ACCUMULATED STRESS, RESERVE CAPACITY, AND DISEASE
Peter A. Levine
ABSTRACT
The underlying theme of this paper
is that the accumulation of stress affects the reserve capacity of an organism,
both in the maintenance of its functional integrity and in the resolution of
subsequent exposures to stress. Stress
is defined in terms of a reaction resulting from stimuli which sufficiently
activate the autonomic nervous system (ANS) and is either resolved or
accumulated depending on whether the pre-stimulus baseline is re-established or
not.
Accumulated stress profoundly
influences the totality of organismic functioning, and is expressed essentially
through three bi-polar effector systems: In the realm of the autonomic, the
effector system is the sympathetic and parasympathetic visceral outflow. For the somatic, it is paired movers, like
extensor/flexors; and metabolically stress is expressed (though less
distinctly) by, for example, catabolic/anabolic and inflammatory/anti-inflammatory
endocrine reactions.
The
response to stress is defined as occurring sequentially in tow phases, chare
and discharge: When the charging
(sympathetic) phase is followed by parasympathetic discharge of equal
magnitude, then pre-activation homeostasis is reestablished and the stress is
said to be resolved. On the other hand,
it is shown that under certain physiologic conditions) and behaviorally where
mobilization – i.e., somatic response to stress—is blocked), the charge phase
is no longer balanced by rebound. In
these cases activation is not resolved and the stress becomes incorporated
within the organism, as a diminished adaptational capacity.
The basic physiologic relations of
the autonomic, sympathetic and parasympathetic, can be
represented by a simple mechanical analogy (the “Zeeman Machine”) which
exhibits properties described by a relatively new branch of mathematical
topology, Catastrophe theory. The
visualization gained by this re-presentation offers new
insights into the nature and mechanisms by which stress accumulates. It also suggests ‘paradigms’ by which stress,
once it has already become internalized, may be successively resolved towards
re-establishing a fuller adaptational range/reserve capacity.
In this regard, various holistic
systems of healing are seen to focus their efforts towards detecting and
treating these accumulation imbalances and reduced capacities even before they
become symptomatic and pathologic. It is
the view of this work that a wide range of “stress diseases” with varied
symptoms and obscure aetiologies are the final—pathologic—expression of this
loss in resiliency.
That the accumulation of stress is
the underlying stratum in certain disease syndromes is tested by measuring
autonomic levels underlying certain blood pressure responses of a hospitalized
population. It is not possible, however,
to measure the sympathetic and parasympathetic components directly (since they
are expressed as a singe output vector, blood pressure). For this reason a systems analysis of the
cardio-vascular system, based on well-known experimental parameters, but with
variable set point and gain levels, is constructed. A set of blood pressure response cures is
generated and compared with the hospitalized population. The fit of these with the experimental data
is surprisingly good. In addition, the
prognosis for five groups in the hospitalized population is predicted
accurately by the model, whereas no such predictions could be made on the basis
of the raw data.
The accumulation of stress, defined
in terms of the autonomic nervous systems.
The concept of an autonomic hypothalamic “hub” around which behavior is
organized and executed is illustrated to clarify some of these extended
relationships. Specifically, the
hypothalamic links between autonomic-endocrine, as well as somatic mobilizing
systems, are examined in the context. In
addition, examples illustrating the potential for the wide and varied
symptomatologies of their “mis-integration” (auto-nomic-endocrine-somatic) in
the stress diseases are presented. Some
possibilities for pre-symptomatic diagnosis, whereby stress accumulation is
detected before the development of debilitating symptoms and tissue
pathologies, are investigated as well.
These stress diseases are shown, in a selected set of examples, to have
underlying patterns of unresolved stress that can be understood in terms of
their topologic configurations in catastrophe space.
ACKNOWLEDGEMENTS
My
heartfelt appreciation to a number of persons who shared freely their time,
resources and energy: To Dixon Jones,
Department of Animal Ecology at the
To my committee
members, both for their encouragement and criticisms, sometimes harsh and
jarring, but always a stimulus to gradual maturing.
And especially to my friends,
without whose support, no way, could this thesis have
been completed. And to John M., Marsha
D. and Megan H., to whom it is dedicated, thanks.
This investigation was supported, in
part, by USPHS training grant #5T01GM00829 from the National Institute of
General Medical Sciences.
TABLE OF CONTENTS
Part
I. Accumulated Stress Page
Section A) Autonomic Stress: Introduction and Definitions . . . . . . . . . . . . . . . . 1
B) Catastrophe Theory. . . . .
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .18
C) The Model. . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . . . . . 32
D) Predictions of the Model. .
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
E) Model Applied to Hospitalized Population. . . . . . . . . . . . . . . . . . .
