Noise

The Noise routine is used to generate a noise stimulus through the DSS system and collect and display neuronal response spike data. It reads the waveform for the noise stimulus from a data file. By default, it uses a noise waveform stored as the dataset id called noise40k in the data file called d:SERWAV. This waveform was created by the GWE program. It has 8192 data points and is composed of frequencies from 9.77 Hz to 40000 Hz. The sample time of each data point is 12.5 microseconds, and the total playback time of the waveform is 102.4 msec.

The user has the option of overriding this default noise waveform by using another waveform. This is done either at the start of the Noise routine, when the option is given, or at any time while running the Noise routine, by using the 'f' command (Fetch Waveform). This command prompts the user for the name of the data file and the data set id of the desired waveform. Note that the file and data set must already exist. One should create them before calling the SER program. (See the GWE program for creating such waveforms.)

Here is a sample menu of the Noise stimulus. A series of 50 noise bursts will be generated, with each burst lasting 200 msec, and with a repetition time of 500 msec between the start of each burst. The stimulus will be generated in both phones, and the SPL for each phone will be 70 dB. (Separate SPL levels for the 2 different DSS's are not available at this time.)

			      Number of Bursts =   50
			       Repetition Time =   500.00 msec
			      Simulus Duration =   200.00 msec
			       Delay of Master =   500 usec
			      Master DSS (1,2) :   2
	      Phone (0=none, 1=1, 2=2, 3=both) :   3
					   SPL =   70 dB
	Slave Sample Time - Master Sample Time =  -0.020 usec
	  Display (0=none, 1=all, 2=no raster) =   1
				       UETCHAN =   0
				      UETCHAN2 =  -1

w=Dset(name), u=Dset(unit#), x=Dset(seq#), d=t=Scr Dump, f=Fetch Waveform
p=P=CTRL-P=Pause, g=G=CTRL-G=Go, e=E=CTRL-Z=Exit, r=R=Retrieve, s=S=Store

Note that since the specified duration time of the stimulus is longer than the playback time of the noise waveform, the waveform will be repeated until the full duration time is reached. If the waveform playback time were greater than the duration time, the waveform would be truncated at the duration time.

Normally, the noise stimulus will be played back at the sampling rate specified in the waveform data set itself. There is, however, a feature which allows for a different sample playback time for the 2 DSS's. This is specified with the parameter Slave Sample Time - Master Sample Time. Here, the Master DSS will play back the waveform at the rate defined in the waveform data set. The Slave DSS will play it back with a sample time equal to this plus the value specified in the Sample Time difference parameter. Note that negative values are allowed, and in this case the sample time of the Slave would then be less than that of the Master. Note also that the possible range of values for this parameter is from -10 usec to 10 usec.

The Delay of Master parameter allows for the stimulus of one of the DSS's to be presented at a slight delay after the stimulus of the other DSS. One must first specify the parameter Master DSS, as either 1 or 2. Then, depending on which DSS is chosen as the master, the delay will be placed on that DSS. Note that negative delays are allowed.

An additional feature of the Noise routine is that the user can now specify 2 different input channels for the collection of spikes through the UET (Unit Event Timer). The UET does have 16 input channels, numbered 0 through 15. Channel numbers 2, 3, and 6 are reserved for internal signals, but the user could connect data inputs to any of the other 13 channels. The user must specify the numbers of the 2 channels to be used. The menu parameters UETCHAN and UETCHAN2 are used for this.

If only one UET input channel is being used, then the user should set either UETCHAN or UETCHAN2 to the value of that channel number, and set the other channel parameter to -1. In any case, a channel number of -1 means no input channel and therefore no data to be displayed.

The output Display of the Noise routine has two parts. There is a plot of the accumulation of spikes generated after every series of noise bursts, plus a dot raster display of the spike times during each repetition interval. The display parameter has 3 possible values: 0 = no display, 1 = display both the spike curve and the dot raster, and 2 = display the spike curve but no dot raster.

Note that if both UETCHAN and UETCHAN2 are set to channel numbers other than -1, then 2 spike curves and 2 dot rasters must be included in the output display. The output corresponding to UETCHAN is plotted as solid vectors, while that corresponding to UETCHAN2 is dashed vectors. In the case of a color display, the dashed vectors should show up in a different color. Note that the program runs considerably faster without any data display.

The arrow keys are used to move around in the menu. There are also several single-character commands available to the user at all times, while in the Noise menu. They are listed here, and described in more detail below.

	g, G, or ^G	Start or restart stimulus
	p, P, or ^P	Stop stimulus, stay in Tone Pips menu
	e, E, or ^Z	Stop stimulus, return to SER main menu
	f or F		Fetch new noise waveform from user file
	d or D		Dump next spike display to printer
	t or T		Dump next spike display to printer
	s or S		Store current menu params to special block
	r or R		Retrieve menu params stored by 's'
	w or W		Save current data, specify data set name
	u or U		Save current data, increment unit #
	x or X		Save current data, increment sequence #

After setting the parameters to the desired values, the series of Noise bursts is started by typing CTRL-G. At the end of each series of bursts, the display (if specified) will be updated and the bursts will be generated again. This continues until the user stops the stimulus. As with the FM Sweep routine, 'e' is used to exit, and 'p' stops the stimulus. Any individual parameter can be changed at any time -- type the new value, then hit 'Return' or one of the arrow keys to move the cursor. Then, stimulus will stop and restart with the new parameter values. Note that, as with the FM Sweep, it is can be faster to change a number of parameters by first typing 'p' to stop the stimulus, then modifying the parameters, then hitting 'g' to restart.

With the Noise stimulus there is the special feature of a private storage block for the menu parameters. To store the current set of menu parameters into this storage block, type 's' or 'S'. Then, continue to run the stimulus, changing any parameter to a new value. To retrieve the set of parameters stored by the S command, type 'r' or 'R'.

There is now a hard copy feature, specified with the 'D' or 'T' option. Simply type D or T and the program will send to the printer a copy of the spike curve and dot raster at the end of the current series of noise bursts. (The display is only updated at the end of each series of bursts, and so both the 'd' and 't' options have been designed to do the same thing, that is, to send a hard copy dump of the next update of the display.) Note that if the Display parameter is set to 2, then there will be no dot raster data in the hard copy. If it is set to 0, the hard copy command will be ignored.

Invalid parameter settings sometimes cause the stimulus to stop. The user must change the appropriate parameter(s), then type 'g' or CTRL-G to restart the stimulus.

To store the response data from the current series of noise bursts, the user can type 'w', 'u', or 'x'. Each of these commands stores the data for the entire series of bursts. The only difference between the 3 commands is how the name for the new data set is chosen. Each command displays the name of the most recently stored data set. The 'w' command prompts the user to type in a new name for the data set to be stored. The 'u' or the 'x' command will try to automatically create a new data set ID name. They look at the previous name, which is understood to have the format 'unit#-sequence#', preceded by an optional string of characters. For example, a valid name would be abc100-12, where the unit number is 100 and the sequence number is 12. The 'u' command will increment the unit number, and the 'x' command will increment the sequence number. If the previous data set name is not in this format, then each of the commands will prompt the user to type in a new data set id name.

The data for a series of noise bursts are stored with the data set type name SRNO, which will distinguish it from data collected during one of the other stimulus options. The schema name for each data set is sch016.

See also the general description of the SER program.
See also the descriptions of the other SER stimulus options:
FM Sweep, Tone Pips, Clicks, Dual Tones