Brain connectivity in
ageing and Alzheimer’s
Does Alzheimer’s disease change how brain regions communicate in the same way as getting older?
SBR Project ID
#BBHM01
Date of project
2017–2026
Location
Sweden and North America
Type of project
Observational Study
Categories
Lead institution and funders
As we grow older, our memory and thinking change. For some people this is a normal part of ageing; for others it is the first sign of Alzheimer’s disease, the most common cause of dementia. Inside the brain, the two often unfold at the same time, which makes them hard to tell apart.
One place to look is in how the brain’s regions work together. Even when we rest, activity in different regions rises and falls, and regions whose activity rises and falls in step are thought to be communicating with each other. Brain scans can measure this, and scientists call it functional connectivityMeasured with resting-state fMRI, which tracks changes in blood oxygen levels that follow brain activity. Strong connectivity between two regions means their activity tends to rise and fall together..
Both ageing and Alzheimer’s disease change functional connectivity. But earlier studies, which mostly looked at one region or network at a time, found a confusing mix: connectivity grew stronger in some places and weaker in others. It has been unclear which of these changes come from ageing and which from Alzheimer’s disease, and whether they help the brain cope or are signs of damage.
In this study, we stepped back to look at the whole brain at once. We used brain scans from 1,102 people in Sweden and North America to ask whether ageing and Alzheimer’s disease reshape functional connectivity in the same way. We also asked what this means for memory and thinking.
57 Million
People
worldwide were living with dementia in 2021. Alzheimer’s disease may contribute to 60–70% of cases, according to the World Health OrganizationRead the World Health Organization’s dementia fact sheet, updated 3 July 2026..
1,102
People
from two separate studies, in Sweden and North America, had their brain scans and Alzheimer’s protein levels analysed in this study.
1,000
Brain regions
of the brain’s outer layer were each compared with every other region to map their connectivity.
Follow this research
Get an email whenever we publish new findings on this project.
AIMS
What we want to know
Our study asked three questions.
For each of 1,000 brain regions, we measured its connectivity with the rest of the brain. Then we looked at how this changes with age, and with the levels of amyloid-betaA protein that clumps into plaques between brain cells in Alzheimer’s disease. and tauA protein that forms tangles inside brain cells in Alzheimer’s disease. Its spread through the brain is closely linked to memory and thinking problems., the two proteins that build up in Alzheimer’s disease.
We compared people of different ages and with different levels of amyloid-beta and tau. We also followed people who had brain scans on two or more visits, to see changes within the same person.
We compared people’s brain connectivity with their scores on memory and thinking tests, taking their age and protein levels into account. We did this separately for people without memory or thinking problems and for people with mild cognitive impairmentMild cognitive impairment: memory or thinking problems that can be measured on tests but are not severe enough to be called dementia. or Alzheimer’s dementia.
IMPORTANCE
Why it matters
If we can tell which changes in the brain come from Alzheimer’s disease rather than from ageing, treatments could be aimed at the disease itself.
Timing matters too. Signs of Alzheimer’s disease can now be found before memory problems begin: blood markers such as p-tau217A form of the tau protein that can be measured in a blood sample. Raised levels are an early sign of Alzheimer’s disease in the brain., for example, can reveal it early. But exactly how ageing and Alzheimer’s disease change the way the brain works is still debated.
Changes in brain connectivity could become a way to spot Alzheimer’s disease early, and to check whether treatments are working.
PROJECT
So we set up a study
We did not collect new data for this study. Instead, we analysed brain scans and test results from two long-running studies. BioFINDER-2The Swedish BioFINDER-2 study, run from Lund University. Opens the BioFINDER website. has followed about 2,960 volunteers in southern Sweden since 2017, from people with no memory problems to people with dementia. The Alzheimer’s Disease Neuroimaging Initiative (ADNI)A study that collects brain scans and other data at over 60 clinical sites in the USA and Canada. Opens the ADNI website. collects similar data in the USA and Canada. Nobody was given a treatment: this is an observational study.
Who took part
Group 1 · 973 people · Sweden
The main group
People taking part in BioFINDER-2. 655 had no measurable memory or thinking problems, though some had noticed changes themselves. 318 had mild cognitive impairment or Alzheimer’s dementia, and were included only if tests showed amyloid-beta build-up.