. . .53
Part II. Applications of the Model
Section
A) The
treatment of accumulated stress:
Introduction. . . . . . . . . . . . . . . .70
B) Holistic Approaches to
Integrative Medicine. . . . . . . . . . . . . . . . . . . . 74
i.
Acupuncture therapy. . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . 76
ii.
Body Structure Approaches:
Alexander and Structural Integration (Rolf). . . . .
. . . . . . . . 79
iii.
Respiratory Vegetotherapy (Reich). . . . . . . . . . . . . . . . . .
. 92
Part III. Organismic Effects of Accumulated Stress
Section A) Anatomic substrata: The hypothalamic hub. . . . . . . . . . . . . . . . . . . . . 96
B) Autonomic endocrine
relationships in accumulated stress. . . . . . . . . .110
C) Autonomic-somatic
relationships. . . . . . . . . . . . . . . . . . . . . . . . . . . . 111
i.
Autism. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . 114
D) Mobilization
(autonomic-somatic discharge). . . . . . . . . . . . . . . . . . . . 117
Part IV. Stress Disease
Section A) The Stress Diseases: compendium and evidence for accumulated
Autonomic states (as predicted by model). . . . . . . . . . . . . . . . . . . . . .130
i.
Nomenclature. . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . 130
ii.
Hypertension. . . . . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . 135
iii.
The Ulcer. . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . 136
iv.
Vasodepressor Syncope. . . . . . . . . . . . .
. . . . . .. . . . . . . . . 137
v.
Anxiety States. . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . .138
vi.
Anorexia Nervosa. . . . . . . . . . . . . . .
. . . . . . . . .. . . . . . . . .138
vii.
Role of Stress in Primarily Infectious Disease. . . . . . . . . . 141
viii.
Aging and Stress. . . . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . 142
ix.
Hyperventilation Syndrome and Clinical Effects. . . . . . . . 143
x.
Childhood Autism. . . . . . . . . . . . . . . . . . . . . . . . . . . .
. . . 146
B) Pre-symptomatic Diagnosis and Preventative Medicine. . . . . . . . . . .148
TABLE OF CONTENTS (cont.)
Page
Part
V. The Servo Analysis of Cardiovascular Dynamics
Section A) Cardiovascular control. . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . .
. . .154
B) Cardiovascular systems
simulation and isolation of sympathetic/
Parasympathetic autonomic components. . . . . . . .
. . . . . . . . . . . . . .164
C) Discussion of simulation. .
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 182
Part
VI. Health, Disease and Integrative
Medicine, Epilogue and Conclusions 203
Appendices
i.
Mechanisms of Acupuncture. . . . . . . . . . . . . . . . . . . . . . .
. . . . . . 208
ii.
Hyperventilation and vegetotherapy. . . . . . . . . .
. . . . . . . . . . . . . 215
iii.
Background Anatomy and Physiology of the
Hypothalamus and Pituitary. . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . 220
iv.
Autonomic Endocrine Relations. . . . . . . . . . . . . . . . . . . . .
. . . . . .227
v.
Consequences of Restricted Behavioral Response
(Mobilization to Stress) . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . 245
vi.
Basic Anatomic-Physiologic Considerations
Of Systems Simulation . . . . . . . . . . . . . . . . .
.. . . . . . . . . . . . . . . .247
References 252
Part I. ACCUMULATED STRESS
Section A. Autonomic Stress: Introduction and Definitions
The term “stress,” despite its universal appearance in the nomenclature of biology and medicine, has been and is used without precise or even consistent definition. This unusual state of affairs must be due to a need in these sciences to describe significant groups of phenomena which simply are not covered adequately by other generic terms or concepts. In Mason’s (1976) words: “The controversy over the definition of the term ‘stress’ does not bear upon the validity of the underlying scientific observations or concepts.”
One of the areas where stress has
variously been considered is in its relation to disease. The accumulation of “stresses and strains”
has in many instances been indicated as a contributory or even primary
factor. Diseases such as hypertension,
ulcers, asthma, heart conditions, and even various neoplastic growths and
certain types of diabetes are widely recognized as having “constitutional” and
“emotional” stress components. More and
more, these factors have been acknowledged by members of the m medical
profession and sciences. Yet there have
really been few, if any systematic means to separate and study these stress
factors and their cumulative effects.
The use of the concept of
homeostasis in the analysis of stress can be useful in eliminating some of the
vagueness from the term, and in suggesting a working definition. The mobilizing energy in the anticipation of
extreme muscular exertion needed for the “life or death” struggle in these
emergency situations. It is of no use
for the animal to maintain an internal consistency if it is eaten in the
process. On the other hand, survival in
the face of emergency, if the organism is unable to return to the previous
non-emergency equilibrium, diminishes the capacity for internal regulation.
