spore::PoissonDblExpNeuron Class Reference

Point process neuron with double-exponential shaped PSCs. More...

#include <poisson_dbl_exp_neuron.h>

Inheritance diagram for spore::PoissonDblExpNeuron:

Collaboration diagram for spore::PoissonDblExpNeuron:

Public Member Functions
	PoissonDblExpNeuron ()
	PoissonDblExpNeuron (const PoissonDblExpNeuron &)
nest::port	send_test_event (nest::Node &, nest::rport, nest::synindex, bool)
void	handle (nest::SpikeEvent &)
void	handle (nest::CurrentEvent &)
void	handle (nest::DataLoggingRequest &)
nest::port	handles_test_event (nest::SpikeEvent &, nest::rport)
nest::port	handles_test_event (nest::CurrentEvent &, nest::rport)
nest::port	handles_test_event (nest::DataLoggingRequest &, nest::rport)
void	get_status (DictionaryDatum &) const
void	set_status (const DictionaryDatum &)
Public Member Functions inherited from spore::TracingNode
	TracingNode ()
virtual	~TracingNode ()
void	get_trace_status (DictionaryDatum &d) const
const_iterator	get_trace (nest::delay steps, trace_id id) const
	Access the trace of id at time step step. More...
const_iterator	get_trace (nest::Time const &time, trace_id id) const
size_t	get_num_traces () const

Friends
class	nest::RecordablesMap< PoissonDblExpNeuron >
class	nest::UniversalDataLogger< PoissonDblExpNeuron >

Additional Inherited Members
Public Types inherited from spore::TracingNode
typedef CircularBuffer< double >::const_iterator	const_iterator
typedef size_t	trace_id
Protected Member Functions inherited from spore::TracingNode
void	init_traces (size_t num_traces)
void	set_trace (nest::delay steps, double v, trace_id id=0)

Detailed Description

Point process neuron with double-exponential shaped PSCs.

PoissonDblExpNeuron is a variant of the spike response model model with double exponential PSP shapes of the from and absolute refractory mechanism.

Spikes are generated randomly according to the current value of the transfer function which operates on the membrane potential. Spike generation is followed by an optional dead time (refractory state). Setting with_reset to true will reset the membrane potential after each spike.

The transfer function can be chosen to be linear, exponential or a sum of both by adjusting three parameters:

where the effective potential and is called the adaptive threshold. By setting c3 = 0, c2 can be used as an offset spike rate for an otherwise linear rate model.

The dead time enables to include refractoriness. If dead time is 0, the number of spikes in one time step might exceed one and is drawn from the Poisson distribution accordingly. Otherwise, the probability for a spike is given by , where h is the simulation time step. If dead_time is smaller than the simulation resolution (time step), it is internally set to the time step. Note that, even if non-refractory neurons are to be modeled, a small value of dead_time, like dead_time=1e-8, might be the value of choice since it uses faster uniform random numbers than dead_time=0, which draws Poisson numbers. Only for very large spike rates (> 1 spike/h) this will cause errors.

The model implements an adaptive threshold. If the neuron spikes, the threshold increases and the membrane potential will take longer to reach it. If the neuron does not spike, the threshold linearly decays over time, increasing the firing probability of the neuron (see [6], parameters: target_rate, target_adaptation_speed).

This model has been adapted from iaf_psc_delta. The default parameters are set to the mean values in [2], which have were matched to spike-train recordings.

Parameters

The following parameters can be set in the status dictionary. (default values and constraints are given in parentheses):

name	type	comment
V_m	double	Membrane potential [mV]
tau_rise_exc	double	Fall time constant of excitatory PSP (2.0, >0.0) [ms]
tau_fall_exc	double	Rise time constant of excitatory PSP (20.0, >0.0) [ms]
tau_rise_inh	double	Fall time constant of inhibitory PSP (1.0, >0.0) [ms]
tau_fall_inh	double	Rise time constant of inhibitory PSP (10.0, >0.0) [ms]
dead_time	double	Duration of the dead time (1.0, &ge 0.0) [ms]
dead_time_random	bool	Should a random dead time be drawn after each spike? (false)
dead_time_shape	int	Shape parameter of dead time gamma distribution (1, &ge 1)
t_ref_remaining	double	Remaining dead time at simulation start (>0.0) ms
with_reset	bool	Should the membrane potential be reset after a spike?
I_e	double	Constant input current (0.0) [pA]
input_conductance	double	Conductance of input currents (1.0) [S]
c_1	double	Slope of linear part of transfer function (0.0) [Hz/mV]
c_2	double	Prefactor of exponential part of transfer function (1.238) [Hz/mV]
c_3	double	Coefficient of exponential non-linearity of transfer function (0.25, &ge 0.0) [1/mV]
target_rate	double	Target rate of neuron for adaptation mechanism (10.0, &ge 0.0) [Hz]
target_adaptation_speed	double	Speed of rate adaptation (0.0, &ge 0.0)

