QUESTIONS & ANSWERS |
Printing problems currently exist with certain HP printers, possibly due to the bidirectional communications of these printers through the dongle.
We have discussed various fixes on the SKIL user group (skil@egroups.com). If you have not
subscribed yet, we highly recommend that you do.
Here are some fixes that have worked for a number of people:
|
|
LEXICOR DAT FILES How does one get the data from the Lexicor program to the SKIL program? The Lexicor program creates .DAT files. They are stored in some folder you designate in Lexicor (changed via Edit; Path). Once the files are acquired by Lexicor, close Lexicor and run SKIL. Create a new report. Once it's created, you will see the Load EEG file icon un-ghost, along with other commands/icons. Click on the Load EEG icon and you get a Windows dialog (files and folders; the top blue bar of the window says "Open EEG File:). Navigate to the folder where you saved the Lexicor file, and click on one of the .DAT files to select and then click the Open button. |
|
MAKING SENSE OF TASK COMPARISONS What does the task comparison EC:EO mean? There are 5 task comparisons possible for grid, topometric, and brain map views -- EC:EO, EC:T1, EC:T2, EO:T1, EO:T2, which refers to Eyes Closed compared to Eyes Open condition, Eyes Closed compared to Task 1, Eyes Closed compared to Task 2, Eyes Open compared to Task 1, and Eyes Open compared to Task 2, respectively. The formula used to generate these comparisons is: (X2-X1)/X1 (x 100, to convert values to percent). X1 and X2 refer to the conditions in the label, e.g., EC = X1, EO = X2 in EC:EO. In the mid-frequency range, this calculation indicates the expected activation that occurs when a subject performs the 2nd condition, in relation to the 1st. How much one activates in eyes open condition relative to the eyes closed baseline, for instance. In the mid-frequency (e.g., alpha) range, we expect negative values from this calculation for all 5 comparison. E.g., subjects activate in EO (compared to EC) and thus exhibit less alpha amplitude in EO. These comparisons must be understood in context. If one or more is significant, examine the individual conditions contributing in this comparison to see whether one or both tasks are primarily responsible for the abnormal activation. |
|
WHICH TASKS TO USE The SKIL TOPOMETRIC OPERATIONS KIT Manual on page 8, states that Task 1 = Audio-Visual processing, and Task 2 = Visuomotor Processing. Since we are comparing our clients data with a normative database I would imagine it is important to administer the same tests that represent Task 1 and 2, respectively, in the database. So, what tests do you recommend be used for Task 1 and 2? Any moderately challenging, continuous task with the appropriate sensorimotor components meets the criteria for the database comparison. Task 1 can be any audio/visual task without a significant motoric component. Task 2 should be a visual tracking task with a significant (and, if possible, continuous) motoric component. We recommend for Task 1 viewing a brief video for content. A common alternative is reading (quietly) a long passage. Both approaches should have a post-task quiz to ensure subject compliance. For task 2, a video game requiring constant tracking or scanning such as "Breakout" is recommended (the database used the CTT, Cont. Tracking Test). Barry has also compared a task of crossing out "e"'s in a passage with Task 2, with good results. |
|
MOVING FILES BETWEEN TWO PCs I had trouble transferring files from my office to my home. For example, I create a SKIL report at work, bring the .SKR file and the .DAT files home, and cannot open the .SKR file there. I have even tried copying the files to the same subdirectories at home as at my office. Duplicating the directory structure is the exact approach I would use too. I presume the operating system is the same. This shouldn't matter, but it could. It may be the case that at one location you installed two versions of SKIL over time, say 1.0 and 1.01, and only 1.01 at the other location and some "program debris" from the 1.0 installation is messing with the file transfer. If this scenario is possible, you probably can fix it by uninstalling all versions of SKIL on each PC (use the Add/Remove programs under the Control Panel in Windows) and then reinstalling the version you use only. The current versions of SKIL (v1.09+) are suppose to install entirely separate from other installations and not use or look for previous support files, but earlier versions appeared to latch on to previously installed files and made us unable to open past reports until we uninstalled all versions like I said above. |
|
RECORDING CONDITIONS I would like to begin a standard protocol for doing a brainmap using the Mindset system. Following the protocol of Joel Lubar, I was wondering if an "at rest" map followed by a "concentration task" map would be appropriate. For example, could the practice visual test from the TOVA be used for the concentration task?
