Linear Filter - lin_01

Overview

The lin_01 model implements a 1-D linear filter followed by spike train generation. It also may contain any number of units in a second layer, defined by unit objects, that combine the spikes from the linear unit with independently generated Poisson spikes.

Configuring the Model

lin_01 models are defined using the following objects and parameters:

#
# lin_01.moo
#

mod_type lin_01

tscale    0.001
tn     1024

<filter>
...
</filter>

<spike_gen>
...
</spike_gen>

<unit>                      # OPTIONAL
...
</unit>
  • tscale - temporal resolution (samples per second)
  • tn - duration of the model (sampling units)
  • <filter> object that specifies parameters for the temporal filter
  • <spike_gen> object that defines the method of spike generation from the final filter output
  • <unit> objects that specify parameters for 2nd layer units may optionally be included

Model Outputs

Model Components

The following command allows users to output the temporal filter itself:

  • write_dog_temporal [outfile] - write the temporal filters.

This includes the temporal kernel and the Gaussian mask, which multiplies the biphasic kernel to adjust the size of the positive and negative lobes.

Model Responses

The following response types can be requested:

  • spikes - spike times
  • f - linear filter output
  • prob - probability of spiking (if Poisson spike generation)

Models of This Type