PTSD - An Engineering Overview
By John Schreiber

What is Post Traumatic Stress Disorder?  As someone who has gone through PTSD, I am rather amazed that no one seems to know the answer.  All they give is a series of symptoms which describe PTSD.  Of course, if you take the opinion, as Freudian psychology suggests, that everything is either sexual or the fault of the parents, you will never understand PTSD.  If, however, you take the engineering viewpoint and reverse engineer the brain, not only will you understand what PTSD is, you will understand that PTSD is actually a critical survival mechanism.

Before beginning, a discussion is required on the process of reverse engineering.  When creating a new design, engineers are given a blank piece of paper and can make all necessary design choices based on the current state of the art.  To reverse engineer the design, it is crucial to know the state of the art at the time of its design, study what the designers knew, and, using your best judgment, design the system in the same manner the original designers would have.  The intent of this process is not to prove the system is designed exactly in that fashion.  We are trying to model how the design functions and try to understand the thought processes within the design.  In the simple model of the brain I will develop, I will make assumptions in how it functions.  I do not care if that unit has not been identified in the brain because that only means it has not been found yet.

Next, in the creation of the model, there were numerous places where I could have made different design choices.  For the sake of brevity, I will only describe the one I have chosen.  For those who wish to challenge what I have chosen, I urge you to create your own model of the brain and identify each one of the branches.  Then you can examine the choices and make your own decision.

 Having made these statements, I will move on to the creation of the model.  To operate, the human mind must generate appropriate responses to a series of random internal (nerve) and external (senses) inputs.  It must act in “real-time” which specifies a need for determinism (predictability).  (I will use the vernacular of engineering in this document for those who wish to investigate the phenomenon more fully.  If you look at a book on real-time systems, the term used will be “determinism” rather than “predictability”.)  This is crucial to understanding the problem.  A real-time system must be deterministic within the response time required by the system.  What does that mean?

To function, one must be able to predict the operation of the system.  If the system is busy when something else must be done, you do not know if your system will be able to complete the new task.  The conscious cannot control the legs as one walks along because it might be thinking about something else when a leg must be moved.  The delay in stopping the thought process would be too great to prevent the person from falling forward.  However, the conscious is capable of operating in a deterministic manner when baking a cake.  No matter what you are thinking about, the conscious is capable of stopping the thought process in sufficient time to take the cake out of the oven to prevent it from burning.

Note that determinism does not imply fast operation.  A “stochastic” or random system will outperform a deterministic system when performing complex operations.  Deterministic systems break tasks down into small chunks which can be performed rapidly.  This allows them to respond to any new task that comes along.  This constant stopping and starting slows them down.  Stochastic systems perform a single task from start to end without interruption.  For complex problems this is more efficient, but it also means they may not be available when a new task must be completed immediately.

Thus, there are two conflicting requirements for the human mind.  It must be deterministic in order to operate in real-time and stochastic to complete long tasks efficiently.  The mind accommodates these requirements by using two separate components: the subconscious for the deterministic functions and the conscious for the stochastic operations.  Completely separate, the requirement to manipulate the same body means the operation appears seamless.

For simple animals with simple lives, they do not require the conscious.  Since turtles don’t do long division, the need to supply the physical elements of the conscious would waste valuable energy.  Many lower order animals get by just fine with just a subconscious.  Therefore, the model of the brain will begin by containing only the subconscious.

The subconscious must perform problems in real-time.  This dictates a deterministic operation which forces certain design decisions on the subconscious.  The subconscious must take a series of inputs and generate an appropriate response.  The simplest method to allow the subconscious to complete this task is to provide it with a look-up table filled with input patterns matched to an appropriate response.  A telephone book is an example of a look-up table.  One has an input: “Simpson, Homer J.”, and an output: “555-1234”.  It is predictable in that the pattern “Simpson, Homer J.” will always produce the output “555-1234”.  For the subconscious, manipulating the legs could consist of patterns and responses in a table such as this:

Figure 1 Patterns/Response Table (1)

Note that the actual tables within the subconscious will contain very detailed movements of individual muscles rather than the whole leg.  For simplicity, which is also why I refer to the phone book, I am describing the concept at a much simpler level than the subconscious must operate at.

As stated earlier, a real-time system is deterministic within the response time required by the system.  As the need for a faster response time increases, the real-time system must be made faster to adapt accordingly.  Using the example of the phone book, let us say that the maximum time allowable to find a phone number is 5 seconds, but the longest time it will take to find a number in the Springfield phone book is 15 seconds.  To operate within the required time, the Springfield phone book cannot be used.  We need a phone book which is small enough such that it will take, at most, 5 seconds to locate a number.  The need for improved response times is not apparent when using the phone book.  In real life, the subconscious is matching input patterns like:

Figure 2 Patterns/Response Table (2)

The faster you respond to a threat, the less likely you are to be eaten.  The need for predictability specifies the use of a look-up table while the requirement to improve the response time dictates that the table must be as small as possible.  Therefore, a method must be found to limit the table size.

A table which contains only the combinations the animal requires would be the smallest and, therefore, fastest to respond.  Naturally this makes one question what these are.  The simple answer to this is that it will use the patterns it has already used.  But if we don’t know what it has already used, we don’t know what to include.

The requirement to limit the table to include only the combinations the animal will use specifies a period of learning.  When the animal is born, the controller will have the ability to add the patterns it uses into the Patterns/Response table.  To get the animal started, we will only include enough Patterns/Response combinations to enable it to survive to its first meal.  After that, the animal is on its own.