I would include two replications of 3-min eyes closed, eyes open, and challenge conditions, in the following order: EC1 EO1 Task1 EO2 Task2 EC2
The visual practice test of TOVA would be a good challenge test. We will shortly be interfacing the TOVA test itself with Mindset mapping in order to perform ERD analyses. ERD (Event response desynchronization) is similar to an ERP (Event response potential), but in the frequency domain. Essentially we average over all trials (grouped by trial classifications such as target/nontarget, correct/incorrect responses) the frequency response one second before the stimulus presentation to 2 seconds after. This provides a baseline (pre-stim-presentation), an orienting response (a rise in 5-7 Hz in particular 125 to 250 ms after stimulus) a cognitive processing response (an ERD at 500 ms to 625 ms, seen as sharp attenuation in 7-11+ Hz) and a post-response synchronization (PRS) 1000 to 1250 ms after stimulus presentation. These three components, compared to baseline, vary on latency and amplitude depending upon subject capabilities and processing demands. We will impliment the ERD database in about 8 months. When we do, it will be with the visual TOVA and possibly some Thinkfast tests. |
|
SAMPLE RATE If I record at 128 srate, when I go to do the automatic artifact detection on SKIL, almost everything is deleted. Should I be recording at 64 srate, and if so, am I losing any data? According to the Lexicor manual, the highest frequency I can record at 64 srate is 1/4 the srate or 16 Hz. The sample rate doesn't interact with artifact detection and elimination.
We recommend a minimum sample rate of 128. (128 srate is fine, you need not go higher for the analyses involved in most clinical applications.)
Did you turn on the highpass before recording? Perhaps the signal is drifting, which makes normal signals reach the criteria for artifact. Having the highpass on for all recording will stop the very low frequency drift (down to a DC signal) that is common with many recordings.
|
|
SMR/BETA After Barry's presentation in Palm Springs, I was left with the following question: What is left side 15-18 Hz training really all about? what is the scientific support for this belief? I am beginning to believe that 12-15 Hz is all that has evidence for success based on Barry's presentation. What are your experiences, beliefs, data to support left side 15-18 Hz training? Barry says the human SMR range is from 12-19 Hz and this large range is often what he trains (or at least used to when I helped him in his clinic). I do not know where he derived this range as the beta range (12-40 Hz) has yet to reveal any unitary mechanisms, unlike alpha. There is some evidence for faster rhythms in beta indicating more processing, but nothing with strict boundaries. The Othmers derived their boundaries (which they vary for some individuals) from clinical experience. One can always be blinded to new information from one's theories, so although I suspect the bands they use have some relevant correlates of function, I would not make hard rules about sides and frequency ranges in beta. Particularly when handedness, gender, and cognitive style are often overlooked.
|
|
HIGHPASS OFF I did not have the highpass filter on before recording, and as you mentioned, there is "drift" in these recordings. Is there a way to correct for this in the artifact detection limits? I see in the "methodology" section, you mention that the sampling rate is set at 256 or better. Is there any advantage to this setting? In the future it would be best to record EEG with the highpass filter on. Mindset users do not need to deal with the highpass as it is always on; Lexicor allows the user to turn the highpass on or off. For these recordings, as with any recordings, you can do the following to augment the automatic artifacting: 1. Increase the amplitude value for artifact detection (change default 3000 value to 3500, 4000, or higher) so that only the largest waveforms are designated as artifact. This may compensate for the drift. 2. Manually reduce the artifact marks, as described in the manual. The 256 sample rate is for Mindset files. Higher sample rates are preferred, but there is only a slight advantage in accuracy for 256 compared to 128 sample rate. Doubling the srate doubles file size, which makes files harder to email or store on floppies (though not such a problem if one's backup & storage system is Zip or Recordable CDs). Mindset files are smaller than Lexicor and thus a 256 sample rate stores at nearly the same byte size as a 128 sample rate Lexicor file. 128 srate for Lexicor is entirely acceptable for comparison to the adult database. Lower sample rates reduce the maximum spectral frequency which can be calculated -- but in conjunction with epoch length. I believe Lexicor makes the default epoch 4 s long for 64 srate, which limits the highest spectral bin to 16 hz. Check to see if the values above 16 Hz are above zero. |
|
