<s>
In	O
neuroscience	O
,	O
the	O
visual	O
P200	B-Algorithm
or	O
P2	O
is	O
a	O
waveform	O
component	O
or	O
feature	O
of	O
the	O
event-related	B-Algorithm
potential	I-Algorithm
(	O
ERP	O
)	O
measured	O
at	O
the	O
human	O
scalp	O
.	O
</s>
<s>
The	O
P2	O
component	O
,	O
also	O
known	O
as	O
the	O
P200	B-Algorithm
,	O
is	O
so	O
named	O
because	O
it	O
is	O
a	O
positive	O
going	O
electrical	O
potential	O
that	O
peaks	O
at	O
about	O
200	O
milliseconds	O
(	O
varying	O
between	O
about	O
150	O
and	O
275	O
ms	O
)	O
after	O
the	O
onset	O
of	O
some	O
external	O
stimulus	O
.	O
</s>
<s>
The	O
amplitude	B-Application
of	O
the	O
peak	O
of	O
the	O
waveform	O
may	O
be	O
modulated	O
by	O
many	O
different	O
aspects	O
of	O
visual	O
stimuli	O
,	O
which	O
allow	O
it	O
to	O
be	O
used	O
for	O
studies	O
of	O
visual	O
cognition	O
and	O
disease	O
.	O
</s>
<s>
The	O
first	O
mentions	O
of	O
an	O
ERP	O
component	O
similar	O
to	O
that	O
of	O
the	O
modern	O
P2	O
were	O
characterized	O
in	O
studies	O
of	O
basic	O
visual	O
and	O
auditory	O
evoked	B-Algorithm
potentials	I-Algorithm
.	O
</s>
<s>
These	O
would	O
later	O
be	O
classified	O
as	O
components	O
of	O
the	O
visual	B-Algorithm
evoked	I-Algorithm
response	I-Algorithm
(	O
VER	O
)	O
,	O
part	O
of	O
which	O
includes	O
the	O
P2	O
.	O
</s>
<s>
The	O
P2	O
follows	O
the	O
visual	B-Algorithm
N1	I-Algorithm
(	O
or	O
auditory	O
N100	B-Algorithm
)	O
and	O
P1	B-Algorithm
waveforms	O
(	O
negativity	O
and	O
positivity	O
at	O
150	O
and	O
100ms	O
respectively	O
)	O
and	O
is	O
followed	O
by	O
the	O
N200	B-Algorithm
,	O
P3	B-Algorithm
,	O
and	O
N400	B-Algorithm
waveforms	O
.	O
</s>
<s>
Originally	O
,	O
the	O
P2	O
was	O
characterized	O
as	O
a	O
sub-component	O
of	O
a	O
complex	O
involving	O
the	O
N1	B-Algorithm
,	O
P1	B-Algorithm
and	O
P2	O
,	O
which	O
was	O
known	O
as	O
the	O
vertex	O
potential	O
and	O
which	O
was	O
classically	O
studied	O
as	O
a	O
unitary	O
phenomenon	O
.	O
</s>
<s>
In	O
particular	O
,	O
the	O
relationship	O
between	O
the	O
N1	B-Algorithm
and	O
P2	O
was	O
thought	O
to	O
be	O
important	O
.	O
</s>
<s>
The	O
difference	O
between	O
the	O
N1	B-Algorithm
and	O
P2	O
,	O
known	O
as	O
the	O
vertex	O
amplitude	B-Application
,	O
was	O
found	O
to	O
be	O
significantly	O
larger	O
for	O
target	O
than	O
non-target	O
stimuli	O
and	O
for	O
rapid	O
attention	O
switching	O
task	O
.	O
</s>
<s>
Further	O
studies	O
have	O
subsequently	O
examined	O
the	O
P2	O
separately	O
from	O
the	O
N1	B-Algorithm
and	O
have	O
found	O
that	O
the	O
amplitude	B-Application
of	O
the	O
P2	O
itself	O
is	O
larger	O
for	O
target	O
stimuli	O
that	O
are	O
less	O
frequent	O
.	O
</s>
<s>
This	O
is	O
similar	O
to	O
the	O
P3	B-Algorithm
,	O
though	O
the	O
P2	O
is	O
usually	O
seen	O