The model of the subconscious will function by having a Pattern Identification Component connected to the nerves and senses providing patterns to a Comparison Unit.  The Comparison Unit will take the input pattern and check it against the patterns contained in each row of the Patterns/Response table.  When the closest match is located, the corresponding response will be sent to the Response Controller which will be hard wired to the muscles to perform the action.  For the purposes of this discussion, the operation of the Pattern Identification Component and the Response Controller are not relevant.  Therefore, it will simply be assumed that they work.

Figure 3 Subconscious Model Elements (1)

Next, I am going to define a rule in the design.  The Comparison Unit must find a match and provide a response when requested.  If no match is found, it must locate the closest match and provide the corresponding response.  To describe this more completely, I will refer to the telephone book example once again.  Let’s say that I want to call Simpson, Homer Joseph, but he does not exist in my phone book.  In that case, I want the response to be the next closest match such as Simpson, Homer J.  Why would I want this?  Why not just return with “no response”.  The answer is that I want my animal to survive.  The actual table will not have names and phone numbers.  It will have entries like “Animal, Bigger than me, Long teeth, Looks hungry” with the response “Fight/Flight”.  I do not want my animal to stand there idly when it encounters “Animal, Bigger than me, Bad breath, Looks hungry”.

In the early stages of its life, the model will react slowly.  Rather than just match a pattern to a response, it must learn what the correct responses are.  Before it can do this, it must identify patterns to place in the Patterns/Response table.  Like a phone book, the Patterns/Response table must be properly organized to ensure the patterns are located quickly.  This means huge gaps must be left between each pattern to allow new entries to be placed properly.  In the phone book if there is no gap between “Simpson, Bart” and “Simpson, Lisa”, what will you do when you meet “Simpson, Homer J.”?  The Comparison Unit must check each row in the table against the signal from the Pattern Identification Component regardless of whether there is any data in the row or not.  That will slow the system down.

As the animal grows, the number of new experiences it encounters will drop off rapidly.  At a certain point, a decision must be made to close the Patterns/Response table to improve the response time.  The period of learning will be allowed for a short time before the table is finalized.  No new entries will be possible after that.  The implication of this design decision is, of course, that we will not be able to teach an old dog a new trick.

For those arguing that animals learn throughout their lives, I will respond that the learning is done in the conscious.  If the subconscious had the ability to learn throughout its life, there would be no requirement for the conscious.  For simple animals with simple lives, the requirement to respond quickly to threats is of far greater advantage than to outwit their opponent.

Improving response time is a critical design feature.  The crucial element limiting the response time is the task of locating a pattern match.  To enhance the reflexes of the animal, the process of locating a match must be carefully engineered.  Enormous reductions in response time can be realized if one can anticipate what the next pattern to emerge from the Pattern Identification Component will be.  That is not as impossible as it may first appear.  When you walk, your left foot always follow your right.  Does your heart not expand and then contract?  Do your eyelids not close and then open?  There are many repetitious patterns in behaviour, and with a simple design change it easy to allow the model to anticipate the next pattern.

A simple example of calling several people in a specific pattern demonstrates how a system can anticipate supposedly random events.  Let’s say that when I have exciting news, I always call “Simpson, Bart” followed by “Van Houten, Millhouse” and then “Wiggum, Ralph”.  These calls are not random.  Initially, the Comparison Unit must start at the beginning of the phone book for each call.  Once the pattern has been established, after I have called “Simpson, Bart”, the Comparison Unit would not need to start its search at the beginning of the phone book.  It could start in the V’s for “Van Houten, Millhouse”.  For Ralph, it could start in the W’s.  The system would respond much faster, and it is surprisingly easy to perform.  All that is required is an “Anticipation” table.

Two new columns are added to the Patterns/Response table.  One will be the row number in the Patterns/Response and the other a reference number to a row in the Anticipation Table.  The Anticipation table holds the next expected Pattern, the row number of the expected pattern in the Patterns/Response table, search window width and a confidence level.  The first time “Simpson, Bart” (pattern) is called, the reference number will be NULL (empty).  While the body goes off to complete the phone call (response), the Anticipation table is queried for the next empty line which, for the sake of argument, will be 678.  When the next pattern emerges from the Pattern Identification Component, the pattern is placed in the Expected Pattern column of row 678 of the Anticipation table.  The search window width will be set equal to the size of the table since we are not confident that a pattern of behaviour will form.  The confidence level is placed at 1.

Figure 4 Patterns/Response and Anticipation Tables Showing Bart Simpson, Millhouse Van Houten and Ralph Wiggum Calling Pattern

The next time “Simpson, Bart” is called, there is a reference number in the Patterns/Response table.  In anticipation of a match, the search will be centered around the “Van Houten, Millhouse” row entry in the Patterns/Response table.  When the next pattern emerges, it is compared against “Van Houten, Millhouse”.  If the pattern does not match, the search window is expanded and the confidence level drops by one.  When it reaches zero, the row is emptied from the Anticipation table and the reference number is set to NULL.  If the pattern matches, the search window is condensed and the confidence level goes up by one.  In another row of the Anticipation table, this process is being done for “Wiggum, Ralph” since he is always called after Millhouse.  By the time the confidence levels are at one thousand, it can be assured that a call to “Simpson, Bart” will be followed by one to “Van Houten, Millhouse” followed by “Wiggum, Ralph”.

