sr

Make plots of mean firing rate (and continuous data - partially implemented) across trials. Error bars show plus and minus one standard error (or standard deviation) around the mean.

The parameters that varies over the set of desired tuning curves should be used as the grouping parameters.

Now, the 'tcode' field, which typically holds the stimulus number or ID value, can be used for grouping. See 'Tcode Grouping' below.

The "condition" statements should be used to select only those stimuli which are members of the tuning curve(s) to be plotted and to exclude baseline or other stimuli.

In the regular parameter section, two baselines may be established using the names "baseline0" and "baseline1", using the format:

  baseline0 groupdef [param_name] [param_val] ... [param_name] [param_val]
  baseline1 groupdef [param_name] [param_val] ... [param_name] [param_val]

If the baseline rate is to be computed from a particular period across all trials, then set the parameters 'bstart' and 'bperiod' to define such an epoch during the trial. When defined, this baseline rate will be subtracted from the tuning curve.

Output types

• append append stats to file.
• plot xplot output.
• data_append append the values to one line of the outfile, separating the values with a space, and ending the line with a newline.
• data columns of floats (no plot headers) *** SO FAR, ONLY W/ TSEQ OPTION ***.
• trial_data print the spike rate value for each trial in each group.
• sr_v_dur for variable duration (vardur) analysis, make a scatter plot of spike rate vs. epoch duration for each group. There must be a vardur parameter specified in the .nda file.
• translation_invar - for translation invariance analysis. See example .nda script below, which includes comments. An xplot file is written that contains three plots: the average firing rate, the maximum firing rate, and the r-value, all as a function of position. The r-value is computed between the responses of all stimuli at the reference position (typically 0.0) and all other positions. Thus, the r-value at 0.0 (or the reference position, if it is not 0.0) should be 1, by construction.
  
• pop_chan_mean - append the channel name and mean response. For use with nda_pop. Useful for writing a list of responses for all channels within a population.

Parameters

• xaxis_convert ori_mirror - Causes a mirroring of x-axis values, assuming they are in degrees of visual angle.
• sem_flag standard error on the mean - by default, this is 1, set it to 0 if you want standard deviation error bars.
• norm_max If set, the tuning curve (after possibly subtracting a baseline) is scaled so that its maximum value is 'norm_max'.
• bstart Start time of epoch for across-trial baseline computation.
• bperiod Duration of epoch for across-trial baseline computation.
• avg_flag (0) If 1, append an across-group average to stat file.
• f1_param [pname] - if defined, 'pname' must contain the TF in Hz as a var or const param. The F1 amplitude will be computed in stead of DC rate. For sinusoidal modulation from 0 to 100 spk/s, F1=DC=50spk/s. For half-wave rectified modulation from 0 to 100 spk/s, F1=50spk/s, DC<50.

Examples are given below.

Tcode Grouping

• TSEQ_0 followed by a list of integers (0,1,...) causes a tuning curve to be plotted for the stimulus numbers in the list. TSEQ_1, etc. may be used for additional tuning curves.
• TSEQ_0_NAME followed by a string specifies a plot name.
• TSEQ_0_X followed by a list of float or integer values specifies the x-axis. The number of values must match that of TSEQ_0. If this is not specified, the value of the last grouping parameter is used for the x-axis.
• See example below

Example - single tuning curve:

#  
#  "ori" varies, ampl is 0 for background stimulus.
#
sr
group 1 ori
plot

xaxis_convert ori_mirror
baseline0 groupdef ampl 0

condition
param_range ampl 1 1024

Example - set of tuning curves (matrix):

#  
#  "dt" and "dr" vary.  A set of plots for each "dt" as a function
#  of "dr" is produced.  Reversing the group parameters changes the
#  axes of the tuning curves.
#
sr
group 2 dt dr
plot

baseline0 groupdef ampl 0

condition
param_range ampl 1 1024

Example - Translation invariance

#
#  This SR analysis output type assumes that the first parameter (in the list
#  following "group" below) is a position parameter, for example 'cx' or 'cy',
#  and the other parameters (following 'cx' below) define the space of stimuli
#  that have been shown at each position.  In this example, the 'direction'
#  has been varied at each 'cx' position.  An alternative example of the
#  'group' line could be:
#
#    group 3 cx shape_id rotation
#
#  which would indicated that a variety of shapes were shown at different
#  rotations at each 'cx' position.
#
sr
group 2 cx direction
translation_invar

chan s2

corr_reference  0.0   # This is an optional parameter that defaults to 0.0
                      # if it is not included in the .nda file.
                      # It determines the x-axis value that will be used as
                      # the reference point to compute r-values for all other
                      # x-axis positions.

#
#  This condition is used to select a y-position ('cy') value, in case 'cy'
#  also varies in the file.  If it does not, then this value is not needed.
#
condition
param_val cy 0.0

Example - Tcode Sequences:

#
#  Three tuning curves are specified.
#  'dir' will be the default x-axis.
#
sr
group 3 ori ampl dir
plot

start 0
period 2000

TSEQ_0  0  1  2  3  4  5  6  7
TSEQ_0_NAME griffiths
TSEQ_0_X 20 21 22 23 24 25 26 27

TSEQ_1  8  9 10 11 12 13 14 15

TSEQ_2 16 17 18 19 20 21 22 23
TSEQ_2_NAME island