Sends: SpikeEvent

Receives: SpikeEvent, CurrentEvent, DataLoggingRequest

References

[1] Multiplicatively interacting point processes and applications to neural modeling (2010) Stefano Cardanobile and Stefan Rotter, Journal of Computational Neuroscience.

[2] Predicting spike timing of neocortical pyramidal neurons by simple threshold models (2006) Jolivet R, Rauch A, Luescher H-R, Gerstner W. J Comput Neurosci 21:35-49.

[3] Pozzorini C, Naud R, Mensi S, Gerstner W (2013) Temporal whitening by power-law adaptation in neocortical neurons. Nat Neurosci 16: 942-948. (uses a similar model of multi-timescale adaptation).

[4] Grytskyy D, Tetzlaff T, Diesmann M and Helias M (2013) A unified view on weakly correlated recurrent networks. Front. Comput. Neurosci. 7:131.

[5] Deger M, Schwalger T, Naud R, Gerstner W (2014) Fluctuations and information filtering in coupled populations of spiking neurons with adaptation. Physical Review E 90:6, 062704.

[6] David Kappel, Robert Legenstein, Stefan Habenschuss, Michael Hsieh and Wolfgang Maass. Reward-based self-configuration of neural circuits. 2017. https://arxiv.org/abs/1704.04238

Author

Kappel, Hsieh; (of pp_psc_delta) July 2009, Deger, Helias; January 2011, Zaytsev; May 2014, Setareh

See also

TracingNode

Constructor & Destructor Documentation

PoissonDblExpNeuron() [1/2]

spore::PoissonDblExpNeuron::PoissonDblExpNeuron

(

)

Default Constructor.

PoissonDblExpNeuron() [2/2]

spore::PoissonDblExpNeuron::PoissonDblExpNeuron

(

const PoissonDblExpNeuron &

n

)

Copy Constructor.

Member Function Documentation

get_status()

void spore::PoissonDblExpNeuron::get_status ( DictionaryDatum &  d ) const

inlinevirtual

Status getter function.

Reimplemented from spore::TracingNode.

handle() [1/3]

void spore::PoissonDblExpNeuron::handle

(

nest::SpikeEvent &

e

)

SpikeEvent handling.

Parameters

e	the event.

handle() [2/3]

void spore::PoissonDblExpNeuron::handle

(

nest::CurrentEvent &

e

)

CurrentEvent handling.

Parameters

e	the event.

handle() [3/3]

void spore::PoissonDblExpNeuron::handle

(

nest::DataLoggingRequest &

e

)

DataLoggingRequest handling.

Parameters

e	the event.

handles_test_event() [1/3]

nest::port spore::PoissonDblExpNeuron::handles_test_event ( nest::SpikeEvent &  , nest::rport  receptor_type  )

inline

PoissonDblExpNeuron test event.

handles_test_event() [2/3]

nest::port spore::PoissonDblExpNeuron::handles_test_event ( nest::CurrentEvent &  , nest::rport  receptor_type  )

inline

PoissonDblExpNeuron test event.

handles_test_event() [3/3]

nest::port spore::PoissonDblExpNeuron::handles_test_event ( nest::DataLoggingRequest &  dlr, nest::rport  receptor_type  )

inline

PoissonDblExpNeuron test event.

send_test_event()

nest::port spore::PoissonDblExpNeuron::send_test_event ( nest::Node &  target, nest::rport  receptor_type, nest::synindex  , bool    )

inline

PoissonDblExpNeuron test event.

set_status()

void spore::PoissonDblExpNeuron::set_status ( const DictionaryDatum &  d )

inlinevirtual

Status setter function.

Reimplemented from spore::TracingNode.

---

The documentation for this class was generated from the following files:

• poisson_dbl_exp_neuron.h
• poisson_dbl_exp_neuron.cpp