The search window width defines the number of rows in the Patterns/Response table to be searched.  As it is condensed, the table is, effectively, smaller.  That improves the response time.  As the confidence level rises, the search window width will become smaller.  Eventually, the search window width is equal to one.  There will be only one entry to search in the Patterns/Response table.  Since it can be the only match, the system will automatically perform the operation without waiting for the next pattern.  This will have a dramatic effect on the speed of the system.  At first, we were searching thousands of row entries to find “Van Houten, Millhouse” and then thousands more to find “Wiggum, Ralph”.  Now we are down to just two.

The same system can be used to reduce the time taken to search the Patterns/Response table for the subconscious.  Knowing that the left leg always follows the right when we walk, the system can narrow its search window in the Patterns/Response table by anticipating the arrival of a pattern indicating the right leg is fully forward.  The response would be to move the left leg.  Rather than search thousands, if not millions, of Patterns/Response combinations, the system will focus on just one.

Moving back to the phone book example for a moment, there are two interesting automated responses that have been generated.  First, if a call is placed to “Van Houten, Millhouse” without a call to “Simpson, Bart”, a call will still be made to “Wiggum, Ralph”.  Second, what happens if we must call “Jones, Jimbo” after “Simpson, Bart”?  The search window after calling Bart is so narrow that the only entry is “Van Houten, Millhouse”.  This will force the system to automatically call the only possible match.  It will still call “Van Houten, Millhouse”. 

A Russian physiologist by the name of Ivan Pavlov used dogs to show what has become known as the conditioned response.  He rang a bell and then dropped a steak in front of a dog.  Seeing the steak, the dog salivated.  He did this repeatedly.  One day he rang the bell but did not drop the steak.  The dog still salivated.

Any systems engineer seeing the model of the subconscious at this point would not be impressed.  Though fast, it will not respond correctly at times and the Anticipation table will require frequent “defragmentation” to keep up its speed.  Because the system constantly looks for patterns, there will be more failed patterns than successful ones.  Huge gaps would appear in the Anticipation table.  The defragmentation process would shrink the table down by eliminating the deleted rows.  It would also arrange the table such that all linked rows are right next to each other in the order they are accessed.  For those who do not understand this process, imagine the Anticipation table as a blackboard.  The constant errors are erased and new entries are scribbled between long established ones.  At the end of the day, it would be pretty messy which would slow down the operation.  It would require reorganizing and a good cleaning to regain its speed.

The need to defragment the Anticipation table specifies a requirement for a period of inactivity in the system.  While it might be possible to defragment on the go, the testing required would slow the system down.  Testing would be required since it is possible to mix reference numbers between an entry coming in and one being moved.  It is faster and far safer to force the system into a condition where few patterns will be generated by the nerves and senses.  The few that are generated would be handled by the thread controller which will be described below.  The controller is then free to defragment without any interruption.  For the sake of argument, this period of inactivity will be known as “Sleep”.

What would happen to our system if it were deprived of sleep?  The system would take increasingly longer times to locate the matched patterns.  This would slow the system down dramatically.  In order to keep response times acceptable, certain functions would have to be reduced or eliminated.  One could infer the mind would have problems focusing on nonessential tasks since the data required for the thought process would be updated at a slower rate.  As the system slowed further, more essential tasks would be dropped such as fine motor control (i.e. the ability to precisely coordinate the movements of one’s limbs).  By turning off fine motor control, the limbs would shake between the limits set by the coarse motor control.  This is analogous to tuning a radio with coarse tuning which only jumps in 1 MHz increments and fine tuning to enable the listener to reach the tenths.  If the desired station is 103.6 MHz, the listener could only jump back and forth between 103 MHz and 104 MHz when fine tuning is off.

The defragmentation process is a risky one since linkages might be lost.  Therefore, a copy of the linkages will be made in a separate table mirroring the Anticipation table in case an error is found.  The defragmentation process is then completed.  Afterwards, a period of testing will begin to ensure the linkages are still valid.

The testing phase will require that the controller’s tables are run as if the patterns were coming in during normal operation.  To prevent the body from moving during this phase, the signal between the Patterns/Response table and the Response Controller will be attenuated until it can barely be detected.  This is like turning down the volume on a radio.  The patterns can then be run to check the links.  If the links fail, the old values are copied from the mirror table and the process repeated.  With the signal to the Response Controller attenuated, it might vaguely interpret the signal as “Dreams” but would not act on the commands.  (For the sake of simplicity, I have not included short and long term memory.  The same process of transferring from short to long term memory would be done during sleep, and the tests would also appear as dreams.)

The full power of the Anticipation table has yet to be realized.  Animals have instincts which are linked series of responses which cannot be accessed in the middle.  How can these be implemented into the design?  The key statement is that they are linked responses.  Therefore, they do not require a pattern to initiate them.  The pattern can be set to “NULL”.  A large number of row entries with NULL patterns matching a response will be placed at the top of the Patterns/Response table.  A rule will be established that the Comparison Unit does not check NULL patterns.  Thus, it is not possible to access the series in the center.  Linking the responses through the Anticipation table allows long strings of responses to be accessed by a single command.  Since the system does not perform pattern tests, there is no way the system can stop the process once it has begun.  These linked series of responses will be called “Instincts”.

