This article will help to answer questions that have been asked, regarding the BrainMaster implementations of Live Z-Score Training (LZT).  To clarify, it should be understood that the use of the BrainDX normative database for Live Z-Score training is a BrainMaster implementation, and is not being deployed by BrainDX.  The manner of use of the BrainDX database for Live Z-Score training is essentially the same as the use of the ANI database.  This includes the use of BrainMaster’s proprietary multivariate “PZOK” method, the Z-Plus extensions, and related capabilities.

It is true that the BrainDX database is based upon static norms, while the ANI live database is based upon dynamic norms.  The result of this is that the BrainDX Live Z-Scores are typically larger in magnitude than the ANI Live Z-Scores, but have the same direction, and both databases agree on what is “normal.”  One advantage of using the static norms is that the maps and z-scores seen during training now match the assessment data and maps, and no conversion between the two sets of norms is necessary.  This eliminates a long-standing confusion and need for mental arithmetic that has been required when assessing using static norms, and training using dynamic norms.  It should also be noted that, because the established method of QEEG-based neurofeedback is to inspect static maps to determine the training requirements, and also to use static maps to assess outcomes, that using the static maps as a reference for training makes sense and does not violate any principles of statistics or practice.  It is, in fact, more seamless and intuitive than using one set of norms for assessment and a different set for training.  It can be confusing, for example, when a focal excess at 1.4 standard deviations on the assessment “disappears” in the live z-scores because it falls below the limit of being considered deviant.  Using static references for live z-scores eliminates this issue, as is shown in the accompanying article comparing maps of each type.

These new methods did not come out of thin air.  They were developed to answer specific questions:

“Why can’t my training z-scores look like my assessment z-scores?”  The answer is, they can.  Using static norms for training achieves exactly this.

“Why do I have to train to a population?  Why can’t I train to an actual EEG of someone with an actual brain, or use z-scores to monitor my client’s progress?”  The answer is, you can.  Using Z-Builder achieves exactly this.

The related articles cited below will help to clarify this situation.  In addition, the illustration below makes it clear that both the BrainDX and the ANI databases, as well as BrainMaster’s “Z-Builder” approach, are complementary methods, that can be used at the discretion of the users.  It is not true that using a static norm in real-time is like comparing “apples to oranges.”  Rather, it is like comparing an apple to a bag of apples, when you know exactly how many apples are in the bag, and you also know that they are all the same kind of apples, and of comparable age.  Moreover, the static norms can be derived from the same data used to derive the dynamic norms, by basic mathematical principles.

An explanation and illustration of how static and dynamic z-scores relate, for population and for individual, can be found at:

http://www.brainm.com/kb/entry/513/

The strong correlation between JTFA and FFT results can be found at:

http://www.brainm.com/kb/entry/537/

Comparison of maps can be found at:

http://www.brainm.com/kb/entry/525/

A demonstration that the JTFA data are Gaussian distributed can be found at:

http://www.brainm.com/kb/entry/491/