I am still not completely clear on the issue of "Standard" versus NeuroGuide
bands (my last question). Isn't the NG Z-score
frequency-band-specific? In other words, if the NG normative database
defines Theta as 4-8 Hz., but Brainm defines it (and trains it) as 4-7 Hz., then isn't there a discrepancy, and doesn't this compromise the accuracy of the NG DLL for training purposes? Or, is the NG DLL defined in single-Hz. increments, so that it applies its Z-scores to each individual frequency, and it's just up to the user to decide which frequencies he or she wants to be defined by each band? Thanks for helping me sort out this issue--it seems like it would make a difference.
My other question (embedded in my first paragraph) involves coherence
training. Will coherence training as defined by the Event Wizard only work between two points, or can it work between three or four points, as well? It looks like the Event Wizard commands (eg., ZCOA) automatically addresses the coherence between whichever points are defined in the training protocol
. What I don't know is if this only works if you only have two points (and therefore only one coherence score for each frequency band), or if it works with three points (and therefore three coherence scores per band) or four points (and therefore six coherence scores per band). Obviously, things get complex very quickly, but being able to train multiple coherences simultaneously to Z-scored norms could be very useful in trying to normalize the brains of the little autistic kids whom I see.
No, the different frequency band definitions does not compromise anything, and there is not a discrepancy. NeuroGuide does all of its own computations, and it is not an issue if the main BrainMaster software is using different bands. You could use BrainMaster to train multiple sliding, variable bands with all kinds of variations, and still use the DLL bands as they are. It is no different than using any kind of weird training bands, then running the data through NeuroGuide for a QEEG
. It does not cause any problems at all. The NG DLL uses its 8 fixed bands, which cannot be changed. In fact it is a beauty of the system that BrainMaster allows you to train any way you like, and the NG DLL is a fixed, solid reference
database. I think you are making it more difficult than it needs to be. Just use the NG DLL for what it is, and use its predefined, fixed bands. If you want the BMr bands to match, you are free to do so. But many people use the BMr software in all of its variations, and use the NG DLL for its purpose. They do not conflict.
The Event Wizard currently only sees coherence between the pairs channels 1-2 and channels 3-4. We are going to upgrade BMr to compute all possible combinations of pairs, but for now this is what we provide via. our software. From the NG DLL, however, you can get any of the up to 6 possible pairs of bands, and access the Z-scores using the proper equation terms, as defined in the Data Dictionary. The Data Dictionary includes the ability to train all possible coherences, using for example PZCOK(range) which means "percent of z scores of coherence which are within plus or minus 'range' of zero". If you use 2 channels, it is 1 coherence per hand, thus 8 coherences (delta, theta, etc.), with 4 channels it is 6 coherences per band, thus 48 coherences. So using Z scores it is easy to grab all the possible scores and train on them in combination. To train them individually would be prohibitively complex, as you note, so we provide these "helper" functions. There are actually dozens of helper functions in the data dictionary that you can use.
I have been using my Atlantis 4x4 for over a year, and have a couple of patients for whom I need to some more specialized coherence training. I have called BrainMaster several times for help, but haven't been able to get what I need. So, I am e-mailing you. First, I need to know if it is possible to train coherence between four separate points, or only two. Second, I want to use the Event Wizard to set up the training protocol, rather than using one of the "prepackaged" options. I get the basics of the Event Wizard, but I do have some questions. Specifically:
1. I assume that I can use the NeuroGuide Z-score
DLL (which I use regularly for amplitude
training), basically by mimicking a couple of your prepared programs--e.g. "Z-Score
Demo Four Coherences Normal" or "Z-Score
Training Five Coherences Normal." I also understand from Bob Thatcher that it's better to train wide, rather than narrow, frequency bands, which is fine.
Yes, you can use our settings files to start with and modify from there. The bands themselves are fixed, as they are built into NeuroGuide.
I agree that using wider bands provides better response generally.
2. What I don't understand, when I get into these two programs, is why there are Events that specify equations that "Do Nothing" and are Disabled--e.g. Events 2, 3, and 4 in the "Five Coherences" program, and Events 3, 4, and 6 in the "Four Coherences" program. I can't see where those equations are flagged for anything or otherwise used, so why are they there?
Some events are set up so you can see a graph, but nothing needs to happen. These would be enabled and visible, but do nothing. The disabled events yiou see are likely left over from previous designs, and just disabled. You can disregard them and use the event numbers for something else. In my version of Four Coherences Normal, Event 2 is set to "If x=E1F" is greater than 0.5 then Award Point (Counter 1) with a refractory period of 1000 milliseconds. This adds points counting, for example.
3. I need to be able to run either the Flash Games Player or the BMrMultimedia/DVD Player (for different patients) during coherence training, but I am not sure how to insure that the Event Wizard is enabling those functions. I assume that flagged equations that award points (e.g. Event #2 on the "Four Coherences" program, or #6 on the "Five Coherences") will enable the Flash Games Player, but I am not clear if this Event will also cause the Multimedia/DVD Player to run. If not, then how do I get the Event Wizard to run a DVD, and am I correct in my assumptions about running the Flash Game Player with the Event Wizard?
If an event has "Control MMP Player" checked, then it will control the MMP player based on whether or not it meets its condition. For example, Event 5 is often set up to do this, as a convention. Event 5 also enables the Flash Player games to proceed. This is a special property of Event 5. The first 5 events have special uses via the Flash Player.
I tried to document this usage in the document which is online at:
4. If I am training coherence (uptraining with one patient, and downtraining with a second) using the NeuroGuide DLL, is there any way to adjust the thresholds on the fly--or to adjust the thresholds easily when I am not training a patient--or do I have to go into the Event Wizard and change the equations (a rather cumbersome procedure!)?
Yes, you can set up dynamic thresholding via the event wizard, by using as the "Rule" value, the damped value of the training variable. This will produce a floating threshold. See the built-in protocol called "Demo alpha amplitude with dynamic thresholding" for an example. You can also use built-in keyboard controlled variables. For example, the built-in "coherence threshold" which is controlled by the "c" and "C" keys, is called "x=CT" in the event wizard.
Finally, I was talking with Bob Thatcher a couple of days ago (we just went to court together and got QEEG's admitted as evidence in court in Oregon for the first time), and he asked about using the NeuroGuide DLL with BrainMaster. I explained that I like it a lot, but that there is a glitch with the frequency bands. Specifically, the "Standard" BrainMaster band frequencies are not the same as those that are used by the NeuroGuide DLL, so it seems that I have to go through each time I set up a protocol and re-define the frequency bands. Again, an annoying pain. Any chance of getting a second button on the Frequency window for the NeuroGuide DLL? That way you could press either the "Standard" or the "DLL" button and get the settings automatically loaded. Just a thought--it would make our jobs easier.
The standard frequncy bands can be made to match the NeuroGuide bands, but I don't see this as a glitch at all. You don't have to redefine the bands. However, in many built-ins, I have made them match for convenience. But why do they have to match? You can train any bands you like with the NG built-in bands, and use any other bands in the main software. We do this all the time. Yes, I could add a button for "Standard NG Bands" and that is a good suggestion.