Figure 5 Patterns/Response and Anticipation Tables For Portion of Fight/Flight Response

An example of an Instinct is the fight/flight response.  During the youth of the animal controlled by this system, it matches the pattern of a predator with the fight/flight response.  After that, when the pattern of that predator emerges from the Pattern Identification Component, the fight/flight response is automatically triggered.  Adrenaline is automatically dumped into the body to enable it to fight for its life if cornered or to run away at the highest possible speed.  A single response triggers the whole process which proceeds as fast as the body can act on the commanded responses.

The next point to consider is that animals do not always respond in the same manner to a given set of circumstances.  For example, an animal will only hunt when it is hungry.  How does this table system handle that situation?

The Pattern Identification Component is connected to the internal nervous system which would provide information about how full the stomach is.  When the Pattern “Stomach full” is sent to the table, it will set the state of hunger to a value of zero.  As the stomach empties, this value will be increased by a “Set state: hunger +2” response which will add two to the state of hunger in the body.  When a prey item is identified, the response will be to check the hunger value with a response like: “Check state: hunger”.  If the hunger value is small, say less than 10, the animal will not attack.  If it is above fifty, it will attack.  If it is above ninety, it will attack with great vigor.

Figure 6 State Variables or "Emotions"

This leads into the concept of emotions.  What are they?  If I were to develop the model further, which is not required to answer “What is PTSD?”, they would hold values which would modify the response from the Patterns/Response table.  For example, if the “Anger” value were at two, a response would be to hit, but the muscles would not hit hard.  A value of ninety would mean the muscles would hit extremely hard.  Other emotions would modify the responses according to the “emotional” state within the controller.  The creature would not know the value of these state variables.  Rather, they would be felt in various portions of the body to force the creature to act.  For example, an emotion such as “Love” would be felt in the heart rather than as a value in the mind.

At this point, the model can walk, and the model can chew bubble gum.  But it cannot walk and chew bubble gum at the same time.  To account for this, all that is required is a Thread Controller.  This is a device which holds the linkage information for a specific series of Patterns/Responses.  In the vernacular of computer programming, the series of linked Patterns/Responses is known as a “thread”.  When a “leg fully extended” pattern is received, the thread assigned to walking would grab it and pass the data to the Comparison Unit.  When the “mouth fully closed” pattern emerges, the chewing thread would grab it.  If the Thread Controller could hold ten threads, the controller could do ten functions at once.

Most of the threads would be used by “house keeping chores” such as beating the heart and breathing.  The few that remain would be cherished by the system.  Holding onto a thread takes up a valuable system resource.  Being forced to release the thread would cause the confidence levels to drop.  Since the performance advantage by repetition is so overwhelming, the system would want to hold onto the thread as long as possible.  Before releasing it, the system would place a desire in the body to complete the thread.

Figure 7 Subconscious Model With Thread Controller and Anticipation Table

This has an amazing affect on the model.  Imagine one has a cup of coffee every morning.  Just as Pavlov’s dog salivated on hearing the bell, the subconscious is expecting a cup of coffee.  The day the person is not able to have the coffee, the Thread Controller must hold onto the thread until the coffee arrives.  Rather than give up, an urge is placed in the body for a cup of coffee.  The individual now needs a cup of coffee.  Driven by the need for fast response times, the subconscious now actively forces the body into forming habits.

The true model of the subconscious would contain more elements, but this is all that is required of the model for the understanding of Post Traumatic Stress Disorder.  To recap the model, its main feature is the Patterns/Response table which is used to match the patterns from the nerves and senses to create an appropriate response.  The need for fast reflexes forces the table to have as few entries as is possible.  These entries will be filled in during a short learning period in the model’s youth.  To enhance the response time, the table will be closed off preventing the model from creating any more matches.  An Anticipation table is used to increase the response time by anticipating what the next pattern will be.  When patterns are repeated, the Anticipation table is able to close the search window onto the anticipated pattern creating the “conditioned” response first demonstrated by Pavlov.  A Thread Controller allows the model to run several series of patterns simultaneously.  “Emotions” are used to adjust the model’s behaviour depending on various internal factors such as hunger.  A period of inactivity (sleep) is required to defragment the Anticipation table, and the testing done on the links during the period of inactivity would appear as dreams.

Comparing the model of the subconscious to the behavior of real animals reveals many similarities.  The model has a period of intense learning at youth followed by an adult stage where learning is limited to combining responses to create routines or habits.  As the years pass, the process of repeating the same acts on a daily basis produces high confidence levels that the acts will be repeated.  This causes the elderly to be increasingly set in their ways, growing more stubborn as time goes on.  The model’s behaviour can be conditioned.  It requires sleep.  It has instincts, and it will even dream.  Thus, it is a good model for the subconscious.

The intent of creating this model was to identify what PTSD is.  The model is done, and PTSD has not even been discussed.  That is because PTSD does not affect the normal operation of the subconscious.  It is an emergency mode to handle a fundamental logic error in the subconscious.  The Patterns/Response table was closed after a period of learning on the logic that everything required to survive had been learned by that point.  But there is a fundamental logic error in that decision: The model assumes that the animal will have encountered all possible threats to its existence during the short learning period of its youth.

