Progress in Neuro-Psychopharmacology & Biological Psychiatry 29 (2005) 1195 – 1200
www.elsevier.com/locate/pnpbp
Review article
Defining stress as a prelude to mapping its neurocircuitry:
No help from allostasis
Trevor A. Day *
School of Biomedical Sciences, University of Newcastle, and the Hunter Medical Research Institute, Newcastle NSW 2308, Australia
Accepted 26 August 2005
Available online 5 October 2005
Abstract
The way in which researchers conceptualise and thus define stress shapes the way in which they approach the task of mapping the brain’s stress
control pathways. Unfortunately, much of the research currently being done on stress neurocircuitry is occurring within a poorly developed
conceptual framework, a framework that limits the depth of the questions that our studies ask, and even our ability to fully appreciate and make
use of the data that they yield. Consequently, any attempt to improve our conceptual framework merits close attention. In that regard it is notable
that in recent years it has been argued that the concept of homeostasis should be supplemented by the concepts of allostasis (literally Fstability
through change_) and allostatic load (in effect, the cost of allostasis). One of the purported benefits of this change has been that it will clarify the
concept of stress. A close review of the arguments leads us to conclude that the introduction of the concept of allostasis has largely occurred as a
result of misunderstandings and misapprehensions concerning the concept of homeostasis. In terms of understanding how the organism operates, it
is not clear that the concepts of Fallostasis_ or Fallostatic load_ offer us anything that was not already apparent, or at least readily derivable, from an
accurate reading of the original concept of homeostasis. Not surprisingly then, these more recently proposed concepts also offer little help in
clarifying our understanding of stress. Indeed, rather than clarifying the concept of stress, the primary effort appears to be directed at subsuming
the concept of stress within the concept of allostasis, which has the inadvertent effect of collapsing the study of homeostatic responses and stress
responses together. This seems to be out of step with the fact that there is now considerable evidence that the brain does indeed possess certain
pathways that merit the title of Fstress neurocircuitry_. The attempt to subsume the concept of stress within the concept of allostasis is also counterproductive in that it distracts stress researchers from the important task of developing conceptual frameworks that allow us to tackle fundamental
issues such as how the organism differentiates stressful from non-stressful challenges.
D 2005 Elsevier Inc. All rights reserved.
Keywords: Allostasis; Allostatic load; Homeostasis; Stress; Stressor
Contents
1.
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1.1. Sterling and Eyer’s version of the concept of allostasis . .
1.2. McEwen’s version of the concept of allostasis . . . . . .
1.3. Allostatic load . . . . . . . . . . . . . . . . . . . . . . .
1.4. Allostasis and allostatic load: where does stress fit in? . .
1.5. Improving our definition of stress . . . . . . . . . . . . .
1.6. Conclusion: further conceptual issues relevant to mapping
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
. . .
. . .
. . .
. . .
. . .
. . .
stress
. . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
neurocircuitry
. . . . . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
1195
1196
1196
1197
1197
1198
1199
1200
1. Introduction
* Tel.: +61 2 492 15629; fax: +61 2 492 15141.
E-mail address: [email protected].
0278-5846/$ - see front matter D 2005 Elsevier Inc. All rights reserved.
doi:10.1016/j.pnpbp.2005.08.005
Recent years have witnessed considerable progress in our
attempts to characterize the brain pathways that control
1196
T.A. Day / Progress in Neuro-Psychopharmacology & Biological Psychiatry 29 (2005) 1195 – 1200
physiological and behavioural responses to stressful stimuli.
Indeed, the annual rate of publication of relevant research
papers is doubling each decade, while the rate of publication
concerning certain sub-topics, such as the role of the amygdala,
has trebled in the last decade alone. Much of this work has
been underpinned by the development of new techniques for
the structural and functional investigation of the brain.
Consider, for example, our ability to map stress neurocircuitry
at the single cell level by simultaneously combining retrograde
tracing with both immunocytochemical determination of
phenotype and the mapping of activity-related transcription
factor expression; although now routine in many laboratories,
techniques such as these were almost unimaginable just a few
decades ago. Inevitably, other new neurobiological techniques
will soon make our present approaches seem primitive.
