Two recent studies suggest that Long COVID may involve long-term neurobiological remodeling - even after mild infection.
One examined the brain under cognitive load.
The other looked at it at rest.
Together, they point to a persistent shift in network organization!🧵
One examined the brain under cognitive load.
The other looked at it at rest.
Together, they point to a persistent shift in network organization!🧵
https://twitter.com/zdenekvrozina/status/2023188837309227210
In the first study (Barnden et al.), the key issue was not where the brain activates -
but how its networks coordinate under mental exertion.
The largest differences appeared during cognitive load.
The regulatory switching system began to fail.
but how its networks coordinate under mental exertion.
The largest differences appeared during cognitive load.
The regulatory switching system began to fail.
The main systems involved were
the salience network - deciding what matters, and executive control circuits - sustaining performance.
And after repeated cognitive effort, the disruption became more pronounced.
That matches the lived experience
I can manage for a while - then it falls apart.
the salience network - deciding what matters, and executive control circuits - sustaining performance.
And after repeated cognitive effort, the disruption became more pronounced.
That matches the lived experience
I can manage for a while - then it falls apart.
Now we have another study.
Hedström et al., Scientific Reports 2026.
Patients after mild infection.
No hospitalization.
And on average 32 months post-infection (sic!)
This is no longer early recovery.
This is long-term.
Hedström et al., Scientific Reports 2026.
Patients after mild infection.
No hospitalization.
And on average 32 months post-infection (sic!)
This is no longer early recovery.
This is long-term.
This time, the researchers looked at the brain at rest.
And they found something striking.
Increased connectivity in the Default Mode Network (DMN).
The DMN is active during internal thought, self-reference, mental simulation -
the brain’s baseline mode.
And they found something striking.
Increased connectivity in the Default Mode Network (DMN).
The DMN is active during internal thought, self-reference, mental simulation -
the brain’s baseline mode.
Patients showed higher intra-network DMN connectivity than controls.
Nearly 3 years after infection.
No structural abnormalities on conventional MRI.
So this is not visible damage.
It is altered network organization.
Nearly 3 years after infection.
No structural abnormalities on conventional MRI.
So this is not visible damage.
It is altered network organization.
Interestingly, after a cognitive task, the group difference disappeared.
That could reflect
altered anticipation of task demands
disrupted switching between rest and task modes
or changes in returning to baseline
The authors remain cautious. But the pattern is there.
That could reflect
altered anticipation of task demands
disrupted switching between rest and task modes
or changes in returning to baseline
The authors remain cautious. But the pattern is there.
Barnden showed that under cognitive load, regulatory switching breaks down.
Hedström shows that the resting baseline itself is altered - long term.
These are not contradictions.
They are two sides of the same picture.
Hedström shows that the resting baseline itself is altered - long term.
These are not contradictions.
They are two sides of the same picture.
The simplest integrated model?
The baseline state is shifted.
And when regulatory demand increases, the switching system cannot compensate.
In other words - the brain starts from a different resting configuration,
and the transition between rest and effort becomes fragile.
The baseline state is shifted.
And when regulatory demand increases, the switching system cannot compensate.
In other words - the brain starts from a different resting configuration,
and the transition between rest and effort becomes fragile.
What makes this particularly important?
These findings are present nearly three years after a mild infection!
That suggests persistent network remodeling.
Not a transient after-effect.
These findings are present nearly three years after a mild infection!
That suggests persistent network remodeling.
Not a transient after-effect.
We are not seeing massive structural destruction.
We are seeing -
reorganization of networks
altered switching dynamics
increased energetic costs
reduced regulatory resilience under load
Even in people who were never hospitalized.
We are seeing -
reorganization of networks
altered switching dynamics
increased energetic costs
reduced regulatory resilience under load
Even in people who were never hospitalized.
This also fits a broader framework seen in other chronic (!) post-infectious and inflammatory conditions, where regulatory brain networks gradually reorganize under prolonged physiological stress.
Subtle.
But persistent.
Subtle.
But persistent.
What is most concerning is not simply the presence of network differences -
but their persistence.
Nearly three years after a mild infection, the brain’s resting architecture remains measurably altered.
That shifts the conversation.
This is not just about recovery time.
It is about long-term neurobiological remodeling.
@szupraha @ZdravkoOnline @adamvojtech86 @adamkova_vera
but their persistence.
Nearly three years after a mild infection, the brain’s resting architecture remains measurably altered.
That shifts the conversation.
This is not just about recovery time.
It is about long-term neurobiological remodeling.
@szupraha @ZdravkoOnline @adamvojtech86 @adamkova_vera
Hedström et al., Brain connectivity and its relation to cognitive function in patients with post-COVID-19 condition after mild infection, Scientific Reports 2026. nature.com/articles/s4159…
If these network changes persist nearly three years after mild infection, the key question becomes - what maintains the shift away from baseline?
Across Long COVID biology, proposed drivers include persistent viral antigens or proteins and chronic immune dysregulation.
Across Long COVID biology, proposed drivers include persistent viral antigens or proteins and chronic immune dysregulation.
Ongoing low-grade neuroimmune signaling can alter synaptic efficiency, network synchronization, and energetic balance - especially in regulatory systems like the salience and executive networks.
That makes long-term remodeling biologically plausible.
That makes long-term remodeling biologically plausible.
Similar network-level cognitive patterns are seen in other chronic infections, including HIV-associated neurocognitive syndrom. HAND.
This does not mean the diseases are the same - but it suggests a shared vulnerability of brain control networks under sustained inflammatory stress.
This does not mean the diseases are the same - but it suggests a shared vulnerability of brain control networks under sustained inflammatory stress.
• • •
Missing some Tweet in this thread? You can try to
force a refresh