Group 2 · 378 people · Sweden
The follow-up group
BioFINDER-2 participants who had brain scans and tau scans on two or more visits, so we could follow changes within the same person over time.
Group 3 · 129 people · USA and Canada
The checking group
People taking part in ADNI: 89 without symptoms and 40 with mild cognitive impairment or dementia. We repeated the main analysis in this group to check the results.
Measuring the disease, not the diagnosis
Most studies sort people by diagnosis: no symptoms, mild cognitive impairmentMild cognitive impairment: memory or thinking problems that can be measured on tests but are not severe enough to be called dementia. or dementia. But a diagnosis is based on symptoms, and people differ in resilienceHow well a person keeps their memory and thinking skills despite disease in the brain.: some keep their memory and thinking skills for longer than others with the same amount of disease.
So instead, we used each person’s amyloid-beta and tau levels to place them on a scale from 0 (no build-up) to 1 (the most advanced build-up in the study). This let us follow the disease itself, step by step, and set it against each person’s age.
What we measured
Functional connectivity
Each person had a resting-state fMRIResting-state functional MRI: a brain scan taken while a person lies still and does no task, recording how activity changes in each part of the brain. scan. We divided the brain’s outer layer into 1,000 regions and measured how closely the activity of every pair of regions rose and fell together. This gave each region a connectivity profile: which regions it keeps in step with, and how strongly. We then scored how similar each region’s profile is to the profiles of all the others.
Alzheimer’s proteins
Amyloid-beta was measured in spinal fluidCerebrospinal fluid: the clear fluid around the brain and spinal cord, sampled with a needle in the lower back. and tau with PET scansPositron emission tomography: a scan that uses a small amount of a radioactive tracer to show where a substance, here tau protein, has built up in the brain., then combined into the protein score from 0 to 1.
Age
Each person’s age at the time of their scan, so the effects of age and of Alzheimer’s proteins could be measured side by side.
Memory and thinking
A combined score from standard memory and thinking tests, and separate scores for memory, executive functionThinking skills used to plan, switch between tasks and control behaviour., language and spatial skills from ten tests.
Other signs of disease
Thinning of the brain’s outer layer, signs of damage to small blood vessels on brain scans, and α-synucleinA protein linked to Parkinson’s disease and Lewy body dementia, measured here in spinal fluid., so we could check whether these explained the results.
How we made sense of the changes
Brain regions with similar jobs tend to have similar connectivity. Regions with similar connectivity sit close together on what scientists call gradients: spectrums that arrange every region of the brain by how similar its connectivity is to that of the others, with the most different regions at opposite ends. Two of these spectrums were central to our study.
The senses-to-thinking spectrum
At one end are the regions that handle vision, touch and movement. At the other are regions that combine information from many sources for complex thinking, such as memory and language. Scientists call this the sensory–association axis.
The meaning-to-control spectrum
At one end are parts of the default mode networkA set of brain regions that is most active when a person is not focused on the outside world., linked to the meaning of words and memories of our own lives. At the other are regions at the sides of the frontal and parietal lobes, used for planning, mental arithmetic and holding information in mind. Scientists call this the representational–executive axis.
For each region, we worked out whether its similarity score went up or down as people got older. That gave us a brain map for age, and we made a second map for the protein score. Then we laid each map over the two spectrums. If the scores of regions at one end of a spectrum went up while those at the other end went down, the change followed that spectrumImagine traffic changes across 1,000 streets in a city. You could list every street, or you could notice that the changes line up from the city centre to the suburbs: the centre got quieter and the suburbs busier. That tells you one city-wide process is at work, such as more people working from home. A different cause, like roadworks along a north–south route, would line up along a different line. Different lines point to different causes..
This tells us whether the changes are scattered or organised. If they follow a spectrum, one process is reshaping connectivity in a consistent way. And if ageing and Alzheimer’s disease follow different spectrums, their effects on the brain can be told apart.
From scan to finding
Steps 1 to 3 were done for each person’s scan. Steps 4 and 5 compare everyone in a group together.
RESULTS
What we have found so far
Here is what the brain scans of 1,102 people showed.
Open each question to see what we found.
Question 1Do ageing and Alzheimer’s disease change brain connectivity in the same way?
No. Ageing and Alzheimer’s disease each reorganise brain connectivity, but in two different patterns:
The ageing pattern
With age, regions used for planning and holding information in mind became more alike in their connectivity. Regions in the default mode network, linked to meaning and personal memories, became more distinct. The changes followed the meaning-to-control spectrum.
