteck A, et al.

Naming vs knowing faces in primary progressive aphasia - A tale of 2 hemispheres

Neurology. Published online August 13 2013