However, while technical improvements will always be
welcome, the need to invest time and resources incorporating
them into our repertoire has perhaps had the inadvertent effect
of distracting many of us from attending to the conceptual
foundations of our research. This is a significant concern
because the way in which we conceptualise and thus define
stress strongly affects the design of our research, right down to
the choice of stimuli we use as stressors. Nevertheless, most
reports concerning the mapping of stress neurocircuitry make
no mention of what the authors consider the term Fstress_ to
mean. If a definition is provided, it is often the conventional
statement that Fstress is the response of the body to any actual
or threatened disturbance of homeostasis’, a definition that (as
we shall discuss below), actually undermines the basic
assumption that the brain contains any distinct Fstress
neurocircuitry_ at all!
It will be apparent then that any serious effort to clarify our
thinking with regard to the way in which we define stress
deserves careful consideration. Accordingly, many researchers
who are engaged in mapping stress neurocircuitry will be well
aware that, over the past decade, there have been over a dozen
major reviews written championing the idea that the concept of
homeostasis needs to be supplemented by the concept of
allostasis, a term which literally means Fstability through
change_. A major reason given for taking up this concept is
that it provides a means of achieving a Fmore precise definition
of stress_ (McEwen and Wingfield, 2003a). This claim has
received remarkably little critical comment (but see Dallman,
2003) and is, therefore, the focus of the present paper.
1.1. Sterling and Eyer’s version of the concept of allostasis
At the outset it is important to realise that there has been
more than one version of the concept of allostasis. The original
was put forward in 1988 in a monograph published by Sterling
and Eyer (Sterling and Eyer, 1988). They seized upon the fact
that certain physiological parameters appear to be reset to new
ranges in accord with circumstance, e.g. blood pressure during
waking vs. sleeping. They argued that this demonstrated that
the concept of homeostasis was wrong and must be replaced by
a new concept, i.e. that Fto maintain stability an organism must
vary all the parameters of its internal milieu and match them
appropriately to environmental demands_. Sterling and Eyer
applied the term allostasis to this concept.
In retrospect, it seems apparent that Sterling and Eyer’s
proposal is founded on a basic misunderstanding of the original
concept of homeostasis. In their monograph they defined
homeostasis as requiring that Fto maintain stability an organism
must hold all the parameters of its internal milieu constant_
(my italics). However, this differs with what Walter Cannon
conceived of when he coined the term homeostasis in the early
20th century. When explaining the concept of homeostasis,
Cannon (1929) made it clear that the primary focus of attention
was the relative stability, despite environmental fluctuations, of
those tissue parameters that were critical to cell survival, e.g.
nutrient availability, oxygen availability, temperature, pH, and
ion concentrations. Cannon explicitly discussed the fact that
other parameters that were relevant to the maintenance of the
peri-cellular environment, e.g. blood pressure, heart rate and
blood sugar levels, although usually staying within certain
Fnormal ranges_ when the body is at rest, would need to be
actively changed at other times to meet tissue needs. Examples
that Cannon used included the way in which Fvigorous
muscular effort_ had to be supported by appropriate cardiovascular changes (Cannon, 1929), and the way in which
Femotional excitement_ led to Fanticipatory_ increases in blood
sugar levels that Fput the organism in readiness for meeting the
demands which will be made upon it_ (Cannon, 1939). In short,
Cannon never suggested that the concept of homeostasis
required that all physiological parameters must be held
constant at all times — just the core tissue parameters critical
to cell survival. Indeed, it seems safe to assume that Cannon
would have regarded the notion that the body operates on the
basis of achieving Fstability through change_ as being integral
to the concept of homeostasis.
It is noteworthy that Sterling and Eyer’s (1988) misunderstanding of Cannon’s concept of homeostasis extended to the
mechanisms by which it might be achieved: they claimed that
homeostasis was supposed to depend entirely on local negative
feedback mechanisms. However, Cannon (1939) explicitly
stated that homeostasis would be achieved by means of Fthe
brain and nerves, the heart, lungs, kidneys and spleen, all
working cooperatively_, i.e. certainly not just local negative
feedback mechanisms.
In summary then, the ultimate problem with Sterling and
Eyer’s concept of allostasis is not that their concept misrepresents the way in which the body operates, but rather that it
is simply an unnecessary re-statement of the concept of
homeostasis.