How will this model react when a new predator emerges?  When the predator approaches a herd of animals whose subconscious is a match to the model, the fight/flight response will not be initiated because the pattern which is selected for the predator does not provide the correct response.  The animals will continue to go about their business.  The predator will then capture one, kill it and consume its body.  That pattern will be recognized by the other animals, and it will trigger the fight/flight response.  They will run away.  But the next time the predator approaches, they will not initiate the fight/flight response because the Comparison Unit does not match the pattern of the predator to the correct response.  The pattern continues until all of the adults are dead.

The controller requires an emergency mode of operation which will allow itself to match the pattern of the new predator to the correct response.  This emergency mode will be carried out as an instinct triggered by a single pattern.  What will this pattern be?

Watching a predator kill a companion will induce stress in the animal.  Therefore, a Stress Measuring Unit will be added to the model.  When the stress rises to a threshold value, the unit will send a signal to the Pattern Identification Component which will generate a pattern.  This pattern will select a response that triggers the instinct which controls this emergency mode of operation.

The emergency mode will begin with a period of numbness to allow the animal to escape.  This is analogous to the numbness after a broken leg which allows the animal to run on the leg to escape.  After this, the model requires an interim solution to protect the animal.  There is an unidentified threat which does not have the proper response assigned.  What happens if it will take thirty days for the emergency mode to complete the process of adding the new threat and there are only twenty-nine adults around?  All of the adults will still be dead.

The interim solution must be quick to implement and be removable once the new instinct is included.  I thought about this problem for a while, contemplating temporary response tables and the like.  Then I realized I do not have to initiate a fear response.  All that is required is to make the animal run away.  Fear is not required to do that.  Pain is equally effective.  Amplifying the gain on the circuitry governing emotions to the maximum level the neurons can produce will induce excruciating pain in the animal.  Regardless of the response selected from the table, the animal will be forced to run away in absolute agony.

What effect does this interim solution have on the animal experiencing this emergency mode?  Imagine if I were to strap a set of earphones onto your head linked through an amplifier to a microphone.  This amplifier will increase the volume to such an extent that you will not be able to detect differences in sound level.  A pin drop would sound the same as a cannon.  Every footstep would cause bleeding in your ears.  Even the gentle words of comfort from a lover would cause ear splitting agony.  The net effect of this interim solution is that the agony causes the animal to become fiercely antisocial.  The interim solution drives the animal into seclusion which is exactly what it is supposed to do.

Now that the animal is temporarily protected, the controller requires a means of adding the new threat onto the table.  At first glance, it would seem apparent that expanding the table to add a new threat, as was done in the model’s youth, would be the solution.  However, the model created is a digital representation of an analog system.  In a digital system, this is easy.  In an analog system, this is very difficult.  Also, adding new items now could produce unwanted errors.  The Comparison Unit will find the closest match to an input pattern.  Adding new patterns to the table may affect the way old patterns are matched.  Like the poor “Screamapillar” of the Simpson’s fame, the model could find itself “sexually attracted to fire” or some equally unpleasant outcome.  Therefore, adding to the Patterns/Response table at this stage is too dangerous.  Another solution is required.

What is the easiest method of linking a pattern (new predator) to an appropriate response (fight/flight)?  The answer is simple: conditioning.  But how do you condition the pattern into the model?

The table is deterministic which means the patterns will always match the same row entries.  The memory of the first encounter with the new predator exists as a series of patterns which produced the desired fight/flight response.  Sending these patterns back into the Patterns/Response table will cause the model to relive the horror of the encounter.  It will also set the confidence levels in the Anticipation table to 2.  Run the patterns through the Patterns/Response table a second time will cause the model to relive the horror of the encounter.  It will close the search windows onto the appropriate matches and raise the confidence level to 3.  Running the patterns through the system thousands and thousands of times will close the search windows onto the appropriate matches and raise the confidence levels.  Run it long enough and the pattern generated at the encounter will automatically trigger the fight/flight response.  Ring a bell, and the dog salivates.  Encounter the new predator a second time, and the animal will be conditioned to run away in terror.  It is that simple.

It is important to realize that the system is not “remembering” the event.  It is actually sending the patterns through the system to generate the actual response.  The model will feel the same emotions of the traumatic event time and time again.  In fact, the system will have to “catch” the fight/flight response to prevent the model from panicking.  If it fails to stop the response, the individual will experience a “panic” (also known as an “anxiety”) attack.  The body will respond with the full fight/flight response with no threat nearby.

There is another point to consider.  The interim solution has amplified the level of the emotions.  Now, as the memories are run through the system, the “little” voice is amplified to a point it could drown out a cannon firing.  It will run the passes repeatedly for every second of every minute of every hour that one is awake.  There is some magic number it must reach, and it will continue until it comes to that number.  As it runs its nighttime tests to check the response in the table, the amplified volume can be detected by the Response Controller.  The Response Controller now receives an order to run the fight/flight instinct, and you wake up screaming like you’re back over there.  And that is all people mean when they say the event has left them “traumatized”.

The Thread Controller will have an interesting effect on the conditioning process.  Predictable threads, such as the beating of the heart, can always pass through the Thread Controller because the patterns are easily identified by the controller.  Random patterns pose a problem, however.  In the initial stages, the Thread Controller will not be able to distinguish between patterns belonging to the traumatic memory and those which are sent normally by the nerves and senses.  Therefore, until the memory has been passed through the system enough times to allow the search windows to close on the appropriate patterns, the Thread Controller will be forced to hold onto (or buffer) all random patterns from the nerves and senses.  When this happens, the model is no longer deterministic within the response time required by the system.  What effect will this have?

