Concept
Neuroplasticity
5 studiesEvidence last moved Sep 20, 2026
Neuroplasticity is the capacity of neural circuits to change their structure or function in response to experience, training or injury. In the human studies here it is inferred from changes in connectivity, corticospinal output and behaviour after stroke — not observed at the level of the connections themselves.
Plasticity is the mechanism rehabilitation is supposed to exploit, so how it is measured decides which therapies look effective. These trials are unusually informative because their honest secondary outcomes show how narrow the gains are, and because one of them identifies who is likely to respond before treatment starts.
Studies
5
Findings
4
4 supporting · 0 challenging · 0 qualifying citations
Open tensions
1
Latest change
Concept page published
Neuroplasticity
Currently
What we know
- The measurable change was in how strongly regions influenced each other, not in rewiring.
- The same stimulation without the pairing did nothing — the contingency was the active ingredient.
- A large share of the apparent variability in response was spatial targeting.
- The primary composite outcome did not move even though the targeted measure and connectivity did.
Largest unresolved question
The mechanistic account and the human evidence sit at levels that cannot confirm one another. Pericyte PDGFRβ signalling was necessary for astrogliosis, remyelination and functional recovery in mice after permanent occlusion, whereas the human trials measure connectivity, force and clinical scales and identify no cellular mechanism at all.
Common misconceptions
Enough practice rewires the adult brain around damage.
The changes recorded here are parameter shifts — reduced local inhibition, increased global coupling, altered hemispheric connectivity — and the behavioural gains they accompany are small and specific. The BCI trial's total Fugl-Meyer score did not favour the active arm.
Increased connectivity after therapy means the patient recovered.
Affected-hemisphere connectivity rose and correlated with wrist-level measures, while the trial's total motor score showed no advantage over the comparison arm. Connectivity is the proxy, not the outcome.
Brain stimulation either works for stroke patients or it does not.
Response to 1 Hz rTMS depended on where pretreatment activation peaked: 86% responded when the peak was within 25 mm of the target and 34% when it was further away. Averaged across unselected patients, the same treatment would look modestly effective.
Related
Claim ledger
What the evidence shows
Drawn from 5 studies in this library. Mix labels say which citation roles are present; they are not a strength score. Supports means evidence for a finding; Challenges means evidence against a stated position; Qualifies marks scope.
The measurable change was in how strongly regions influenced each other, not in rewiring.
Post-stroke reorganisation shows up as altered network parameters rather than new pathways. Modelling gave stroke patients reduced local neural inhibition and increased global coupling relative to controls, and specific pre-therapy parameters related to how much patients subsequently recovered.
The same stimulation without the pairing did nothing — the contingency was the active ingredient.
Timing is what separates an effective protocol from an ineffective one. Peripheral electrical stimulation delivered in time with movement-related cortical potentials increased ankle strength by 7.33 N (9.75% over baseline) and voluntary activation by 6.99 percentage points, while sham stimulation produced no such change.
A large share of the apparent variability in response was spatial targeting.
Who responds can be predicted before treatment. Patients whose pretreatment activation peaked within 25 mm of the stimulation target responded to 1 Hz rTMS at 86%, against 34% for those whose peak lay further away, with response defined as at least a 5-point motor-scale gain.
The primary composite outcome did not move even though the targeted measure and connectivity did.
Behavioural gains are narrower than the connectivity changes accompanying them. Motor-imagery-contingent BCI feedback improved wrist extension strength (adjusted mean difference 0.52, 95% CI 0.03–1.00) and raised affected-hemisphere connectivity correlating with that measure, but did not improve the total Fugl-Meyer score over the control arm.
Debates
Tensions and limits
Some items are genuine disagreements on the same question. Others mark different assays, populations, or outcomes.
The mechanistic account and the human evidence sit at levels that cannot confirm one another. Pericyte PDGFRβ signalling was necessary for astrogliosis, remyelination and functional recovery in mice after permanent occlusion, whereas the human trials measure connectivity, force and clinical scales and identify no cellular mechanism at all.