The basic idea to be built upon here
is that activation of emergency response and the functions of efficient
cellular activity are often, if not basically, incompatible. Further, they are timed and balanced
dynamically to the service of organismic survival, the acute adaptive
response of Cannon’s sympathetico-adrenomendullary system having temporarily
a higher priority than the ongoing activities of cellular homeostasis.
In studying factors controlling the
adrenal medulla, Cannon and his students found that the control of this gland
was carried out by the Autonomic Nervous System (ANS). It was also being discovered that regulation
of such automatic control functions as blood pressure, temperature,
ventilation, osmolarity, and energy balance were also in the province of the
ANS. Since the activity of the adrenal
medulla is regulated by the sympathetic branch of the ANS, this meant that the
same division of the ANS participated both in an array of minute, continuous
internal adjustments as well as in preparing the organism for flight and fight
reactions. Only during these extreme
conditions, he reasoned, did the sympathetic division “takeover” and
temporarily suppress the normal delicate regulation of the internal milieu and
restitution of cellular function (which he felt was served by finely graded
reciprocal shifts between autonomic states, i.e., both sympathetic and
parasympathetic).
The “shades of grey” wherein the
organism may not be able either to fully “mobilize” towards meeting
emergency conditions or to make completely the transition back from emergency
to “normal” situations with their much smaller and more precise requirements,
were not derived in Cannon’s era. It is
the classification and understanding of these phenomena that is a primary
concern of this paper.
A major shift in perspective comes
about when one considers that patterns of autonomic function are plastic and
therefore subject to modification by experience. And, while Cannon’s discoveries apply to
animals in the wild, they are almost certainly not sufficient in ;understanding “modern civilized man” or even animals in
a laboratory environment. As Smelik
(1972) aptly puts it:
“It could happen some twenty years ago, that animals
were transferred to the experimental room to undergo a stressful procedure, and
that the experimenters were not aware of the fact that the simple opening of
the cage and handling had already activated the adrenal system. It appeared that actually not only the
harmful stimulus or the life-endangering situation elicits the adaptive reflex,
but the anticipation of danger already triggers off the alarm reaction.”
This
anticipation, which occurs both in humans and in wild animals, is of obvious
natural survival value. In humans and
laboratory animals, however, the usual mobilization which follows in the
wild is suppressed or absent. Only the perception of this reaction, which,
in humans, is probably fear, is present as an acute
state. Chronic anxiety[1]
can have profound autonomic and hormonal influences—a fact fully agreed upon by
most clinicians and researchers in the field of psychosomatic medicine.
Thus
it will be of great importance in the study of the various stress syndromes, to
understand the potential mechanisms for “accumulation of stress,” i.e., for the
transition from an acute response towards a chronic limitation in overall
organismic function and efficiency.
Yet
understanding of the role of stress as an underlying factor in the process of
health and disease has been in such complete disarray that it has recently
prompted a re-examination of this crucial arena. The initial volumes of the newly formed Journal
of Human Stress (Vol. 1; Nos. 1, 3, 4) contain a discourse between two
of the most prominent figures in stress research today: Hans Selye and James Mason.
Two
basic issues dealt with in this debate are the generality vs. specificity of
stress, and whether the concept is more properly tied to stimulus or to
internal response dimensions. Selye
defines stress wholly in terms of a specific stereotyped response
(pituitary-adrenocortical) which is evoked, generally, by all noxious stimulus
agents. Mason sees this same response,
however, as but one of several endocrine reactions to what he considers a
relatively specific group of stimuli—those which have “psychological
components.”[2]
It
will be worthwhile to outline, in single steps.
The meaning and scope of “stress” and its relation to health
and disease, as it will be used in this dissertation. The facet of stress to be dealt with here is
its effect on the autonomic nervous system (Autonomic Stress)[3],
the mechanisms by which it accumulates and its relation to an organism’s
potential or reserve capacity to meet further stress; as well as the eventual
pathological breakdown and manifestation of the various symptoms of the
so-called “stress diseases” as this capacity becomes sufficiently
diminished. This is not to imply an
absolute threshold relationship between the accumulation of AS, the eventual
breakdown in disease and the manifestation of symptom pathologies. It does imply the existence of lawful
processes in the transition between health and disease, which can be understood
with a degree of quantitative rigor.