for	O
more	O
simple	O
features	O
than	O
the	O
P3	B-Algorithm
.	O
</s>
<s>
In	O
the	O
auditory	O
domain	O
,	O
there	O
is	O
evidence	O
of	O
enhanced	O
P2	O
amplitudes	B-Application
even	O
when	O
a	O
target	O
stimuli	O
is	O
not	O
embedded	O
in	O
a	O
series	O
of	O
identical	O
stimuli	O
.	O
</s>
<s>
In	O
these	O
instances	O
,	O
enhanced	O
P2	O
amplitudes	B-Application
have	O
been	O
associated	O
with	O
auditory	O
learning	O
and	O
repeated	O
stimulus	O
exposure	O
.	O
</s>
<s>
Enhanced	O
P2	O
amplitudes	B-Application
have	O
been	O
reported	O
in	O
musicians	O
with	O
extensive	O
listening	O
experience	O
as	O
well	O
as	O
laboratory	O
based	O
auditory	O
training	O
experiments	O
.	O
</s>
<s>
A	O
significant	O
finding	O
is	O
that	O
P2	O
amplitude	B-Application
changes	O
are	O
sometimes	O
seen	O
independent	O
of	O
N1	B-Algorithm
amplitude	B-Application
changes	O
,	O
again	O
suggesting	O
some	O
degree	O
of	O
independence	O
of	O
N1	B-Algorithm
,	O
and	O
P2	O
latencies	O
and	O
amplitudes	B-Application
appear	O
to	O
be	O
affected	O
by	O
old	O
age	O
.	O
</s>
<s>
Like	O
other	O
evoked-response	O
potentials	O
,	O
the	O
presence	O
of	O
the	O
P2	O
is	O
revealed	O
in	O
the	O
waveform	O
of	O
the	O
EEG	B-Application
recorded	O
by	O
time-locking	O
data	O
from	O
trials	O
to	O
the	O
onset	O
of	O
the	O
stimulus	O
,	O
in	O
appropriate	O
paradigms	O
.	O
</s>
<s>
The	O
component	O
is	O
evoked	O
as	O
part	O
of	O
the	O
normal	O
response	O
to	O
visual	O
stimuli	O
,	O
but	O
the	O
amplitude	B-Application
and	O
latency	O
(	O
delay	O
between	O
stimulus	O
and	O
response	O
)	O
may	O
be	O
affected	O
by	O
exogenous	O
factors	O
,	O
such	O
as	O
repeated	O
visual	O
stimuli	O
.	O
</s>
<s>
The	O
visual	O
P2	O
has	O
been	O
studied	O
in	O
the	O
context	O
of	O
visual	O
priming	O
paradigms	O
,	O
oddball	B-Algorithm
paradigms	I-Algorithm
(	O
where	O
the	O
amplitude	B-Application
is	O
enhanced	O
to	O
targets	O
)	O
,	O
and	O
studies	O
of	O
repetition	O
in	O
language	O
.	O
</s>
<s>
With	O
regard	O
to	O
the	O
auditory	O
P2	O
,	O
the	O
primary	O
paradigm	O
used	O
to	O
study	O
manipulations	O
of	O
this	O
type	O
of	O
sensory	O
information	O
is	O
the	O
auditory	O
oddball	B-Algorithm
task	I-Algorithm
.	O
</s>
<s>
In	O
general	O
,	O
increases	O
in	O
the	O
attentiveness	O
of	O
the	O
subject	O
lead	O
to	O
decreased	O
amplitude	B-Application
of	O
the	O
P2	O
.	O
</s>
<s>
More	O
probable	O
targets	O
also	O
lead	O
to	O
decreased	O
amplitude	B-Application
of	O
the	O
P2	O
,	O
which	O
is	O
sensitive	O
to	O
the	O
number	O
of	O
non-target	O
(	O
distracter	O
)	O
features	O
in	O
a	O
visual	O
search	O
.	O
</s>
<s>
The	O
amplitude	B-Application
of	O
the	O
P2	O
is	O
greater	O
when	O
the	O
visual	O
search	O
is	O
more	O
efficient	O