The mechanistic account and the human evidence sit at levels that cannot confirm one another. Pericyte PDGFRβ signalling was necessary for astrogliosis, remyelination and functional recovery in mice after permanent occlusion, whereas the human trials measure connectivity, force and clinical scales and identify no cellular mechanism at all.
PaperFren reads this as a limit on how far one study travels — different assays, populations, or outcomes — not a forced fight between papers.
Timeline
How understanding moved
Study years are when the paper was published. Evidence edits are dated changes to this page's claims. Explanations are when PaperFren added a Discovery — not a claim that the science happened that day.
2026
- Brain maps predict response to magnetic stroke therapy
Concept page published
Neuroplasticity
Change log
What changed
Dated edits to this page's evidence: studies added or removed from a claim, claims added or withdrawn, and new explanations tagged here. Rewordings are not listed.
- Concept page published
Papers
5 studies in this library bear on Neuroplasticity, ordered by citations.
- Do pericytes help the brain recover after a stroke?
In PDGFRβ-deficient mice after permanent MCAO, weaker pericyte-driven infarct repair tracked with less peri-infarct astrogliosis, poorer oligodendrogenesis/remyelination, and worse functional recovery.
- Modeling Stroke Recovery with Virtual Brains
By simulating individual brain activity, researchers found that chronic stroke changes how local and global brain regions interact, and these alterations can predict how well patients regain motor function.
- Does motor-imagery-contingent BCI feedback help chronic stroke?
In 25 people with chronic stroke, 4 weeks of BCI-FES with MI-contingent feedback improved wrist extensor strength more than matched MI-independent (sham-like) feedback.
- Can timed nerve stimulation boost ankle strength after stroke?
Syncing physical movements with targeted electrical pulses to the leg nerve significantly increases ankle strength and brain-to-muscle signal delivery in stroke survivors.
- Brain maps predict response to magnetic stroke therapy
Post-stroke patients whose brain activity peaks close to the stimulation target respond significantly better to magnetic brain stimulation.
Compare studies
Select 2–10 studies. Design and N are labels, not a ranking.
Nothing selected yet.
Questions
What is still open
The mechanistic account and the human evidence sit at levels that cannot confirm one another. Pericyte PDGFRβ signalling was necessary for astrogliosis, remyelination and functional recovery in mice after permanent occlusion, whereas the human trials measure connectivity, force and clinical scales and identify no cellular mechanism at all.
Ask PaperFren about Neuroplasticity
Study this conceptflashcards and short-answer questions
Why is a sham-stimulation arm indispensable in a paired-stimulation trial?
Because the theoretical claim is about contingency, not about stimulation. Pairing peripheral electrical stimulation with movement-related cortical potentials raised ankle strength by 7.33 N and voluntary activation by 6.99 percentage points; the inactive sham produced no comparable change. Without that arm, the same result would be equally consistent with stimulation having a non-specific effect, which would not support a plasticity interpretation.
A trial's connectivity measure improves but its primary clinical score does not. How should that be reported?
As a null on the primary outcome with a mechanistic signal alongside it. In the BCI trial, contingent feedback improved wrist extension (0.52, 95% CI 0.03–1.00) and raised affected-hemisphere connectivity correlating with that measure, while total Fugl-Meyer did not favour the active arm. Leading with the connectivity change would promote a secondary, mechanistic result over the outcome the trial was designed to test.
What does an 86% versus 34% response split tell you about published stimulation trials?
That unselected trials may be averaging over two different populations. When pretreatment activation peaked within 25 mm of the target, 86% of patients met the response threshold; further away, 34% did. A trial recruiting without that measurement would report an intermediate effect that describes neither group, which is one explanation for inconsistent stimulation results. The design here is an unmatched case-control, so it identifies a predictor rather than proving targeting causes response.