1.2. McEwen’s version of the concept of allostasis
Sterling and Eyer’s version of the concept of allostasis
received no attention in the literature until 1993, when it was
picked up in an influential review written by two prominent
neuroscientists, Bruce McEwen and Eliot Stellar (McEwen and
Stellar, 1993). Stellar passed away shortly after their paper
appeared, but McEwen has continued to champion the concept
in a dozen or so reviews published over the last decade. Some
T.A. Day / Progress in Neuro-Psychopharmacology & Biological Psychiatry 29 (2005) 1195 – 1200
of these reviews have been written in collaboration with
scientists from adjunct fields, thereby increasing exposure of
his ideas across biomedical sub-disciplines.
Not unreasonably, McEwen’s version of the concept of
allostasis has evolved over the past decade. Thus in the first
review allostasis was defined as Fthe ability of the body to
increase or decrease vital function to a new steady state on
challenge_ (McEwen and Stellar, 1993). In the second review,
co-authored with Schulkin and Gold (Schulkin et al., 1994),
allostasis was defined as Fthe regulation of many variables over
time in maintaining stability to meet changing circumstance_.
Notably, in the latter review it was also stated that allostatic
mechanisms differed from homeostatic mechanisms in that
they were tied to anticipation of needs rather than being
triggered by a detected deviation from a setpoint. This claim is
rather ironic as, in his writings, Cannon had explicitly raised
the issue of anticipation in relation to the mechanisms
responsible for maintenance of homeostasis but in fact never
discussed setpoints! The notion that the central nervous system
might be able to encode setpoints was developed by researchers
who came after Cannon; moreover, the centrality of setpoints to
the maintenance of homeostasis is now under question
(Berridge, 2004).
In more recent reviews the definition of allostasis favored by
McEwen has shifted again, now being described as Fthe process
for actively maintaining homeostasis_ (e.g. McEwen, 2000a,b;
McEwen and Wingfield, 2003a). Because this definition has
remained stable over several years, it is this one that we shall
now focus on.
It is critical to appreciate that McEwen’s version of the
concept of allostasis differs significantly from Sterling and
Eyer’s in that he does not reject the concept of homeostasis.
Thus, while Sterling and Eyer argued that the concept of
allostasis should entirely supersede the concept of homeostasis,
McEwen describes them as being complementary (e.g.
McEwen, 2000b). Indeed, because McEwen’s usage of the
term allostasis is now as a label for the processes that achieve
homeostasis, what McEwen would refer to as Fallostatic
mechanisms_ are actually identical to Cannon’s Fhomeostatic
mechanisms_ (Cannon, 1939).
A critical question then is whether having a specific name
(allostasis) for the processes that achieve homeostasis is useful.
McEwen has argued that it is, because it Fclarifies an inherent
ambiguity in the term_ (McEwen, 2000b). The ambiguity
referred to arises from the fact that in some cases the term
homeostasis has been used to denote the process of achieving
relative stability – which is wrong – as well as the actual state of
relative stability of core tissue parameters –which is correct.
McEwen of course is right to point out that this error can be
seen in the literature, but does this demonstrate Finherent
ambiguity_? Cannon at least was perfectly clear about the
difference between the endpoint (homeostasis) and the
mechanisms required to achieve it, rather whimsically speaking
of the Fnice devices_ used by the body to achieve homeostasis,
devices that he of course referred to as Fhomeostatic mechanisms_ (Cannon, 1939). So, while McEwen’s version of the
concept of allostasis is not founded on the misunderstandings
1197
concerning homeostasis that featured in Sterling and Eyer’s
(1988) monograph, it nevertheless reflects a failure to
distinguish between a misrepresentation of the concept of
homeostasis (by later workers) and any inherent problem with
Cannon’s original concept. In summary, it is my view that
allostasis, as championed by McEwen, does not represent a
fresh conceptual insight into how the organism operates, but
just a new label for processes already described, i.e. the
operation of homeostatic mechanisms. However, before reaching any conclusion as to the impact of the concept of allostasis
on our understanding of stress, it is appropriate to first consider
the related concept of allostatic load.
