Human Brain Mapping,
Journal Year:
2020,
Volume and Issue:
42(4), P. 1070 - 1101
Published: Nov. 20, 2020
Abstract
Lesion
symptom
mapping
(LSM)
tools
are
used
on
brain
injury
data
to
identify
the
neural
structures
critical
for
a
given
behavior
or
symptom.
Univariate
lesion
(ULSM)
methods
provide
statistical
comparisons
of
behavioral
test
scores
in
patients
with
and
without
voxel
by
basis.
More
recently,
multivariate
(MLSM)
have
been
developed
that
consider
effects
all
lesioned
voxels
one
model
simultaneously.
In
current
study,
we
much‐needed
systematic
comparison
several
ULSM
MLSM
methods,
using
both
synthetic
real
potential
strengths
weaknesses
approaches.
We
tested
spatial
precision
each
LSM
method
single
dual
(network
type)
anatomical
target
simulations
across
location,
sample
size,
noise
level,
smoothing.
Additionally,
performed
false
positive
characteristics
associated
method's
spurious
findings.
Simulations
showed
no
clear
superiority
either
overall,
but
rather
highlighted
specific
advantages
different
methods.
No
produced
thresholded
map
exclusively
delineated
regions
behavior.
Thus,
indicated,
depending
particular
study
design,
hypotheses,
size.
Overall,
recommend
use
tandem
enhance
confidence
results:
Brain
foci
identified
as
significant
types
unlikely
be
can
confidently
reported
robust
results.
Biology,
Journal Year:
2025,
Volume and Issue:
14(4), P. 395 - 395
Published: April 10, 2025
The
Default
Mode
Network
has
been
extensively
studied
in
recent
decades
due
to
its
central
role
higher
cognitive
processes
and
relevance
for
understanding
mental
disorders.
This
neural
network,
characterized
by
synchronized
coherent
activity
at
rest,
is
intrinsically
linked
self-reflection,
exploration,
social
interaction,
emotional
processing.
Our
of
the
DMN
extends
beyond
humans
non-human
animals,
where
it
observed
various
species,
highlighting
evolutionary
basis
adaptive
significance
throughout
phylogenetic
history.
Additionally,
plays
a
crucial
brain
development
during
childhood
adolescence,
influencing
fundamental
processes.
literature
review
aims
provide
comprehensive
overview
DMN,
addressing
structural,
functional,
aspects,
as
well
impact
from
infancy
adulthood.
By
gaining
deeper
organization
function
we
can
advance
our
knowledge
mechanisms
that
underlie
cognition,
behavior,
health.
This,
turn,
lead
more
effective
therapeutic
strategies
range
neuropsychiatric
conditions.
Neuropsychologia,
Journal Year:
2018,
Volume and Issue:
115, P. 101 - 111
Published: March 18, 2018
This
study
investigated
how
sample
size
affects
the
reproducibility
of
findings
from
univariate
voxel-based
lesion-deficit
analyses
(e.g.,
lesion-symptom
mapping
and
morphometry).
Our
effect
interest
was
strength
between
brain
damage
speech
articulation
difficulties,
as
measured
in
terms
proportion
variance
explained.
First,
we
identified
a
region
by
searching
on
voxel-by-voxel
basis
for
areas
where
greater
lesion
load
associated
with
poorer
using
large
360
right-handed
English-speaking
stroke
survivors.
We
then
randomly
drew
thousands
bootstrap
samples
this
data
set
that
included
either
30,
60,
90,
120,
180,
or
patients.
For
each
resample,
recorded
estimates
p
values
after
conducting
exactly
same
analysis
within
previously
holding
all
procedures
constant.
The
results
show
(1)
often
small
sizes
heterogeneous
population
fail
to
be
detected;
(2)
its
statistical
significance
varies
size;
(3)
low-powered
studies
(due
sizes)
can
greatly
over-estimate
well
under-estimate
sizes;
(4)
(N
≥
90)
yield
highly
significant
even
when
are
so
they
become
trivial
practical
terms.
implications
these
interpreting
discussed.
Human Brain Mapping,
Journal Year:
2018,
Volume and Issue:
40(5), P. 1381 - 1390
Published: Dec. 13, 2018
Multivariate
lesion
behaviour
mapping
based
on
machine
learning
algorithms
has
recently
been
suggested
to
complement
the
methods
of
anatomo-behavioural
approaches
in
cognitive
neuroscience.
Several
studies
applied
and
validated
support
vector
regression-based
symptom
(SVR-LSM)
map
relations.
However,
this
promising
method,
as
well
multivariate
approach
per
se,
still
bears
many
open
questions.
By
using
large
samples
three
simulation
experiments,
present
study
empirically
tested
validity
several
methodological
aspects.
We
found
that
(i)
correction
for
multiple
comparisons
is
required
current
implementation
SVR-LSM,
(ii)
sample
sizes
at
least
100-120
subjects
are
optimally
model
voxel-wise
location
(iii)
SVR-LSM
susceptible
misplacement
statistical
topographies
along
brain's
vasculature
a
similar
extent
mass-univariate
analyses.
Human Brain Mapping,
Journal Year:
2020,
Volume and Issue:
42(4), P. 1070 - 1101
Published: Nov. 20, 2020
Abstract
Lesion
symptom
mapping
(LSM)
tools
are
used
on
brain
injury
data
to
identify
the
neural
structures
critical
for
a
given
behavior
or
symptom.
Univariate
lesion
(ULSM)
methods
provide
statistical
comparisons
of
behavioral
test
scores
in
patients
with
and
without
voxel
by
basis.
More
recently,
multivariate
(MLSM)
have
been
developed
that
consider
effects
all
lesioned
voxels
one
model
simultaneously.
In
current
study,
we
much‐needed
systematic
comparison
several
ULSM
MLSM
methods,
using
both
synthetic
real
potential
strengths
weaknesses
approaches.
We
tested
spatial
precision
each
LSM
method
single
dual
(network
type)
anatomical
target
simulations
across
location,
sample
size,
noise
level,
smoothing.
Additionally,
performed
false
positive
characteristics
associated
method's
spurious
findings.
Simulations
showed
no
clear
superiority
either
overall,
but
rather
highlighted
specific
advantages
different
methods.
No
produced
thresholded
map
exclusively
delineated
regions
behavior.
Thus,
indicated,
depending
particular
study
design,
hypotheses,
size.
Overall,
recommend
use
tandem
enhance
confidence
results:
Brain
foci
identified
as
significant
types
unlikely
be
can
confidently
reported
robust
results.