The
next step is to formulate and define the phenomenological and neurological
mechanisms by which AS accumulates over time, and how that leads progressively
to limitation in an organism’s capacity to respond appropriately to further
stress (dis-ease) and then finally to the appearance of the “stress disease.”[4]
Stress
is defined as a process whereby a stimulus elicits activation of the
autonomic nervous system (ANS) to such a degree that return to the homeostatic
balance can be interfered with.[5]
Stress
is then further defined in terms of a dichotomy which divides it into two
forms: resolved and unresolved or
accumulated. In a particular situation
it is both the nature of the stressful stimulus and the present “capacity”
of the organism to “respond” to this stress. This will determine whether the situation is
resolved or whether it becomes “internalized” within the organism as a
decreased capacity to resolve future stress.
It
is proposed, in other words, that stress be defined in terms of a pattern of
autonomic reactions which are not necessarily reversed. When initial conditions are re-established,
the stress is said to be resolved. On
the other hand, when the autonomic stress response is evoked but doe not return
to its initial state, it is defined as accumulated, and consequently, the
autonomic response characteristic to subsequent arousal is fundamentally
altered.[6]
In
addition to the involvement of the ANS in the reaction the figure illustrates
that not only somato-visceral behavior but endocrine responses participate as
well. The two-way arrows allow for more
generality. In addition, the two way
flow a, a1 illustrates that the state of the ANS, as well as the
nature and magnitude of the stimulus influence one another.
The
cycle by which stress is resolved is then defined as follows:
Curve
a represents a totally unresolved stress residual, while b and c
are partially accumulated.
In summary, then, for a stress to be
resolved the shift of autonomic activity evoked by the stimulus must be
restored to the pre-stimulus value. If
the level does not return to that baseline, the stress reaction is said to be
unresolved and a residual stress accumulates, modifying the baseline of
autonomic activity.
The mechanisms by which stress is accumulated
are central to the development of this paper, and are intimately related to the
fact that autonomic activity is expressed, at the effector level, by the
interplay of two component branches, the sympathetic and parasympathetic
division.[7]
For reasons which will become clear
as the theme of this dissertation is further developed, the autonomic stress
response is divided into two primary components, charge and discharge,
as shown below:
While stress has been defined in
terms of autonomic activities, care should be taken not to think of the
autonomic system as a functionally distinct efferent channel isolated from the
central nervous or peripheral somatic systems.[8]
As early as 1925 Hess distinguished
between ‘ergotropic’ (E) and ‘trophotropic’ (T) reactions. The former consisted of sympathetic
discharges which were always combined with heightened activity of the somatic
muscular system and cortical arousal, while the latter involved parasympathetic
discharges and inhibition of somatic and central functions. Indeed, the major function of the autonomic
charging was, as Cannon first realized, a preparation and mobilization towards
flight or fight. This depended upon the
capacity for intense and highly organized motor behavior. As this behavioral response was terminated, a
return to the pre-stress autonomic baseline would serve again the ongoing
homeostasis. Thus the complete cycle by
which stress activation is regulated can be diagrammed as follows:
magnitude of the discharge is considered to be determined primarily by the
intensity, rate and duration of the charge phase, the mobilization being more
of a catalyst than entering into the dynamics of the discharge phase directly.[9]
Cannon’s emergency reaction can be
restated, then, in terms of the cycle, as a three phase response involving: (1)
autonomic (sympathetico-adrenal) activation; (2) motor response (mobilization)
and (3) return to pre-activation levels.
Normally (in the wild) these three
phases would occur sequentially, each one leading to the next: the activation evoked by threatening stimuli
is supported by and organized into appropriate motor response. This is followed by the phase of discharge
into neutral equilibrium again.[10]
It is only in this context of the
organismic adaptive response that autonomic activation “makes sense”; the
various components are appropriately phased so as to reinforce an integrated
response and to insure a return to the pre-stress level of ongoing cellular
maintenance. Thus the development in the
course of evolution of highly specialized mechanisms to respond to and organize
for extreme emergency, with their obvious survival value, would have required
parallel machinery to have evolved insuring that these responses acted only
during the time when threat was actually present.
The normal mechanisms by which this
balance is established appear to be basically similar to those discovered by
Sherrington in his pioneer work (1906) for spinal reflexes. He found that changes in the state of excitation
are followed by compensatory phenomena.
This inhibition of a reflex by its antagonist subsequently enhances the
contraction of agonist. In Sherrington’s
words, the “inhibition is followed by rebound to super activity.” Similar phenomena also occur at higher levels
of the CNS, particularly in the hypothalamus, and involve both the ergotropic
(E) and trophotropic (T) systems.[11] These effects, studied by Gellhorn and
associates (1943, 1958, 1959a,b) can be summarized
briefly as follows (Gellhorn, 1969):
1. Excitation of the ergotropic
system: Brief supra-threshold
stimulation of the ergotropic division of the hypothalamus which increases
blood pressure and heart rate during stimulation is followed by a sudden
decrease in blood pressure and heart rate immediately after stimulation. This trophotropic rebound is directly related
to the intensity of the preceding sympathetic excitation regardless of whether
increasing degrees of excitation had been produced by changes in voltage, frequency,
duration of stimulation or similar factors.
