The Fallacy of Training Machines

[Jitsuman]

^^^ Ronin makes sense.

Short post.
Makes sense.

What a guy.

[EnaeS]

Yes, state-dependent memory, learning, and behaviour (SDMLB) is a bandwidth issue and involves all encoding systems: cognitive, molecular, behavioral, socio-cultural.

If people get down to the 'irreducible' aspects of information processing, they meet (metaphorically) two parallel gates: data-stream and hypothesis stream.

Data refers to sensory data alone - its immediate recording and translation into representation centrally (CNS). Data driven procesing is slow and reactive, its behind the curve of events. Hypotheses processing refers to the cognitive interpretation of the data, and, this always involves an overlay so that the data is modified. Human intelligence is characterized by the ability to bandwidth hypotheses (on the positive side) and on the negative side, to over generate them - which is the origin of both ilusions and delusions rspectively. Nevertheless, hypotheses driven processing is fast, rather than slow, and hs teh potential to be ahead of the curve.

Ideally, both gates/pathways work together so that raw data is interpreted off from, in such a way as to allow the most efficient predictions of real-time and future events to be made. This is the only way we can avoid being a reactive 'nerve-net' and be overwhelmed under pressure simply by the progrwession ofevents.

Any decent entrainment in MA will include both methods (of course as its the normal baseline for cognitive functioning). The 'trick' however is how to manage these gates, and avoid the slow-reactive pitfall, or, the abstract overgeneralization trap that comes with cultural overriding of 'natural' hypotheses development.

The place to start is with data-processing, because this is raw and immediate, and far more elementary than any sbsequent hypothesis overlay. An important quetsion is "what is the minimal amount of data input necessary to shape the human form, and its motion". If you have this, then you have the basics of a refined map that carries minimal loading. The lower the load the more spare capacity is available - and spare capacity = time when under pessure.

This goes back to the 'feature detectors' in the previousp post. Frogs have no flexible 'hypothesis' gate, its either 'on' or its 'off', so their behavior is very limited.

When the data stream is minimally (but effectively) modelled, its time to build up the hypotheses stream, as this is the leading aspect of the battle-computer (by analogy) program.

Good hypotheses in combat, come from a baseline working knowledge of the human form - its motion characteristics, its relative position to map points (on itself), and 'probablility computations' about the bandwidth of potential available in any aspected structural orientation (relative position). If you can get this, you can jump ahead in time and generate even more spare capacity - and again spare capacity is speed defined as rate of change of structure in the 'intended' direction of the goal (completion). Completion is usually defined as 'finishing the 'target' - but this is context specific.

The 'goal' should be present at he beginning, and guide all action towards itself (teleologically). Error-correcting feedback can modify action along the way, but still be directed by probablitity computations. If not, the program can get distracted, RAM fills up, people panic, gas-out, get hit, run-away or whatever.

The human form is the key, the process must deconstruct it.

this is what i gathered from your post, am i on point?

mappping: learning how the human body works and moves. what attacks can come from were and at what angles and ranges, how it balances and how to disrupt the motions of the body

data stream: sensory input, slow, real time/reaction to opponent, more thinking required than hypothesis stream

hypothesis stream: prediction of movements, faster and less thought needed than data stream, predictions based on previous experience and in the moment experience, predictions become more accurate/requires less data as experience is built

[Torakaka]

I think that, IF you use equipment in a "dead and static" state ( just punching a punching bag like most do), while you MAy develop some skills, you are short changing your development.
Of that there is no doubt.
There is also no doubt that the vast majority don't use equipment to their full potential.

I can see something like this:

A typical workout consiting of warm ups and such, followed by:

HB work to work on your "punching power" and combinations - 5% or workout time
Pad work to work on your cardio and speed - 5%
Mitt work to work on your hand-eye, foot work, timing and such - 5%

Sparring ( contact to be dependant on level and what is being trained) to put it ALL together. - 30%


I think that, developing the skills that equipment were designed for and then applying them in the SAME workout, gives you the best of both worlds.
You work on your kicking "power" without having to crush the legs and body of your partner, yet you still apply it on a living, breathing, reacting, person after.

what about the other 55%?

