Systematically underestimating one’s own cognitive decline may be a sign of Alzheimer’s

By Federica Cacciamani, Luisa Sambati, Marion Houot, Marie-Odile Habert, Bruno Dubois, Stéphane Epelbaum, on behalf of the INSIGHT-PreAD study group

Keywords: Alzheimer's disease Amyloid Anosognosia

With several drugs currently being tested, a very early diagnosis of Alzheimer’s disease becomes even more important. But how to do that?

For years it has been thought that older adults who perceive a decline and complain about their memory should be monitored because they may be developing Alzheimer’s dementia. But in fact, most older adults complain about their memory, as memory changes are normal with aging.

We have bucked the trend by using an advanced method of data analysis and discovering that:

  • Those who have complained about their memory (or language, attention…) for years, while their family/close friends do not notice any changes, are not at greater risk of progressing to Alzheimer’s dementia. The purely subjective and lasting (over years) perception of memory loss could be linked to depression, anxiety, or other conditions.
  • Conversely, the situation in which family/friends notice a decline, albeit slight, that the individual systematically underestimates should orient the clinician towards suspected Alzheimer’s disease
Published in Alzheimer's Research & Therapy the 14/10/2020

Full title Awareness of cognitive decline trajectories in asymptomatic individuals at risk for AD

Article abstract:


Lack of awareness of cognitive decline (ACD) is common in late-stage Alzheimer’s disease (AD). Recent studies showed that ACD can also be reduced in the early stages.


We described different trends of evolution of ACD over 3 years in a cohort of memory-complainers and their association to amyloid burden and brain metabolism. We studied the impact of ACD at baseline on cognitive scores’ evolution and the association between longitudinal changes in ACD and in cognitive score.


76.8% of subjects constantly had an accurate ACD (reference class). 18.95% showed a steadily heightened ACD and were comparable to those with accurate ACD in terms of demographic characteristics and AD biomarkers. 4.25% constantly showed low ACD, had significantly higher amyloid burden than the reference class, and were mostly men. We found no overall effect of baseline ACD on cognitive scores’ evolution and no association between longitudinal changes in ACD and in cognitive scores.


ACD begins to decrease during the preclinical phase in a group of individuals, who are of great interest and need to be further characterized.