The Alzheimer’s pattern
As Alzheimer’s proteins built up, regions for vision, touch and movement became more distinct. Regions that combine information for memory, language and complex thinking became more alike. The changes followed the senses-to-thinking spectrum.
“More alike” means a region’s connectivity profile became more similar to those of the rest of the brain – more generic. “More distinct” means it became more specialised and unique. We can’t yet say whether these shifts harm thinking, help the brain cope, or both. We found the same two patterns in the 129 people from North America.
The patterns also fit how each process affects thinking. In the main group, older age was more strongly linked to lower scores for executive functionThinking skills used to plan, switch between tasks and control behaviour., and Alzheimer’s proteins to lower memory scores.
“Our analyses showed that brain communication reorganizes in one specific pattern during normal aging, and in a very distinct and different pattern as Alzheimer’s pathology accumulates.”
Question 2At what age, and how early in Alzheimer’s disease, do changes in brain connectivity appear?
Early. The Alzheimer’s pattern was clearest at early and middle levels of protein build-up, and it was already there in people with no memory or thinking problems. The ageing pattern was clearest between the ages of 55 and 70, and had essentially gone after 80.
“What surprised us was that the distinctive pattern of changes in brain communication was already apparent in people with low levels of Alzheimer’s pathology who were still cognitively unimpaired.”
Question 3Are changes in brain connectivity linked to memory and thinking?
Yes. Among people without symptoms or amyloid-beta build-up, lower scores on a combined test of memory and thinking went with changes along the meaning-to-control spectrum. But these ran in the opposite direction to ageing: regions used for planning and holding information in mind were more distinct, not more alike. In older people without symptoms, lower scores went with more of the Alzheimer’s pattern. Once memory and thinking problems had begun, lower scores went with more of the Alzheimer’s pattern, whatever the protein level or age.
What lower memory and thinking scores went with
No symptoms (average age 62)
Changes along the meaning-to-control spectrum, in the opposite direction to the ageing pattern
No symptoms, older people
More of the Alzheimer’s pattern
With memory or thinking problems
More of the Alzheimer’s pattern, whatever the protein level or age
UNEXPLORED
Unexplored questions
Like all good research, this study answered some questions and raised others. Explore the topics below to see what we still need to know.
Tap or click a pin to open each one.
FAQ
Frequently asked questions
Can a brain scan now tell whether I am developing Alzheimer’s disease?
No. These are patterns seen across groups of people, not a test for one person. Whether the patterns can be measured in one person, and whether they predict later decline, is still being studied.
How is this different from a blood test for Alzheimer’s disease?
Blood markers such as p-tau217 can already show signs of Alzheimer’s disease before memory or thinking problems appear. This study looks at something else: how brain connectivity changes as Alzheimer’s proteins build up, and how that differs from ordinary ageing.
If a brain region becomes “more alike”, does it stop working?
Not necessarily. “More alike” describes a region’s connectivity, not whether it works: its connectivity profile becomes more similar to those of the rest of the brain. We don’t yet know whether that change affects what the region does.
Could these changes be the brain’s way of coping?
Possibly. When we looked at whether the changes affected test scores, the Alzheimer’s-related changes tended, if anything, to help memory. But changes that seemed to help one skill tended to come at a cost to another, and these effects were small and uncertain. We think the brain may be under strain, with thinking drawing on limited resources, but we can’t yet say whether the changes are harmful, helpful or both.
Do memory problems in older age always mean Alzheimer’s disease?
No. Ageing on its own affects some thinking skills, especially executive functionThinking skills used to plan, switch between tasks and control behaviour., such as planning and keeping information in mind. In this study, ageing and Alzheimer’s proteins were linked to different patterns of change in brain connectivity.
TIMELINE
The long journey
ADNI begins
The first phase of the Alzheimer’s Disease Neuroimaging Initiative starts collecting brain scans and other data across the USA and Canada. Our checking group comes from this study.
BioFINDER-2 begins
The Swedish BioFINDER-2 study starts at Lund University. Over the following years it collects the brain scans, spinal-fluid samples and tau scans used for our main and follow-up groups.
Published in Nature Neuroscience
The findings are published in the journal Nature NeuroscienceRead the original research paper: Rittmo et al., Nature Neuroscience, published online 1 October 2026.. Work to test the patterns in individual people continues.