1.3. Allostatic load
At the same time as adopting and modifying Sterling and
Eyer’s concept of allostasis, McEwen and Stellar introduced
the concept of allostatic load (McEwen and Stellar, 1993).
They defined allostatic load, in effect, as the strain placed on
the organism through its attempts to maintain homeostasis
against all challenges, whether daily events or more enduring
life-style and life-cycle related factors, such as obesity, drugtaking, or reproduction. All of these things would contribute to
allostatic load and thus occasion wear-and-tear by requiring the
body to work harder to maintain the stability of core tissue
parameters critical to tissue and thus organism survival. They
pointed out that excessive allostatic load could be expected to
produce illness and/or accelerate ageing, something of obvious
clinical significance. McEwen and colleagues subsequently
Foperationalised_ the concept of allostatic load by selecting
certain physiochemical variables for measurement (e.g. cortisol
and adrenal catecholamine levels) and testing whether,
collectively, they predicted subsequent health (McEwen and
Seeman, 1999; Seeman et al., 1997). Most will agree that,
while there may be methodological issues that need to be
addressed with regard to the Foperationalisation_ of allostatic
load (i.e. the choice of variables to be measured; see
Schnorpfeil et al., 2003), this attempt to map a relationship
between Fload_ and health has considerable merit. But does this
require the concepts of allostasis and allostatic load? I believe
not. If allostasis is defined simply as the process by which
homeostasis is maintained, then it would seem that this
interesting endeavor could have been just as easily launched
on the basis of the concepts of homeostasis and homeostatic
load.
1.4. Allostasis and allostatic load: where does stress fit in?
McEwen and his co-authors have stated that their interest in
the concept of allostasis has been at least partly motivated by a
dissatisfaction with the fact that Fthe concepts of stress and
homeostasis have been used in ambiguous ways_ (McEwen and
Wingfield, 2003a) and a hope that Fthe allostasis terminology
helps (us) to get inside (the terms) stress and homeostasis and
(thus) understand better what they mean_ (McEwen and
Wingfield, 2003b). In the preceding sections, however, it has
been argued that, rather than improving our understanding of
1198
T.A. Day / Progress in Neuro-Psychopharmacology & Biological Psychiatry 29 (2005) 1195 – 1200
homeostasis, both versions of the concept of allostasis –
Sterling and Eyer’s, and McEwen’s – have stemmed from
misunderstandings or at least misapprehensions concerning
the concept of homeostasis. Nevertheless, it is still interesting
to consider what McEwen and colleagues have to say about the
relationship between allostasis, allostatic load, and stress.
It is clear from his reviews that McEwen regards stress as a
term that is overused and ambiguous. A particular concern is
that while the terms stress response and stress mediators are
generally taken to have negative connotations, these mechanisms actually evolved because, in the short term, they provide
an adaptive advantage; their damaging effects only arise when
stress is excessive, or when stress-inactivation systems malfunction (McEwen, 1998). McEwen also points out that there
can be a significant disconnection between an individual’s
apparent experience of stress and the circulating levels of the
signalling agents that have come to be known as stress
mediators. For example, despite always experiencing a prewaking morning rise in glucocorticoid levels, few individuals
regard themselves as being stressed while still asleep, nightmares aside (McEwen, 2000a). McEwen proposes that it is
possible to eliminate these inherent contradictions by subsuming Fstress_ into the broader construct of allostasis. Thus, in a
recent review, McEwen and Wingfield (McEwen and Wingfield, 2003a) wrote that F(the term) stress will be used to
describe events that are threatening to the individual and which
elicit physiological and behavioural responses as part of
allostasis in addition to that imposed by the normal life cycle_
(my italics). They propose, in effect, that stress is just one type
of challenge that can activate of allostatic (or, as I prefer,
homeostatic) responses. Accordingly, we can summarise their
position as follows: life is a series of challenges; some are part
of the normal life cycle; some can be described as stressors; all
of these challenges must be met, i.e. homeostasis must be
maintained; the process of maintaining homeostasis (a process
they would refer to as allostasis) involves wear and tear (which
they refer to as allostatic load) which can impact adversely on
health. This re-statement of McEwen and Wingfield’s thesis
may seem banal but reading it with the bracketed words
eliminated will demonstrate that understanding their thesis does
not require the adoption of allostasis terminology. The critical
question that remains then is this: does the concept of allostasis
help us to better define stress? I suggest that the answer is Fno_.