Figure 8 Subconscious With Thread Controller Buffer

The conscious will be sending data to the subconscious through the Pattern Identification Component, and it will be expecting a response within a certain time.  But patterns not identified by the Thread Controller will be buffered until the fight/flight response has been detected by the Comparison Unit.  This causes a delay between a request and its response.  Imagine a person enduring the initial stages of PTSD who wants to look left.  There will be a delay, sometimes several seconds, before the eyes move.  Another pause will exist before an arm is moved.  This phenomenon is known as a “sense of detachment”.  It literally feels like you are “looking through someone else’s eyes” as one woman described.

Once the search windows have closed onto the patterns, the Thread Controller is capable of filtering out the traumatic encounter patterns from the others.  Thus, there is no need to buffer the unidentified patterns.  The sensation goes away.

When the Thread Controller is able to send all patterns through the system with no delay, the PTSD sufferer is presented with a unique state of mind.  Their mind is processing the normal patterns associated with everyday life intertwined with the patterns of a traumatic event.  It is as if they are living two very different lives at the same time.  And they are fully aware of both.

Remember how making a call to Millhouse without first calling Bart would still cause a call to Ralph?  If one must use the patterns within the memory during the conditioning process, the system must cut the links out of the thread or the confidence levels will drop.  Taking the example of a pilot experiencing PTSD, once the PTSD process is complete, any pattern contained in the new thread will initiate fight/flight.  If the pilot is forced back into the cockpit before the process is complete, the patterns relating to cockpit experience and flying will force the confidence levels down.  Rather than allow the confidence levels to drop, the model must remove the patterns from the thread.  This is not difficult since it only requires changing the reference number.  Thus, the pilot who is forced to fly again shortly after the traumatic event will still be able to fly.  Once the process is complete, the pilot will not be able to fly because the patterns associated with flying will be included in the thread which will initiate fight/flight.

The requirement for the Thread Controller to hold onto patterns has created the need for a new element in the design; namely a buffer.  A buffer is just a temporary storage space for patterns.  It will be a “FIFO” or First-In-First-Out design to ensure the patterns are manipulated in the correct order.  It will be capable of containing only a certain number of patterns which poses another question: What happens when the buffer is full?

When the buffer becomes full, no new patterns can be added and the subconscious will lose information.  When this happens, the process is operating completely randomly which is unacceptable for a real-time system.  To maintain determinism, the subconscious must shut down certain functions.  One function which generates many patterns is the conscious.  By disassociating itself from the conscious, the subconscious is able to maintain control of the body in a predictable manner.  When this happens, the conscious is no longer aware of the body.  For those who do not believe this condition can exist, consider “sleep”.  The conscious is not aware of what the subconscious is doing during sleep except for the occasional dream.

I have experienced this condition which is best described as “blind rage”.  I would be walking along consumed by the rage generated by the event and then I would be several kilometers away having crossed several major highways.  Scary at first, I came to accept that my subconscious was following the correct procedure for crossing roads.

PTSD has a number of other symptoms associated with it such as night sweats.  These are not felt by all who have endured PTSD.  I did not have night sweats or flashbacks for that matter.  I did have severe bowel cramps, though.  These additional symptoms vary from case to case depending on the nature of the patterns in the traumatic thread.  I did not fear for my personal safety during the event which triggered my PTSD.  Thus, my subconscious did not accidentally trigger the mechanism which must exist to awaken an individual to possible danger while sleeping.  Therefore, I did not have flashbacks.  The bowel cramps were a direct cause of the rage I experienced during my trauma.  Thus, the nature of the incident triggering PTSD will have an effect on some of the symptoms.

Some of the symptoms described above may not be felt by everyone going through PTSD.  For example, if the series of patterns in the traumatic thread is very short, the thread controller will be able to distinguish which patterns belong to the thread quite rapidly.  The individual may not be aware of the “sense of detachment” that others may feel for quite some time.

And that is all PTSD is.  Post Traumatic Stress Disorder is an emergency mode within the subconscious designed to generate an instinctive response to a new threat to its existence after a single encounter.  Despite its horrid reputation, on its own, PTSD will not take an exceptionally long time to run its course.  Its effects can be easily lessened through constant exercise.

How does this work?  The whole key is the Thread Controller.  If one lays down on a couch, probably about half of all patterns will be devoted to the traumatic episode.  By moving ones arms and legs about, there are more patterns for routine motion to pass through.  In this case, probably ten percent of the patterns going through the system are from the traumatic episode.  Movement dilutes the traumatic event.

If this is the case, then why are so many people’s lives dramatically altered by PTSD?  To answer this question I will use my old analogy (from which this document grew) that PTSD is the mental equivalent of a broken bone.  On its own, a broken bone will take only about two months to heal.  When it takes longer, the problem must be investigated.  Imagine the person so overwhelmed by the pain of the broken bone that they keep hitting the bone.  This keeps breaking the bone again, and it will never heal.  And that is one reason why some people’s lives are so dramatically altered by PTSD.

To explain this further, I must introduce the conscious into the model.  The need for the conscious grew out of the deficiencies in the subconscious.  A major deficiency was the inability to learn after a short period in its youth.  Rather than slow the response time of the subconscious by allowing it to learn more, a completely separate component was created.  In many respects, the conscious is the exact opposite of the subconscious.  While the subconscious is predictable (deterministic),  the conscious responds randomly (stochastically).  They are completely separate entities with separate agendas.