IMPACT
What’s happened
Here’s some of what we’ve published and contributed to. We’ll keep adding to this as the work develops.
Published in Nature Neuroscience
Analysis code openly available
FUTURE
Hopes for the future
Through this research, we hope to:
help doctors spot changes in brain connectivity caused by Alzheimer’s disease before memory problems appear
make it possible to track brain connectivity to see whether treatments for Alzheimer’s disease are working
find out whether, if these patterns turn out to be harmful, treatments such as non-invasive brain stimulation could adjust them
TEAM
Who are we?
Lead Investigators
Dr Jacob Vogel
Assistant Professor
Lund University, Sweden
Jonathan Rittmo
PhD Student
Lund University, Sweden
Co-investigators
Nicolai Franzmeier
LMU Munich, Germany
Munich Cluster for Systems Neurology (SyNergy), Germany
University of Gothenburg, Sweden
Olof Strandberg
Lund University, Sweden
Léa Chauveau
Lund University, Sweden
Theodore D. Satterthwaite
University of Pennsylvania, USA
Laura E. M. Wisse
Lund University, Sweden
Nicola Spotorno
Lund University, Sweden
Harry H. Behjat
Lund University, Sweden
Amir Dehsarvi
LMU Munich, Germany
Danielle van Westen
Lund University, Sweden
Skåne University Hospital, Sweden
Toomas Erik Anijärv
Lund University, Sweden
Susan M. Landau
University of California, Berkeley, USA
Sebastian Palmqvist
Lund University, Sweden
Skåne University Hospital, Sweden
Shorena Janelidze
Lund University, Sweden
Erik Stomrud
Lund University, Sweden
Skåne University Hospital, Sweden
Rik Ossenkoppele
Lund University, Sweden
Vrije Universiteit Amsterdam, the Netherlands
Amsterdam Neuroscience, the Netherlands
Niklas Mattsson-Carlgren
Lund University, Sweden
Skåne University Hospital, Sweden
Oskar Hansson
Lund University, Sweden
The study also used data from the Alzheimer’s Disease Neuroimaging Initiative, whose investigators are listed as an author group on the paper.
Sweden
Lund University
Lund, Sweden
Skåne University Hospital
Lund and Malmö, Sweden
University of Gothenburg
Gothenburg, Sweden
Germany
LMU Munich
Munich, Germany
Munich Cluster for Systems Neurology (SyNergy)
Munich, Germany
The Netherlands
Vrije Universiteit Amsterdam
Amsterdam, the Netherlands
Amsterdam Neuroscience
Research institute of Amsterdam UMC, Amsterdam, the Netherlands
USA and Canada
University of Pennsylvania
Philadelphia, USA
University of California, Berkeley
Berkeley, USA
Alzheimer’s Disease Neuroimaging Initiative (ADNI)
Data from clinical sites in the USA and Canada
Funders
Knut and Alice Wallenberg Foundation
SciLifeLab Wallenberg Data Driven Life Science Program (KAW 2020.0239)
Swedish Alzheimer Foundation
Alzheimerfonden, Sweden
Swedish Research Council
Vetenskapsrådet, Sweden
European Research Council
Funding for BioFINDER-2 (ADG-101096455)
Alzheimer’s Association
Funding for BioFINDER-2, USA
Cure Alzheimer’s Fund
Funding for BioFINDER-2, USA
Parkinson Foundation of Sweden
Funding for BioFINDER-2 (Parkinsonfonden)
ERA PerMed
Funding for BioFINDER-2 (ERAPERMED2021-184)
National Institute on Aging, National Institutes of Health
Funding for ADNI data collection, USA
CONTACT
Get involved
Questions about the project? Get in touch with the research team.
Research team
Lead investigator
Dr Jacob Vogel
Department of Clinical Sciences Malmö, Lund University
Media enquiries
Anna Hellgren
Stay connected
Jacob Vogel’s lab BioFINDER ADNI Lund UniversityFollow this research
Get an email whenever we publish new findings on this project.
Join the conversation
If a brain scan could show signs of Alzheimer’s disease before any memory problems began, would you want to know?
Should research money for dementia go first to finding the disease earlier, or to finding treatments for people who already have symptoms?
Do you think of memory changes in later life as a normal part of getting older, or as something to check with a doctor – and why?
