Val,

 

I've been reading on the subject a little, so I was wondering if your NLD approach necessarily has to be based on some kind of normative EEG data.  In other words, how do you know that neurology is getting "better".  Also, do you train with rewards and inhibits?  I can understand regular EEG training somewhat, because it is fairly straightforward.

 

John Hooper

 

On 3/31/08, Val Brown <val@...> wrote:

John:

I'm not sure what you mean by "handle". Do you mean "predict"? If you mean predict (as
in what is likely to work well, not well, etc) NLD can give very good predictions and,
basically, what you've said descriptively is what can be given in more particular
mathematical terms. One can derive probable values for "too much" or "too little" of
various of the "elements" added in at various times -- and the time component is critical
because of (among other things) that issue of sensitive dependance on initial conditions.
When things are done and in what sequence they are done can all be critically important
for the actual outcomes observed.

It is also generally easier to destroy than it is to create, esp in the specific case. I can tell
you what to do to make absolutely certain that roses will NOT grow in a certain area. I can
also tell you the things to do to make if much more likely that roses will grow but I don't
know of anyone who tell you precisely what to do to make a particular size of bloom
happen at a precise junction of a certain "branch" of the rose bush.

val

--- In neurofeedcommunity@yahoogroups.com, "John Hooper" <jwhooper@...> wrote:
>
> >
> > If neurofeedback were linear we would expect to see linear graphs for
> > outcomes and
> > responses within training sessions themselves. We don't see this. We would
> > also expect
> > to see that clients should respond in an enormously different way when
> > given "general"
> > training as opposed to when highly specialized training is given. *Condition
> > A should
> > NECESSITATE approach A and EXCLUDE approach B, C, and D but also should
> > exclude the
> > use of a general approach as a way to address a specific problem.* We also
> > don't see this.
> > So called "general" approaches work rather well despite the presumption of
> > "single specific
> > disorder, single specific etiology, single specific intervention".
> > Moreover, specific
> > approaches have general effects as well -- also not what we would expect
> > for a linear
> > process.
>
>
> Thank you for clearing up one of the great sources of confusion and
> annoyance I had in years past. Different, apparently successful,
> practitioners often gave contradictory advice for how to train for the very
> same disorder. All argued vehemently that the results spoke for
> themselves. I must admit, at the time I wrongly stated that BOTH protocols,
> being opposite, could not be effective if NF training actually worked, which
> I began to doubt.
>
> So, whether you start shoveling on the right or the left of a sandpile, it
> doesn't really matter. In the end, you get rid of the sand. When the sand
> is gone, the grass underneath can grow again. Leaving a foot, or even a few
> inches of sand will not get the job done. The grass growing and the size of
> the sandpile do not have a linear relationship.
>
> Of course, in this simple example, almost any protocol will do. Add
> sunlight, fertilizer, and water to the equation and making the grass grow
> isn't perfectly straightforward, even with the sand gone. There could be
> too much water, sun, or fertilizer as well as too little.
>
> How does non-linear control handle all of this?
>