The data set is organized by recording session. Each session was recorded on a different date. For each session, there is a data file (*.dm6), a channel file (*.chn), and a log file (*.log).

File names

The name of each file is coded as follows:

characters 1-2:	monkey name (ge, lu, or ti)
characters 3-4:	session id
character 5:	s (meaning "stimulus")
character 6:	stimulus type (1-4 for the 4 different stimulus types, or "a" for all stimulus types) [See Info:conditions.txt]
character 7:	r (meaning "response") [See Info:conditions.txt]
character 8:	response type (4 for correct "go", 5 for correct "no-go", or "a" for "all" response types, i.e. both correct go and no-go)
character 9:	channel subset id (an arbitrary letter specifying the channel subset)
extension:	"dm6" indicates that 60-hz removal was performed on the data

The data file

The data file is organized as trials x channels x time points, where trials is the slowest changing variable (i.e. the outermost loop).

The data are Local Field Potential time series sampled at 200 samples per second, i.e. 5 msec per data point. The Local Field Potential amplitude units are microvolts.

For data files with the dm6 extension, trials having large amplitude deviations have been eliminated, and 60-Hz line frequency contamination has been reduced according to the method of sinusoidal estimation by Multitaper Spectral Analysis (Mitra & Pesaran, Analysis of Dynamic Brain Imaging Data, Biophysical Journal, 76:691-708, 1999).

NB: No other transformations were applied to the time series in the dm6 files. In our analysis, ensemble mean subtraction and linear detrending are commonly performed on the dm6 file data as preprocessing steps prior to further analysis steps.

The channel file

The channel names are 4-character codes listed in the order that they appear in the data file. The codes can be located on the maps in Maps:*.eps by finding the corresponding numbers in the master files in Maps:*_master.txt. In addition, extended channel names are also found in the master files.

The log file

The log file is organized with one trial per line. There are 13 entries per line:

entry 1:	monkey name and session id
entry 2:	the word "trial" indicating that the next entry is the trial number
entry 3:	trial number
entry 4:	the word "sty" indicating that the next entry is the stimulus type
entry 5:	stimulus type (1-4)
entry 6:	the word "sttm" indicating that the next entry is the stimulus time in data points
entry 7:	the stimulus time in data points
entry 8:	the word "rty" indicating that the next entry is the response type
entry 9:	response type (4 or 5)
entry 10:	the word "rstm" indicating that the next entry is the response time in data points
entry 11:	the response time in data points
entry 12:	the word "cty" indicating that the next entry is the contingency type
entry 13:	contingency type (1 or 2)