2. Excitation of the
trophotropic system: Stimulation of the
intralaminar thalamic nuclei with currents at a low frequency (3 to 5/sec)
which produces recruitment (waxing and waning of potentials in thalamus,
caudate nucleus and cortex) is followed after stimulation by a typical arousal
pattern in the cortex consisting of potentials of low amplitude and high
frequency.
It may therefore be said the ergotropic patterns
elicited by diencephalic stimuli are followed on cessation of stimulation by trophotropic
patters and vice versa. These rebound
phenomena tend to maintain ergotropic-trophotropic balance.
Thus the processes of charge and discharge can be viewed in terms of the hypothalamic response to excitation: the process of charge being the build-up of central sympathetic activity and its shift to the parasympathetic:
It is the normal reciprocal relation
of sympathetic and parasympathetic, the, as we see in the above figure, which
insures homeostatic return to baseline autonomic activity.
On the other hand, Gellhorn (1937,
1968a) has found numerous cases where the above homeostatic processes are
effective only to a limited degree, and a “tuning” of either branch, at
the expense of the other, becomes evident.
For example, recall, if the hypothalamus is stimulated at one ergotropic
(sympathetic_ site, with a brief suprathreshold stimulus, a characteristic rise
in blood pressure (BP) and heart rate (HR) will result. This is followed by a trophotropic
(parasympathetic) rebound (decrease of BP and HR). If at another ergotropic site in the
posterior hypothalamus a more prolonged, near threshold stimulus is
applied, little or not ergotropic discharge occurs. When, however, the two stimuli are combined
so that the brief suprathreshold stimulus is applied in the middle of the
prolonged subthreshold one, the normal supra-stimulus does not produce a
trophotropic rebound. The minimal
subthreshold ergotropic excitation counteracts the trophotropic discharge which
followed the suprathreshold one when it was applied alone. (The explanation might be that normal
suprathreshold stimulation of the ergotropic system inhibits the trophotropic
and then trophotropic rebound is a release from inhibition. But if it is not inhibited enough in the
first place no release excitation occurs.)
Further, in those instances where the trophotropic rebound does not
occur, ergotropic “afterdischarges” do.
That is, instead of the ergotropic stimulation being followed by a
trophotropic rebound it is followed by its own reactivation.
These observations suggest that ergotropic
afterdischarges produced by various combinations of increasing frequency,
intensity, or duration of hypothalamic stimulation might also counteract the
homeostatically acting rebound phenomena.
To test this hypothesis Gellhorn
applied hypothalamic stimulation with increasing duration. Two phases were observed: if the ergotropic stimulation is terminated
in from two to eight seconds, the trophotropic rebound is
increased along with the magnitude of the preceding ergotropic excitation; but
with stimulation periods of ten to fifteen seconds (or more) the
trophotropic rebound is progressively reduced while the ergotropic afterdischarge
increases (Gellhorn, 1959). These
two responses, along with their respective charge/discharge curves (C-D) are
compared in the following figure:
We see above, then, the possibility of a physiological mechanism accounting for the accumulation of stress in a failure of the charge-discharge mechanism to complete a full cycle.
This imbalancing effect can readily
become progressive due to a phenomenon Gellhorn calls “tuning”: “In a state of sympathetic tuning, the
reactivity of the sympathetic division of the hypothalamus is enhanced and that
of the parasympathetic division is lessened.
Similarly, in a state of parasympathetic tuning, the parasympathetic
responsiveness of the hypothalamus is augmented, whereas its sympathetic
reactivity is lessened.” (Gellhorn, 1967a). Simply, if one branch of the ANS, for
whatever reason, becomes dominant,[12]
then the responsiveness of the other becomes diminished over a period of time;
which is to say that the tuning has become enhanced (and will lead to further
tuning of that branch). In this way the
restorative homeostatic potential is diminished. In addition, Gellhorn notes phenomena whereby
one branch becomes tuned to such a degree that “reversal” occurs: Stimuli which normally evoke an ergotropic response
will, in a trophotropically tuned situation, elicit instead a trophotropic
response.
Obviously, understanding the
dynamics of these processes and the “real life situations” which initiate (and
which block) them will be important to the understanding, prevention, and
treatment of clinical conditions deriving from this loss of reciprocal
capacity.