Indeed, in practice, the adoption of allostasis terminology seems
to constitute an attempt to avoid the term Fstress_ altogether;
thus allostasis is substituted for stress response, allostatic
mediators for stress mediators, and allostatic load for stressrelated damage. This has the inadvertent effect of collapsing the
study of homeostatic responses and stress responses together.
McEwen and Wingfield’s view, noted above, that stress is
just one type of challenge able to trigger homeostatic
mechanisms, highlights a broader problem inherent in the
definition of stress that is most commonly encountered in
biomedical research papers, i.e. that Fstress is the body’s
response to any actual or threatened disturbance of homeostasis_ (Johnson et al., 1992). Brief reflection reveals that this
definition carries the implication that all threats to homeostasis,
no matter how minor and transient, must be viewed as stressors
— even eating lunch (which will trigger a rise in blood glucose
level and thus trigger a homeostatic response). Given that
McEwen sees stress as just one type of challenge to
homeostasis (the others being life-style and life-cycle related),
it would seem inconsistent for him to support this common
definition of stress. However, it is not clear that he recognises
this contradiction as, in one of his most recent reviews, he
repeats a form of this common definition without demurring
(McEwen and Wingfield, 2003a).
In summary, the foregoing analysis leads to the conclusion
that a clearer definition of stress is not inherent in the concepts
of allostasis and allostatic load per se. However, the position
outlined by McEwen and Wingfield (2003a) concerning the
(contributory) role of stressful events in triggering homeostatic
mechanisms highlights an inherent problem in the all-toocommon definition of stress as Fthe body’s response to any
actual or threatened disturbance of homeostasis_. Accordingly,
we shall now take the opportunity to explore this problem
further, in the hope that it provides some insights into ways in
which we might improve our ability to conceptualise and
define stress.
1.5. Improving our definition of stress
From the perspective of those interested in mapping Fstress
neurocircuitry_ it would seem self-evident that they would
expect any definition of stress to be such as to allow for the
possibility that the brain does indeed contain a sub-set of neural
circuits that merit the title Fstress neurocircuitry_. It is ironic
then that so many papers concerning the mapping of stress
neurocircuitry explicitly or implicitly take their definition of
stress to be: Fthe response of the body to any actual or
threatened disturbance of homeostasis_. This definition carries
the implication that all threats to homeostasis must be viewed
as stressors. Given that almost every part of the brain is
involved directly or indirectly in the survival of the organism
and thus the defence of homeostasis, this definition virtually
dictates that almost all brain circuitry should be deemed Fstress
neurocircuitry_ — not a very useful position at all. In
retrospect, the critical point is that this common definition of
stress inadvertently conflates two fundamentally different types
of response that can be used by the body to meet a challenge:
(i) A selective response that involves the targeting of a subset
of cells or tissues so as to achieve a circumscribed
adjustment appropriate for handling a specific minor
challenge. Examples might include the release of insulin
in response to a glucose load, a transient increase in
alertness when an unanticipated sound is heard, or
cutaneous vasoconstriction in response to a small fall in
air temperature. In each case, this response would likely
suffice to avoid or rectify any disturbance of homeostasis.
(ii) A non-selective response that involves simultaneous
targeting of multiple tissue types and organ systems so
as to achieve a major redistribution of the body’s
resources in response to, or anticipation of, a challenge
T.A. Day / Progress in Neuro-Psychopharmacology & Biological Psychiatry 29 (2005) 1195 – 1200
1199
unlikely to be met by a selective response. Examples
would include adrenocortical, adrenomedullary and
cognitive activation either when an unanticipated sound
was recognised as the first sign of an attack, or when a
fall in air temperature proved to be so severe that
selective physiological responses (e.g. cutaneous vasoconstriction) or behavioural responses (e.g. putting on
more clothing) would not suffice to protect body
temperature.
neurocircuitry activated during a non-hypotensive haemorrhage
is not Fstress neurocircuitry_. In summary then, researchers who
wish to map stress neurocircuitry need to differentiate in their
studies between challenges that elicit first tier (selective) versus
second tier (non-selective, stress) responses. However, what we
currently see in the literature are many studies where, simply
because an applied stimulus has elicited a response (indeed,
any response), it is assumed that Fstress neurocircuitry_ is
involved in the generation of that response.
