This program presents a field of "noise" symbols that sometimes contains an imbedded "signal". The subject's task
on each trial is to decide whether or not the signal was present. If present, the subject should push the '/' key, if
absent, the correct response is the 'Z' key. No feedback is presented after the trial. (If a key other than the Z or / is
pressed then a tone is presented.) The response need not be speeded--reaction time is not recorded. The appropriate
data are the proportions of "present" and "absent" responses for trials that did or did not include a signal. The
program allows considerable control over the timing of trial events, and the nature of the signal and noise symbols.
Running the experiment
If your computer operates at two speeds this program should be run at the faster speed. To run the experiment, type SIGDTECT and hit "Enter". When asked to "Press any key to begin", pressing the 'P' key will allow you to change various parameters that determine the design of the experiment and the timing of trial events. Pressing a key other than 'P' will cause the program to begin the first trial immediately. The parameters that can be safely changed are listed below, along with their probable default values.
1. Number of trials per block:
(default=40)
2. Total number of blocks per session:
(default=5)
3. Number of noise elements in display:
(default=150)
4. Number of practice blocks
(default=0)
5. Inter-trial interval (msec):
(default=750)
6. Duration of error tone (msec):
(default=500)
7. Type of noise element (.,+,*,square,X,diamond):
(default=1)
8. Type of signal element:
(default=2)
12. Duration of stimulus exposure (msec):
(default=50)
13. Probability of target present (%):
(default=50)
14. Minimum x of active display area:
(default=8500)
15. Maximum x of active display area:
(default=23500)
16. Minimum y of active display area:
(default=8500)
17. Maximum y of active display area:
(default=23500)
20. Maximum allowed RT (msec):
(default=5000)
Note. The locations of the noise and signal elements are determined randomly on each trial.
Analyzing the data (You will need analysis program available below)
To analyze the data type SIGANAL and hit "Enter" from the directory in which the experiment was run. To view the summary analysis TYPE or PRINT the file ASIG.ANL (i.e., issue the command: TYPE ASIG.ANL "Enter"). If you want to view the raw data, TYPE or PRINT the file DSIG.DAT. Each time that SIGDTECT is run, the program will destroy all previous DSIG.DAT data files. Each time that SIGANAL is run, it will destroy all old ASIG.ANL files. Therefore, be sure to look at the data before the experiment is run again.