A detailed model of the conscious is not required to explain the problem.  In fact, no model is required at all.  To understand the difficulties some people have with PTSD, one need only accept that the conscious has the ability to induce sufficient stress to trigger the PTSD response.  This is done through a variety of means.  The simplest to explain and easiest to overcome is “feedback”.  (Technically “positive feedback” but I will shorten the term to “feedback”.)

Before explaining this further, I would like to point out to anyone who is familiar with the US Army’s Bulletproofing program that this is the portion of the whole puzzle their program focuses on.  They reduce the chance of their soldiers having PTSD by teaching them what to expect.  This reduces the chance they will allow stress to build when confronted by the signs of high stress.  The concept is to keep the stress below the threshold which triggers PTSD.  It will fail when, independent of how the conscious reacts, there is sufficient stress generated in the subconscious to trigger PTSD.

When a microphone is placed in front of a speaker, the rapidly building sound level is due to feedback.  An example for describing the effect feedback has on the mind are panic attacks.  Note that I am also providing this explanation to illustrate how they can be controlled.

Every now and then during the process of pushing the patterns associated with the traumatic event through the system, the fight/flight response is not prevented from being acted upon.  When the fight/flight response is initiated, a mechanism releases adrenaline into the body to give power to the muscles to either fight or run away.  This is not a switch but a valve which regulates the flow of adrenaline depending on the amount of stress felt by the individual.  During the actual event, the conscious was focused on the event.  The focus was external to the body.  The mechanism released the amount of adrenaline appropriate for the stress level generated by event.  Too busy at the time to notice, the conscious was more than likely unaware of the increase in adrenaline.  And if aware, the extra energy was welcomed.

When the fight/flight response is triggered accidentally, the adrenaline valve opens to give the body energy.  But there is no external threat for the conscious to focus on.  The traumatic event becomes the surge of adrenaline in the system.  This increases the stress.  The mechanism controlling the valve senses the additional stress and opens the valve further to give the energy appropriate for the stress level.  The conscious detects the additional adrenaline and panics even further.  The stress rises causing more adrenaline to be released.  Eventually, the valve is opened as far as it can be.  The conscious is in full panic now with the heart beating furiously. 

The panic attack was caused by an inadvertent triggering of the fight/flight response.  It grew in intensity because the conscious focused on the level of adrenaline in the system.  This caused the feedback effect that fully opened the adrenaline valve.

To prevent panic attacks from growing, my immediate response is to divert my attention externally.  I open my jacket if I am hot.  I will bend down, untie and tie my shoes again.  If in the car, I will change radio stations and adjust the climate controls.  I ignore the adrenaline in my system.  After using these techniques long enough, I was able to stop a panic attack by merely saying the alphabet backwards.  After having hundreds of panic attacks, my mind is now conditioned to respond with diversion techniques automatically.  It has been so effective that I am no longer plagued by panic attacks.  This is either because I no longer experience them or my subconscious is conditioned so finely that they are stopped before I can detect them.

In the same manner, someone experiencing PTSD can be overwhelmed by the symptoms.  The anxiety over what the body is doing can easily cause stress in an individual.  During this time, the subconscious is generating huge amounts of stress as it replays the traumatic event.  When the conscious focuses its attention internally on the stress, it generates its own stress.  When the sum total of the stress in the conscious and subconscious exceeds the PTSD trigger threshold, the response begins again.  This resets the system and the whole process starts over.  Since triggering the PTSD response returns the mind to a state emotional numbness, I called this cycle “powering down”.  Everything shut down and I was left with the complete emotional silence of “the void” as I referred to it.

This is counterproductive since the subconscious must perform the PTSD operation from beginning to end with various tests which it must pass in the middle.  For example, if one is playing the “what if” game, the memory will be changed slightly.  The patterns will not pass through the system to generate the fight/flight response causing the subconscious to fail that pass.  This creates stress in the subconscious which is felt in the conscious as anger.  The stress rises and, if it rises too much, will trigger the PTSD response.

To prevent the PTSD response from triggering again, the individual must ignore the subconscious as it goes about its business.  This is like ignoring a bad headache which I am now able to do as long as no one asks how my headache is.  Then my attention returns to it, and I feel it again.  The whole trick is to relax and get huge amounts of long term, low intensity exercise.  The key is to keep the arms and legs moving.  Since it is normal for the conscious to want to review the event, a means must be provided to do this.  There is nothing wrong with reviewing the event as long as it is done externally.  This means writing the experience down (pen on paper to force the subconscious to carefully control the movement of the fingers) in the initial stages.  After that, verbal discussions can be done in a support group.  Never, ever attempt to interfere with the process internally since this will elongate the process.

When discussing the event, there is nothing wrong with becoming emotional.  If you must cry, then cry.  This does not signify a relapse.  Just stop telling the story to reduce the input of stress and allow your emotions to settle.  I find comfort in the tears since they are a sign that I am still human and not some emotionless machine.

Another form of feedback is peer pressure.  The individual is pressured to put the incident behind them.  The person is told to get back onto the mythical horse that roams about somewhere.  (I could never find it though I did locate the odd jackass or two.)  When the individual is not able to do this, the belief is that the individual is not trying.  In my case, I was ordered by a higher ranking officer not to let the incident bother me.  He questioned why I was disobeying his order when I still had problems.  This type of pressure is as counterproductive as forcing and individual with a broken leg to run a marathon.