[Steve Richards]

this is what i gathered from your post, am i on point?

mappping: learning how the human body works and moves. what attacks can come from were and at what angles and ranges, how it balances and how to disrupt the motions of the body

data stream: sensory input, slow, real time/reaction to opponent, more thinking required than hypothesis stream

hypothesis stream: prediction of movements, faster and less thought needed than data stream, predictions based on previous experience and in the moment experience, predictions become more accurate/requires less data as experience is built

Thats it, bang on point.

Steve.

[Steve Richards]

I think that, IF you use equipment in a "dead and static" state ( just punching a punching bag like most do), while you MAy develop some skills, you are short changing your development.
Of that there is no doubt.
There is also no doubt that the vast majority don't use equipment to their full potential.

I can see something like this:

A typical workout consiting of warm ups and such, followed by:

HB work to work on your "punching power" and combinations - 5% or workout time
Pad work to work on your cardio and speed - 5%
Mitt work to work on your hand-eye, foot work, timing and such - 5%

Sparring ( contact to be dependant on level and what is being trained) to put it ALL together. - 30%


I think that, developing the skills that equipment were designed for and then applying them in the SAME workout, gives you the best of both worlds.
You work on your kicking "power" without having to crush the legs and body of your partner, yet you still apply it on a living, breathing, reacting, person after.

I can't disagree with you, only suggest an additional perspective.

[Ronin]

what about the other 55%?

warm ups and cool downs of course.

[Ronin]

this is what i gathered from your post, am i on point?

mappping: learning how the human body works and moves. what attacks can come from were and at what angles and ranges, how it balances and how to disrupt the motions of the body

data stream: sensory input, slow, real time/reaction to opponent, more thinking required than hypothesis stream

hypothesis stream: prediction of movements, faster and less thought needed than data stream, predictions based on previous experience and in the moment experience, predictions become more accurate/requires less data as experience is built

Sparring defined.

[dramaboy]

Maybe its just me, but this seems over complicated for something as simple as fighting.
Or maybe its just they way you are trying to explain it.

I think he has combat situations in mind, rather than simple "fighting".
I think most of this doesn't matter if you want to train for comps/fun, even pro.

Tomas

[JFS USA]

Sparring defined.

Sparring honestly, with integrity defined.

The whole notion of being able to "get there from here" plays out as a defect across many parameters. Braketed at the extremes Form Fairies and Machine Slaves.

I don't find the continually raised "problems" associated with a sparring dominant approach, e.g., injuries, to be very compelling. It is far more a matter of quality vs. quantity and cycling helps to provide ample recovery time.

IMO, adjunct practice should be focused almost exclusively on honing of skills with some "fun" tossed in for the novelty factor. Good for the Mind & good for the Body.

[Omar]

Quoted from another board:

Step-1
Investigate and explore structure of the skeleton and connective tissues by applying pressure and impact in such a manner as to create a relationship between structure and impact.

Progressive and percussive impact is best applied on a shortened stroke at first. This shortening of the stroke allows for minimal/manageable energy to be generated while the curves and straights are sorted out.

Curves are ideal, but straights are more realistic for most people when they begin playing with impact without prior experience in getting hit.

Most people will have at least one curve that they naturally gravitate toward. The coach/facilitator must be very attentive to this since it's a point of entry to tjq's broader set of curves that might otherwise be overlooked or actually discarded/denatured - a casualty of a generic/un-natural method being adopted.

Straights are very apparent and easily understood since they deal with the mechanical structure we all share rather than the unique intrinsic structure of the individual.

Begin this investigation by placing a palm on the forehead of your partner and have them neutralize the pressure you apply by aligning their leg, back, and head so that the pressure is felt in the heel.

Once this line is understood, roll your palm on their forehead so that the pressure attacks the structure from slightly different angles. Do the same exercise with your palm on one of their shoulders, their chest, hip, and knee. This is where you begin to understand a person's natural curves and their basic peng.

Begin working with progressive impact. That is, impact that is applied slowly and increases in pressure but not speed. The same line is established in each of the positions as were done with the palm, but now you're putting your fist against them. Be sure that your own structure replicates that of one being applied to punching-proper. This is important.

When you start to incorporate percussive hitting, you can broaden the tactical spectrum by 'hitting' with your fist, shoulder, elbow, foot, and knee. Again, it's important to replicate the structure and motion which is present with the efficient use of each weapon.

I didn't ask him if I could use his post so I am leaving his name off. But I think the content is relevant to this discussion.