<s>
Prefrontal	B-Algorithm
cortex	I-Algorithm
basal	I-Algorithm
ganglia	I-Algorithm
working	I-Algorithm
memory	I-Algorithm
(	O
PBWM	B-Algorithm
)	O
is	O
an	O
algorithm	O
that	O
models	B-Application
working	O
memory	O
in	O
the	O
prefrontal	O
cortex	O
and	O
the	O
basal	O
ganglia	O
.	O
</s>
<s>
It	O
can	O
be	O
compared	O
to	O
long	B-Algorithm
short-term	I-Algorithm
memory	I-Algorithm
(	O
LSTM	B-Algorithm
)	O
in	O
functionality	O
,	O
but	O
is	O
more	O
biologically	O
explainable	O
.	O
</s>
<s>
It	O
is	O
used	O
as	O
part	O
of	O
the	O
Leabra	B-Algorithm
framework	O
and	O
was	O
implemented	O
in	O
Emergent	B-Algorithm
in	O
2019	O
.	O
</s>
<s>
Although	O
many	O
computational	O
models	B-Application
of	O
working	O
memory	O
have	O
been	O
developed	O
,	O
the	O
mechanistic	O
basis	O
of	O
executive	O
function	O
remains	O
elusive	O
.	O
</s>
<s>
PBWM	B-Algorithm
is	O
a	O
computational	O
model	O
of	O
the	O
prefrontal	O
cortex	O
to	O
control	O
both	O
itself	O
and	O
other	O
brain	O
areas	O
in	O
a	O
strategic	O
,	O
task-appropriate	O
manner	O
.	O
</s>
<s>
The	O
model	O
's	O
performance	O
compares	O
favorably	O
with	O
standard	O
backpropagation-based	O
temporal	O
learning	O
mechanisms	O
on	O
the	O
challenging	O
1-2-AX	O
working	O
memory	O
task	O
,	O
and	O
other	O
benchmark	O
working	O
memory	O
tasks	O
.	O
</s>
<s>
In	O
the	O
PBWM	B-Algorithm
model	O
in	O
Emergent	B-Algorithm
,	O
the	O
matrices	O
represent	O
the	O
striatum	O
.	O
</s>
<s>
SNrThal	O
represents	O
the	O
substantia	O
nigra	O
pars	O
reticulata	O
(	O
SNr	O
)	O
and	O
the	O
associated	O
area	O
of	O
the	O
thalamus	O
,	O
which	O
produce	O
a	O
competition	O
among	O
the	O
Go/NoGo	O
units	O
within	O
a	O
given	O
stripe	O
and	O
mediates	O
competition	O
using	O
k-winners-take-all	B-Algorithm
dynamics	O
.	O
</s>
<s>
This	O
layer	O
models	B-Application
midbrain	O
dopamine	O
neurons	O
.	O
</s>