With feedback, the conscious is responding to the symptoms of PTSD to increase the stress and elongate the healing process.  The conscious can have another effect on PTSD, and it is this effect which has given it the fearsome reputation that it has.  The conscious has the ability to, independent of any stress in the subconscious, build up sufficient stress to trigger the PTSD response.  When triggered in this manner, the common term is a “nervous breakdown.”  Individuals who cannot control the stress in their lives trigger the PTSD response without any traumatic event.  The subconscious will then cycle the series of incidents which built up the stress through the system generating growing paranoia.  I have talked with individuals who have had a nervous breakdown.  (The woman who claimed it was like “looking through someone else’s eyes” had suffered a nervous breakdown.)  What I found remarkable were the similarities in their experiences and what I endured going through PTSD.

The same phenomenon can happen to those going through PTSD.  If the conscious has difficulty dealing with the event which triggered PTSD, it can build up enough stress to trigger PTSD again.  If the event did not produce sufficient stress to trigger PTSD initially, the stress generated within the conscious can raise the level high enough to trigger the PTSD response.  Thus, the conscious can trigger PTSD months or years after the actual traumatic event.

This phenomenon is the reason why American Vietnam veterans has such difficulty with PTSD.  They responded to the call of duty as their fathers had done in the Second World War.  They followed orders just like their fathers had done in the Second World War.  They watched their friends die just as their fathers had done in the Second World War.  But, when they returned, they did so one at a time rather than with the regiment.  They were spat upon in the airport.  The mutual support of their comrades was gone, and they did not get to see images of concentration camps which would allow their conscious minds to accept their experience as worthwhile and necessary.  They wondered if they had done the right thing.

This created conflict in their conscious minds which built up stress.  The stress rose until it was sufficient to trigger the PTSD response.  Their subconscious’ went back through the memories to locate a traumatic event and fed the memory patterns through the system.  Thus, they believed the problem arose due to the traumatic event.  In fact, the PTSD response was triggered by their inability to deal with the stress in their lives.  The subconscious just grabbed hold of the traumatic event believing it was the reason the PTSD response was triggered.  In this case, the conscious was no longer responding to the stress as in the case of feedback, but independently generating its own.

The veterans of the Second World War cannot claim they went through the same experience as those in Vietnam.  Their minds were able to heal because they had the support of the community.  The veterans of Vietnam were affected so dramatically because they did not have the support of the community.  And that is the same condition problem facing Canadian soldiers sent on peacekeeping missions.

Imagine a situation in Canada where a young woman is being held in a house by a serial killer.  Her screams as she is being raped can be heard on the street.  The police are called, and the SWAT team surrounds the house.  They wait until her screams are choked out of her.  The SWAT team continues to wait outside until they see the serial killer exit the back door.  Then they approach the house with cameras to document the incident.  Pictures are taken and turned over to the prosecutors.  When the serial killer is picked up, he is brought to trial and forced to pay for his crime.

This scenario would never happen in Canada.  The public outcry would ring across the country.  Canadians would demand to know why the SWAT team was not sent in to save the woman.  Heads would roll, and careers would end.  This situation would not be accepted in Canada.  But that is the exact situation Canadian soldiers sent on peacekeeping missions were forced to confront.

Canadian soldiers sent on peacekeeping missions in the former Yugoslavia and central Africa did not have the equipment or manpower to stop the ethnic cleansing taking place.  Confrontation would have caused the wholesale slaughter of Canadian units.  The government, basking in the accolades of its peacekeeping tradition, fiercely protected the “everything is good” image when the situation was clearly out of control.  The soldiers were neither withdrawn to safer locations or sent in to put an end to the ethnic cleansing.  They stayed as witnesses to the atrocities.  Those who had joined to serve as “peacekeepers” had no problem with this situation for they knew that the documentation would be used in the final war crimes trials.

But those who had joined to serve as soldiers only saw that the lives of innocent men, women and children were being sacrificed to protect their own lives.  In the conscious minds of these men, this was unacceptable.  They would rather have gone into combat, placing their own bodies between the antagonists and the victims to put a stop to it even if it meant they would die.  And it is that conflict which now churns in their minds, raising the level of stress and triggering the PTSD response over and over again.

I do not believe in the concept of the broken soldier.  These soldiers have to understand that the failing was in the Canadian population rather than them.  They had the courage – the courage in spades – to try to end the ethnic cleansing.  But we would not let them because we did not have the courage to face the trauma of their deaths.  The weakness was in us, not them.

Can they ever be healed?  Perhaps not, but conditioning is a powerful tool which can be used to prevent memories from resurfacing.  This is not equivalent holding them inside.  These memories must be dealt with externally, preferably within a support group.  Over time, as long as they do not build anger against the memory, the subconscious can be taught that bringing back those memories only causes pain.  At that point, the subconscious must decide whether the cycle of pain needs to be repeated.  When the subconscious knows there is no answer, only pain, there is no need to run the memories again.  The men can live in peace again.

John Schreiber

Author’s Note:  This document was written to assist Canadian soldiers who are enduring PTSD.  It may be reprinted without the author’s consent or compensation by any organization or person who is seeking to help those enduring PTSD.  The author may be reached for comment at: mailto:johnschr@sympatico.ca

Some of you who don't have any background in technical fields have read the above article and wondered what was he talking about. Because he has gotten so many email he has also written the same article in common terms which can be found here.