BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//talks.cam.ac.uk//v3//EN
BEGIN:VTIMEZONE
TZID:Europe/London
BEGIN:DAYLIGHT
TZOFFSETFROM:+0000
TZOFFSETTO:+0100
TZNAME:BST
DTSTART:19700329T010000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=-1SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:+0100
TZOFFSETTO:+0000
TZNAME:GMT
DTSTART:19701025T020000
RRULE:FREQ=YEARLY;BYMONTH=10;BYDAY=-1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
CATEGORIES:Computational Neuroscience
SUMMARY:Modeling a non-linear EPSPs integration site in de
ndrites and its impact on computational capacities
- Romain Cazé\, Group for Neural Theory\, Ecole N
ormal Superieure\, Paris
DTSTART;TZID=Europe/London:20111124T160000
DTEND;TZID=Europe/London:20111124T170000
UID:TALK34777AThttp://talks.cam.ac.uk
URL:http://talks.cam.ac.uk/talk/index/34777
DESCRIPTION:Integration of excitatory post synaptic potentials
(EPSPs) in dendrites can be locally non-linear. I
n a dendritic branch\, synchronous EPSPs can cause
a local voltage response which overshoots their a
rithmetic sum\, the so-called dendritic spike. Thi
s leads to the theoretical proposal that a neuron
can be seen as a network of non-linear subunits\,
a model with a higher computational capacity than
classic models. However\, in practice dendritic mo
rphology can severely reduce the number of possibl
e subunits. This can be an issue\, because the num
ber of independent subunits directly determines th
e increase in computational capacities. Thus\, in
this study\, we take the most conservative hypothe
sis: we suppose the existence of a single non-line
ar dendritic subunit on top of linear somatic inte
gration.\nTo model this hypothesis we use a binary
neuron model\, the threshold linear unit also kno
wn as the Perceptron\, and we add to it a non-line
ar subunit. This subunit locally sums its synaptic
inputs\, and passes it though an all-or-none tran
sfer function\, modeling a dendritic spike. We use
systematic parameter searches on this model to an
swer two questions: Are they computations - Boolea
n functions – enabled by a non-linear subunit ? Ho
w well a computation which is possible with a non-
linear subunit can be approximated without ?\nWe m
ade two observations: first\, a non-linear subunit
coupled with linear integration enables original
computations\, for instance the feature binding pr
oblem\; second\, this computation is the hardest t
o approximate without a non-linear subunit. We use
these observations to implement the binding probl
em in a proof of principle biophysical model made
of a soma and a single dendritic compartment. This
demonstrates that a very simple dendritic morphol
ogy\, a single branch\, already enhances the compu
tational capacities of single neuron.\n
LOCATION:Cambridge University Engineering Department\, CBL
Rm #438 (http://learning.eng.cam.ac.uk/Public/Dire
ctions)
CONTACT:Dr. Cristina Savin
END:VEVENT
END:VCALENDAR