SKIL REPORT WON'T SAVE I am aware of one example of the save function not working on a report someone has worked on. In this case, the EEG data was stored on a CD disk and the SKIL program was loading EEG files directly off the CD (instead of loading EEG files from the hard drive). This may be a problem for some PCs. The solution is simple: Before running the SKIL program, copy all EEG files you wish to analyze to a folder on your hard drive. Also, be sure to remove the CD disk from its drive so you don't accidentally load a file from the CD during your work. We recommend that you create dedicated folders to hold your EEG data and SKIL reports (with names like EEG Data and SKIL Reports, why not). You can always delete the EEG files you transfered to the EEG data folder once you complete the report for this dataset. If you wish to keep the EEG data on your hard drive indefinitely (given the large size of hard drives nowadays you might), consider creating subfolders inside the EEG Data folder named after each recording session (e.g., "JD April0399" or "TommyK EEG Session 1") Also, disable any screen-saver program or any other unessential program that runs in the backgrounds. These may interrupt processing and interfere with saving files. |
|
SKIL REPORT CRASHED As for the SKIL report crash, you must load at least one EEG file for the report to be saved correctly. It's a bug, but it's easily dealt with. You do not have to do any work on the files immediately, just load them into the report and that should allow later retrieval of a saved report. 32 Megs of RAM is okay, but it is about the minimum. If your reports have numerous graphics, especially brain & covariance maps, you'll see much shorter display and printing times with more RAM. Also, you may have to disable any screen-saver program or any other unessential program that runs in the backgrounds. These programs may interrupt processing and cause the crash. |
|
PRE/POST TRAINING ASSESSMENT How do I compare progress of a child before and after neurotherapy?? How Can I compare task 1 before and after. * Which brings me to the most IMPORTANT question of all.... What are the details of the tasks.... Please send that urgently so that meaningful comparisons and analyses can be made. * What criteria is there for inclusion/exclusion of subjects into the database norms * I will be brainmapping dozens of children soon.. I cannot do it without these details. We are currently completing an update the of software (v1.1) which will include replications of all conditions, allowing mapping and comparison of Pre/Post or Rep1/Rep2 of all conditions. Task 1 is a passive visual processing task -- passive in terms of gross motoric responses. Reading comprehension, video viewing, and similar tasks will do. Task 2 is a motorically-active visual processing task which includes tracking, math problems, and similar tasks. We will be updating the manual with the upgraded software to make this clearer. Presently this system includes an adult database. Although a child database is planned, its inclusion will not be until much later this year at the earliest. This does not mean you cannot analyze children's data with the program and either graph pre/post -- however the norm and +/- 2 SD marks should be ignored in one's conclusions (as they are for adults). This will still provide a relative change between conditions and replications/treatments. |
|
CLIPBOARD CONTROL How does one delete an eeg file from a report and replace it with another. Say if one finds it is the wrong one or whatever... You may like to consider putting an on screen label in figures that have been selected for inclusion on the report to say what figure number itis. It would make writing the report findings etc. much easier. I tried copying MS Word 7 onto the clipboard and into the report, and I find that I get all the control codes copied as well as the text... The upcoming versions 1.1 will allow removal of appended EEG files from a report file. (The present version appends all EEG files opened to a report, so although these unwanted files don't effect the report generation, they increase the report file size unnecessarily) Version 1.1 will allow sorting of figures which should reduce the figure inclusion problem. I have yet to see it implemented so I'm not sure if those figures not included will be placed at the end or treated differently. For MS Word documents: Copying selections of a Word document should maintain the formatting, but not the command codes... hmmm. If you save the Word document as Text Only and then open the text file with any word processing program (even Word will do). This will strip out the commands, leaving the basic formatting only (i.e., line breaks, spaces). Then copy the selections over. |
|
SOFTWARE INSTALLATION ERROR I tried to load it by using "setup", then "browse", then click on the only exe file, but came up with "internal error". What might be happening? Are you using Windows 98 -- as Windows 95 doesn't use the command "browse". Try this:
|