It will be argued in subsequent
chapters that situations which militate against the resolution of stress (and
for its accumulation) can be grouped into three basic types, which are not
meant to be absolute but broad and partially independent classes:
1) Those in which the level of
activation has become so intense that the organism’s central processing
machinery is unable to integrate the stress into an appropriate mode of
discharge.
2) Those in which the buildup
of charge is so slow (i.e., as in chronic low grade “environmental” or “social”
stress) that the mechanisms of rebound are not activated and in which a more
acute (though by itself moderate and resolvable) stress response is evoked on
that background and becomes accumulated.
3) This in which the somatic
(motoric) component of the discharge has been blocked from full or appropriate
expression.
Section
B. Catastrophe Theory
The question is how can the existence of accumulated stress be “proven,” as well as its level measured in humans? To do this requires that the various parameters of stress by first defined in a mathematical form so that specific quantitative as well as qualitative predictions can be formulated and specifically tested. The strategy taken in the subsequent sections will be to look at mathematical-topological properties which can be expected directly from the most basic (and minimal number of) well known physiologic properties of the ANS.
To begin, one of the most
fundamental properties of the ANS (and of the nervous system in general) is the
phenomenon, demonstrated by Gellhorn, that stimulation of either branch with
brief electrical or natural stimuli evokes compensatory rebound of the opposite
one; i.e., sympathetic stimulation evokes a secondary parasympathetic response
and vice-versa.
Reciprocal activation of sympathetic
(S) and parasympathetic (PS) autonomic components does not take place
instantaneously but with a significant measurable delay and with only a small
degree of overshoot. It exhibits, as an
energy system, then, properties characteristic of a high degree of frictional
damping.
These basic properties, i.e.,
reciprocity, friction and delay, can be represented by the following simple
arrangement of an inertial disc, pivoted at its center “O”, and with two
elastic bands fastened to a point on its circumference, the free ends of which
are held at points along a straight line through O (figure I).
The sympathetic (S) and
Parasympathetic (PS) activity are represented respectively by the two
bands. If the disc is twisted in a
clockwise direction, then the band representing parasympathetic activity is
stretched or “charged,”
While
a counterclockwise turn activates the one labeled sympathetic. Stretching (“charging”) one (by rotating the
disc) diminishes the other’s charge or tension in a manner described by
Gellhorn, and when either a clockwise (PS) or a counterclockwise (S) turn is
released, it will—depending on friction and the disc’s inertia- -return past
the neutral position, discharge into the opposing branch and then tend towards
the neutral position, illustrating also the phenomenon of rebound.[13]
Description of this “machine” so far
gives no added information on the relations of the autonomic components (S and
PS) in the accumulation of autonomic stress (AS). It is, though, with some malice to
forethought, that this ridiculously simple machine, similar to one invented by
Zeeman, exhibits certain essential behaviors described by “Catastrophe Theory,”
a branch of mathematics theory new to this decade. This theory, in conjunction with control
systems analysis, will set the foundation for a model of the ongoing process of
health. Health is defined in terms of
full autonomic range; dis-ease as a lessening in this capacity; and disease as
the abrupt discontinuous changes in behavior and energy metabolism which
characterize pathologic stress diseases.
Towards these ends, basic ideas from Catastrophe theory will be
explored, and some less than obvious, unexpected mechanisms for the
accumulation of stress developed.
Rene Thom, in what has been termed
“an intellectual revolution,” (Stewart, 1975), has developed a theory which
comes to the conclusion that almost all systems, which—in a mechanical
analogy—have a high degree of friction, fall into only seven types. Thom calls them catastrophes to accent the
quality of sudden change. The theory itself is quite elaborate and its proof
rests upon “techniques of great sophistication.” Fortunately, there have been, in spite of its
newness, two very excellent explicatory articles by Stewart (1975) and Zeeman
(1976), which are drawn on in this section.
The behavior of a system, in Thom’s
theory, is governed by an “energy function” E—which is not necessarily the
actual physical energy. If we suppose
that the state of the system can be described by a singe variable x, a graph of
E against x can be plotted. Figure II, 1
is an example. The equilibrium states
correspond to values of E where the graph is horizontal. There are several “stationary values: on this
particular graph: minima at s1, x3, x7, maxima at x2,
x5 and inflection points at x4, x6. The minimum points correspond to stable
equilibria, i.e., to regions to which the system will return after a slight
disturbance, while maxima and points of inflexion correspond to unstable
equilibria.[14]
About the only requirement for the
system is that it tends rapidly to a steady state
equilibrium. Thus, frictional mechanical
systems (force is proportional to velocity rather than acceleration) provide a
good specific example with which to illustrate the basic tenets of the theory.