Most readers will recognise this idea – that the body can
mount qualitatively different responses according to the severity
of the challenge – as echoing Selye’s definition of stress as Fthe
non-specific response of the body to any demand_ (Selye, 1936,
1976, 1980). But while many biomedical researchers have
hailed Selye as Fthe father of stress research_ and have also
followed his lead in using the term stress to refer to the response
to a challenge and stressor to refer to the challenge, the nuances
of his definition of stress have often gone unappreciated. This is
commonly due to uncertainty as to what Selye actually meant
by Fnon-specific response_ (Selye, 1976, 1980). I suggest that
Fnon-specific_ can be read in two ways: firstly, as denoting
outcomes that are common to all stressors, e.g. adrenocortical
activation, and secondly as denoting non-selective or global
changes that re-orientate almost all of the organism’s cognitive
and physiological systems towards meeting the challenge at
hand. Thus, Selye’s definition can be taken as differentiating
between selective (and normally adequate) responses to minor
challenges, and non-selective (emergency) responses to major
challenges which can be termed stressors. Consequently, an
appropriate re-wording of Selye’s definition of stress might be
that: stress is the body’s multi-system response to any challenge
that overwhelms, or is judged likely to overwhelm, selective
homeostatic response mechanisms.
In the Introduction to this paper it was stated that the way in
which we conceptualise and thus define stress strongly affects
the design of our research, right down to the choice of stimuli
we use as stressors. To now illustrate this point in relation to
the above definition of stress, let us consider haemorrhage.
Because it undoubtedly constitutes a homeostatic challenge,
many would consider any haemorrhage to be a stressor and any
neuronal populations activated by it to be part of the brain’s
stress neurocircuitry. However, the response to haemorrhage
differs, qualitatively as well as quantitatively, according to the
severity of the haemorrhage. Non-hypotensive haemorrhage
triggers the release of vasopressin into the circulation, which
constitutes a selective and highly appropriate first tier response
to this challenge because vasopressin, as well as suppressing
diuresis, acts as a potent vasoconstrictor of mesenteric vessels
and thus shunts bloods from a major but non-essential vascular
bed to the heart and brain. In contrast, hypotensive haemorrhage can trigger not only vasopressin but also glucocorticoid
and adrenal catecholamine secretion, clearly a non-selective
response. From the preceding discussion it will be apparent that
adoption of Selye’s re-worded definition of stress requires us to
accept that a non-hypotensive haemorrhage should not be
considered to be a stressor and that, consequently, the
1.6. Conclusion: further conceptual issues relevant to mapping
stress neurocircuitry
Students of Western philosophy sometimes adopt the
position that nothing new has been said since Socrates, Plato
and Aristotle, and that there is in fact no more to be said. Some
readers may feel that such a claim is being made in this paper,
but for Cannon and Selye in relation to the concepts of
homeostasis and stress. Certainly I have argued that the
introduction of the concepts of allostasis and allostatic load
constitutes little more than a re-labelling exercise. However, far
from believing that the last word has been said, it is clear that
considerable scope for fresh thinking remains in relation to the
topic of stress. For example, the proposed adoption of Selye’s
(re-worded) definition of stress as Fthe body’s multi-system
response to any challenge that overwhelms, or is judged likely
to overwhelm, selective homeostatic response mechanisms_
still leaves a significant problem: it provides no insights into to
the nature of the stimuli that can act as stressors. It effectively
leaves us with a definition of stressors as simply being those
stimuli that do elicit a stress response. However, neurobiologists involved in mapping stress neurocircuitry have already
begun to accumulate data consistent with the notion that the
brain categorises stressors and uses at least partially separate,
category-specific neural pathways to generate subsequent stress
responses. Thus there have been many studies that have
provided data suggesting that that the brain uses different
neurocircuitry to generate stress responses to stimuli that cause
actual disturbances of homeostasis (Fphysical_ stressors) versus
stimuli that threaten the individual’s current or anticipated state
(Fpsychological_ stressors) (Dayas et al., 2001; Herman et al.,
2003; Sawchenko et al., 2000). However, as interesting as this
work has been, it remains empirically based. We sorely need a
conceptual framework that offers some means of predicting
which stimuli or situations serve to signify a potential threat to
the individual (still subject, one presumes, to down-grading to
Fnon-stressful status_ based upon a simultaneous appraisal of
context and of resources relevant to meeting the challenge).
