Current Biology,
Journal Year:
2020,
Volume and Issue:
30(6), P. 988 - 1001.e4
Published: March 1, 2020
Animals
integrate
information
from
different
sensory
modalities,
body
parts,
and
time
points
to
inform
behavioral
choice,
but
the
relevant
comparisons
underlying
neural
circuits
are
still
largely
unknown.
We
use
grooming
behavior
of
Drosophila
melanogaster
as
a
model
investigate
that
govern
motor
sequence.
Flies
perform
movements
spontaneously,
when
covered
with
dust,
they
clean
their
bodies
following
an
anterior-to-posterior
After
investigating
modalities
could
detect
we
focus
on
mechanosensory
bristle
neurons,
whose
optogenetic
activation
induces
similar
Computational
modeling
predicts
higher
input
strength
head
will
cause
anterior
occur
first.
test
this
prediction
using
competition
assay
whereby
two
targeted
light
beams
independently
activate
neurons
parts.
find
initial
choice
movement
is
determined
by
ratio
inputs
In
dust-covered
flies,
change
result
successful
cleaning
movements.
Simulations
our
suggest
results
in
sequence
progression.
One
possibility
flies
frequent
between
posterior
inputs,
changing
ratios
drive
choices.
Alternatively,
may
track
temporal
given
part
measure
effectiveness.
The
first
hypothesis
supported
experiments:
iterative
spatial
parts
essential
for
organizing
PLoS Computational Biology,
Journal Year:
2021,
Volume and Issue:
17(8), P. e1009205 - e1009205
Published: Aug. 10, 2021
The
Drosophila
mushroom
body
exhibits
dopamine
dependent
synaptic
plasticity
that
underlies
the
acquisition
of
associative
memories.
Recordings
neurons
in
this
system
have
identified
signals
related
to
external
reinforcement
such
as
reward
and
punishment.
However,
other
factors
including
locomotion,
novelty,
expectation,
internal
state
also
recently
been
shown
modulate
neurons.
This
heterogeneity
is
at
odds
with
typical
modeling
approaches
which
these
are
assumed
encode
a
global,
scalar
error
signal.
How
coordinated
presence
heterogeneity?
We
develop
approach
infers
pattern
activity
sufficient
solve
defined
behavioral
tasks,
given
architectural
constraints
informed
by
knowledge
circuitry.
Model
exhibit
diverse
tuning
task
parameters
while
nonetheless
producing
coherent
learned
behaviors.
Notably,
prediction
emerges
mode
population
distributed
across
Our
results
provide
mechanistic
framework
accounts
for
during
learning
behavior.
Predatory
animals
pursue
prey
in
a
noisy
sensory
landscape,
deciding
when
to
continue
or
abandon
their
chase.
The
mosquito
Aedes
aegypti
is
micropredator
that
first
detects
humans
at
distance
through
cues
such
as
carbon
dioxide.
As
nears
its
target,
it
senses
more
proximal
body
heat
guide
meal
of
blood.
How
long
the
search
for
blood
continues
after
initial
detection
human
not
known.
Here,
we
show
5
s
optogenetic
pulse
fictive
dioxide
induced
persistent
behavioral
state
female
mosquitoes
lasted
than
10
min.
This
highly
specific
females
searching
and
was
recently
blood-fed
males,
who
do
feed
on
In
males
lack
gene
fruitless
,
which
controls
social
behaviors
other
insects,
long-lasting
behavior
response
resembling
predatory
females.
Finally,
triggered
by
enabled
engorge
mimic
offered
up
14
min
stimulus.
Our
results
demonstrate
internal
allows
integrate
multiple
over
timescales,
an
ability
key
success
apex
humans.
The Journal of Comparative Neurology,
Journal Year:
2019,
Volume and Issue:
528(6), P. 906 - 934
Published: Oct. 18, 2019
Abstract
The
central
complex
(CX)
comprises
a
group
of
midline
neuropils
in
the
insect
brain,
consisting
protocerebral
bridge
(PB),
upper
(CBU)
and
lower
division
(CBL)
body
pair
globular
noduli.
It
receives
prominent
input
from
visual
system
plays
major
role
spatial
orientation
animals.
Vertical
slices
horizontal
layers
CX
are
formed
by
columnar,
tangential,
pontine
neurons.
While
columnar
neurons
have
been
analyzed
detail,
especially
fruit
fly
desert
locust,
understanding
organization
tangential
cells
is
still
rudimentary.
As
basis
for
future
functional
studies,
we
studied
morphologies
locust
Schistocerca
gregaria
.
Intracellular
dye
injections
revealed
43
different
types
neuron,
8
PB,
5
CBL,
24
CBU,
2
noduli,
4
innervating
multiple
substructures.
Cell
bodies
these
were
located
11
clusters
cell
rind.
Judging
presence
fine
versus
beaded
terminals,
vast
majority
provide
into
CX,
lateral
(LX),
superior
protocerebrum,
posterior
slope,
other
surrounding
brain
areas,
but
not
directly
mushroom
bodies.
Connections
largely
subunit‐
partly
layer‐specific.
No
direct
connections
found
between
CBU
CBL.
Instead,
both
subdivisions
connected
parallel
with
PB
distinct
Current Biology,
Journal Year:
2020,
Volume and Issue:
30(6), P. 988 - 1001.e4
Published: March 1, 2020
Animals
integrate
information
from
different
sensory
modalities,
body
parts,
and
time
points
to
inform
behavioral
choice,
but
the
relevant
comparisons
underlying
neural
circuits
are
still
largely
unknown.
We
use
grooming
behavior
of
Drosophila
melanogaster
as
a
model
investigate
that
govern
motor
sequence.
Flies
perform
movements
spontaneously,
when
covered
with
dust,
they
clean
their
bodies
following
an
anterior-to-posterior
After
investigating
modalities
could
detect
we
focus
on
mechanosensory
bristle
neurons,
whose
optogenetic
activation
induces
similar
Computational
modeling
predicts
higher
input
strength
head
will
cause
anterior
occur
first.
test
this
prediction
using
competition
assay
whereby
two
targeted
light
beams
independently
activate
neurons
parts.
find
initial
choice
movement
is
determined
by
ratio
inputs
In
dust-covered
flies,
change
result
successful
cleaning
movements.
Simulations
our
suggest
results
in
sequence
progression.
One
possibility
flies
frequent
between
posterior
inputs,
changing
ratios
drive
choices.
Alternatively,
may
track
temporal
given
part
measure
effectiveness.
The
first
hypothesis
supported
experiments:
iterative
spatial
parts
essential
for
organizing