A simple physical model which
displays all the relevant phenomena is Zeeman’s “Catastrophe Machine,”
illustrated in figure II, 2: the device,
made up of a circular disc, pivoted at the center and free to rotate, with two
elastic bands attached to its edge, has already been described. In the formal machine, however, the remaining
end of one piece is fixed at point Q, while the other end P is free to move in
the plane of the machine.
Experimentally, it can be shown that
the diamond shaped area ABCD has the following property: if the free point P is outside ABCD, the
rotation angle of the disc (8) has only one stable equilibrium position; but if
P is inside this region there are two stable equilibria. Thus, if the disc is twisted or rotated by an
external force, it will return to the same position (if P is outside that
region), just as a ball rolling down the side of a closed trough will settle to
the bottom. If, however, P is within ABCD,
the disc will fall into one of two positions.
In general, if P is moved smoothly, the equilibrium position of the disc
will (in the absence of any additional forces) also change smoothly. If, however, P moves across the edge of the
region enclosed by ABCD, the disc may make a sudden jump from one equilibrium
position to a completely different one.
In figure II, 3, as P moves along the path UVWXYZ, a jump occurs in the
position of the disc as it passes out of the diamond at Y (but not as it enters
at V). Thus, the behavior of the disc
exhibits hysteresis and does not reverse when the path traced by the free end P
is reversed.
This behavior, which seems
mysterious at first, is readily understood if we look at the energy function E
(which, in the case of the Zeeman machine, represents the energy stored in the
elastic bands) for positions of P traversing along the line UVWXYZ (fig. II,
2). When the point is moved outside the
diamond area (e.g., around Y or Z), there is only a single minimum. Inside the area, at W or X, there are two
minima on either side of a central maximum.
At the edges Z and Y, one of the minima has now formed into the maximum,
giving a point of inflection.
Immediately, as P moves outside the diamond, this inflection disappears
completely, so that as P moves from U to Z the disc starts off in the initial
minimum position and, because of the friction, stays at the minimum all the way
across Y. At Y, however, as this minimum
disappears, the disc is “forced” to jump suddenly into the only remaining
minimum, which is some considerable distance away.
This process can be visualized by
drawing a three dimensional graph of these equilibrium positions as a function
of the free end P. A mathematical
analysis of the machine leads to the graph depicted in figure III. The folded surface represents the equilibrium
values for x, and is called the “Behavior Surface.” For any given position of P (a,b) (control points), a vertical line can be drawn, which
cuts the behavior surface at 1, 2, or 3 points.
The lower plane is called the “control surface.” The vertical height of the line corresponds
to the equilibrium value(s) of x. If the
control point P lies outside the shaded region, then only one value of x is
possible. (The shaded region corresponds
to part of the diamond shaped region in figure II, 2.) If, however, the point P lies inside the
shaded region, there are three values that it can take because of the fold in
the surface: one on the upper sheet, one
in the middle, and one on the lower.
Thus, as point P is moved along the path UVWXYZ (i.e., the control
surface), the state of the disc is represented by a point on the behavior
surface vertically above P, and “friction” causes this point to stay on the
same sheet of the surface as long as this is possible. As P moves through V no trouble occurs, but
when P finally moves through Y there is a fold in the upper sheet and the disc
location falls off the edge, onto the lower one, with a sudden jump. It can be shown that all the jumps
which can possibly occur are incorporated into a single simple geometrical
picture like figure III.
In summary, then, the energy which
is minimized in this system is the potential energy stored in the elastic
bands.
The
disc, therefore, rotates until the tension on the two bands is at a
minimum. At that position the machine is
said to be in a stable equilibrium, and unless energy is appropriately added,
the machine must remain at the equilibrium point. The process that keeps it in equilibrium is
called the dynamic, and relates the dependent behavior surface
variable(s) to the independent control surface variables. The dynamic has two functions: First, it holds the behavior point firmly on
the top or bottom sheet of the behavior surface. That is, if the disc is rotated by an
external force and then released (as in the sympathetic-parasympathetic
analogue described in figure I), it is the dynamic which brings it sharply back
to one of the two equilibria. Secondly,
when the behavior point crosses the fold curve, it is the dynamic that causes
the catastrophic jump from one sheet, that is, from one behavior, to
another. So it is the movement of the
control point along the control surface which, through the dynamic of the
system, results in the path taken on the behavior surface.
Thom studied much more general
situations—systems that could be described by a finite set of variables, X, Y,
Z. . . (behavior variables) and controlled by a second
finite set of variables, A, B, C. . . (control
variables) under an energy function E which varied with A, B, C. . . and X, Y,
Z. . . . Thom’s theorem says that with
only a small group of exceptions, it is always possible to effect a smooth
reversible change of coordinates in such a way that in the neighborhood of a
given point the system exhibits on of seven types of behavior.[15] Thus, through catastrophe theory, one can
deduce the shape of the entire surface merely from the fact that the behavior
is bimodal for some control points.