(	O
selective	O
attention	O
)	O
,	O
but	O
this	O
does	O
not	O
affect	O
the	O
latency	O
.	O
</s>
<s>
Other	O
characteristics	O
,	O
such	O
as	O
attention	O
,	O
repetition	O
and	O
probability	O
of	O
the	O
stimulus	O
also	O
impact	O
the	O
amplitude	B-Application
of	O
the	O
P2	O
.	O
</s>
<s>
Differences	O
in	O
P200	B-Algorithm
peak	O
amplitude	B-Application
suggest	O
that	O
anterior	O
and	O
posterior	O
distributional	O
differences	O
are	O
elicited	O
during	O
encoding	O
of	O
words	O
for	O
rote	O
and	O
elaborative	O
memory	O
tasks	O
.	O
</s>
<s>
While	O
encoding	O
the	O
words	O
across	O
both	O
memory	O
tasks	O
,	O
participants	O
who	O
subsequently	O
recalled	O
less	O
generated	O
larger	O
frontal	O
amplitudes	B-Application
and	O
smaller	O
parietal/occipital	O
amplitudes	B-Application
than	O
those	O
who	O
recalled	O
more	O
.	O
</s>
<s>
Also	O
,	O
researchers	O
have	O
found	O
that	O
the	O
P200	B-Algorithm
(	O
overlapping	O
with	O
the	O
P300	B-Algorithm
)	O
was	O
elicited	O
in	O
a	O
digit	O
span	O
task	O
when	O
participants	O
heard	O
the	O
reverse	O
order	O
of	O
a	O
digit	O
series	O
that	O
they	O
previously	O
heard	O
.	O
</s>
<s>
The	O
P2	O
amplitude	B-Application
was	O
bigger	O
for	O
words	O
that	O
had	O
been	O
seen	O
before	O
.	O
</s>
<s>
This	O
indicates	O
that	O
P2	O
amplitude	B-Application
is	O
modulated	O
by	O
aspects	O
of	O
recognition	O
and	O
that	O
there	O
is	O
a	O
hemispheric	O
difference	O
(	O
which	O
may	O
be	O
important	O
for	O
language	O
processing	O
,	O
see	O
below	O
)	O
.	O
</s>
<s>
Researchers	O
recorded	O
visual	O
evoked	B-Algorithm
potentials	I-Algorithm
in	O
response	O
to	O
non-stereoscopic	O
two-dimensional	O
and	O
three-dimensional	O
images	O
in	O
order	O
to	O
study	O
neurophysiological	O
correlates	O
of	O
depth	O
perception	O
.	O
</s>
<s>
In	O
this	O
study	O
,	O
P2	O
amplitude	B-Application
was	O
significantly	O
larger	O
in	O
the	O
condition	O
with	O
three-dimensional	O
convex	O
and	O
concave	O
images	O
,	O
than	O
in	O
condition	O
with	O
two-dimensional	O
images	O
.	O
</s>
<s>
Much	O
in	O
line	O
with	O
observations	O
of	O
traditional	O
visual	O
search	O
paradigms	O
,	O
the	O
application	O
of	O
P2	O
studies	O
to	O
language	O
research	O
has	O
shown	O
that	O
the	O
amplitude	B-Application
of	O
the	O
P2	O
is	O
sensitive	O
to	O
both	O
the	O
orthographic	O
combinability	O
and	O
phonological	O
consistency	O
(	O
neighborhood	O
sizes	O
for	O
similar	O
appearing	O
and	O
similar	O
sounding	O
words	O
)	O
in	O
the	O
reading	O
of	O
Chinese	O
phonograms	O
.	O
</s>
<s>
High	O
combinability	O
and	O
consistency	O
Chinese	O
characters	O
elicited	O
lower	O
P2	O
amplitudes	B-Application
than	O
low	O
combinability	O
and	O
low	O
consistency	O
characters	O
.	O
</s>