The Zeeman machine and the basic
topology of the Cusp Catastrophe is a very simple system; and the question, of
course, arises as to whether the theory applies realistically too much more
complex systems such as the central nervous system. An energy minimum in a physical system, e.g.,
the Zeeman machine, is a special instance of a concept called an “attractor.” This particular case is an example of the
simplest kind of attractor, the single stable state. It is like a magnet or gravity acting on a
trough well. Everything within its range
of influence is drawn toward it. It is
under the influence of this attractor that the system assumes a state of static
equilibrium.
More generally, the attractor of a
system, in dynamic equilibrium, consists of the entire stable cycle of
states through which the system passes.
The bowed string of a violin, for example, repeats the same cycles of
positions over and over at its particular resonant frequency. This cycle of positions represents an
attracator of the bowed string system (Zeeman).
While attractors can be single
states, they are more likely to be stable cycles of states, i.e.,
“higher dimensional analogs” of stable states.
As various parts of complex systems, such as found in the brain,
influence one another, these attractors would wax and wane with varying degrees
of rapidity, one attractor giving way to another. As this process goes on, the stability of the
system also undergoes alteration, and there is the potential for a catastrophic
jump in state. According Thom’s theory,
though, ALL possible jumps between equilibrium attractors are determined by the
seven catastrophes. This applied
strictly to the subset of point attractors, but nonetheless the elementary
catastrophes can, according to Zeeman, provide meaningful models for behavior
as complex as the brain: “The models are
explicitly and sometimes disarmingly simple, but the powerful mathematical
theory on which they are based implicitly allows for the complexity of the
underlying neural network.”
Zeeman lists five characteristic qualities
common to all cusp catastrophes: 1)
Bimodality of behavior; 2) sudden transitions between states; 3) hysteresis:
the transition between top and bottom sheet behavior does not take place at the
same point; 4) an inaccessible region; and 5) divergence (large differences in
the final state of the system resulting from small perturbations of the initial
state).
Now, independent of the complexity
of a system, according to Zeeman, “if any one characteristic is apparent in
a process, the other four should be looked for, and if more than one is found,
then the process should be considered a candidate for description as a cusp
catastrophe.”
Of these five criteria, the role of the autonomic systems in stress behavior and disease meets at least two of them and possible four. (Using the concept of behavioral motility (bm), the fifth criterion, an inaccessible region, is not measurable.)
The first criterion, bimodality, is the basic behavior of the sympathetic-parasympathetic system. That sudden transitions occur both between and within these systems is demonstrated both by the physiological work of Gellhorn and from the wealth of animal observations by Konrad Lorenz and other ethologists. For example, the dynamics of “decision” whereby an aroused animal either fights or takes flight, i.e., exhibits fear or aggressive behavior, has been demonstrated by Zeeman to conform with the predictions of a simple cusp catastrophe. Also, many stress diseases exhibits discontinuous remission or abrupt changes in symptoms such as extreme excitability and depression, as well as being triggered often by events which appear relatively minor to individuals who are not so predisposed. Even the phenomenon of hysteresis seems to characterize the disease process. In a simulation of anorexia nervosa (see section IV, A, The Stress Diseases), Zeeman clearly demonstrates this property.
In summary, despite the vast
complexity of center integrative processes, and perhaps because of its discrete
division into bipolar output elements, the involvement of the autonomic nervous
system in “stress phenomena” appears well suited to catastrophe theory.
Section
C. The Model
Towards relating autonomic and
accumulated stress behavior, we look at the Catastrophe diagram, shown in
figure IV. The control variables are
sympathetic (S) and parasympathetic (PS activity, while the behavior variable
is labeled behavioral
motility (bm).[16] The neutral point of the system is the
baseline level of autonomic activity (So, PSO), corresponding to a behavioral level bmO(SO, PSo)), which is neither active
nor quiescent, but in “restful alertness” (with the potential for a shift in
either direction).
Let us first examine the behavior of
the system for the case (1) of low to moderate levels of sympathetic
activation. The behavior curve and its
projection onto the control plane (see figures V and VI) follow Gellhorn’s
observations of parasympathetic rebound from sympathetic excitations (solid
line). In figures V and VI, the dotted
curve represents the behavior resulting from a higher level of sympathetic
arousal. The behavioral motility (bm for
both of these curves follows a path which peaks, then subsides smoothly to a
low level of motility (lower than the initial pre-stimulus state) and then
returns to that value, re-establishing homeostasis. This entire process is carried out
continuously and reversibly.