One way of addressing this problem may be to adopt an
evolutionary perspective, an approach that has been explored
by psychologists in an attempt to explain the origins of
common psychopathologies (Brune, 2002; Gilbert, 2001).
Applying this approach to the issue of stress, it could be
argued that for a stimulus to be assessed as a psychological
stressor, it must signify to the individual the thwarting of
genetically pre-programmed goals, such goals having evolved
because their pursuit increases the inclusive fitness of the
1200
T.A. Day / Progress in Neuro-Psychopharmacology & Biological Psychiatry 29 (2005) 1195 – 1200
individual. By taking into account the environment of
evolutionary relevance, it may then be possible to develop a
series of testable hypotheses as to the identity of the core goals.
Some likely goals, however, will be immediately apparent, e.g.
position within a social hierarchy, access to a sexual partner,
etc. The task that will need to be tackled is to determine
whether each of these goal systems is underpinned, at least in
some degree, by unique neurocircuitry. Clearly development of
this approach is at a very early stage, but it has the potential to
offer new insights to the organization of the brain’s stress
neurocircuitry.
References
Berridge, K.C., 2004. Motivation concepts in behavioral neuroscience. Physiol.
Behav. 81 (2), 179 – 209.
Brune, M., 2002. Toward an integration of interpersonal and biological
processes: evolutionary psychiatry as an empirically testable framework
for psychiatric research. Psychiatry 65 (1), 48 – 57.
Cannon, W.B., 1929. Organization for physiological homeostasis. Physiol. Rev.
9, 399 – 431.
Cannon, W.B., 1939. The Wisdom of the Body. Revised and Enlarged Edition.
W.W. Norton, New York.
Dallman, M.F., 2003. Stress by any other name . . .? Horm. Behav. 43 (1),
18 – 20.
Dayas, C.V., Buller, K.M., Crane, J.W., Xu, Y., Day, T.A., 2001. Stressor
categorization: acute physical and psychological stressors elicit distinctive
recruitment patterns in the amygdala and in medullary noradrenergic cell
groups. Eur. J. Neurosci. 14 (7), 1143 – 1152.
Gilbert, P., 2001. Evolutionary approaches to psychopathology: the role of
natural defences. Aust. N. Z. J. Psychiatry 35 (1), 17 – 27.
Herman, J.P., Figueiredo, H., Mueller, N.K., Ulrich-Lai, Y., Ostrander, M.M.,
Choi, D.C., Cullinan, W.E., 2003. Central mechanisms of stress integration:
hierarchical circuitry controlling hypothalamo-pituitary-adrenocortical responsiveness. Front. Neuroendocrinol. 24 (3), 151 – 180.
Johnson, E.O., Kamilaris, T.C., Chrousos, G.P., Gold, P.W., 1992. Mechanisms
of stress: a dynamic overview of hormonal and behavioral homeostasis.
Neurosci. Biobehav. Rev. 16 (2), 115 – 130.
McEwen, B.S., 1998. Protective and damaging effects of stress mediators. N.
Engl. J. Med. 338 (3), 171 – 179.
McEwen, B.S., 2000a. Allostasis and allostatic load: implications for
neuropsychopharmacology. Neuropsychopharmacology 22 (2), 108 – 124.
McEwen, B.S., 2000b. The neurobiology of stress: from serendipity to clinical
relevance. Brain Res. 886 (1 – 2), 172 – 189.
McEwen, B.S., Seeman, T., 1999. Protective and damaging effects of mediators
of stress. Elaborating and testing the concepts of allostasis and allostatic
load. Ann N. Y. Acad. Sci. 896, 30 – 47.
McEwen, B.S., Stellar, E., 1993. Stress and the individual. Mechanisms leading
to disease. Arch. Intern. Med. 153 (18), 2093 – 2101.
McEwen, B.S., Wingfield, J.C., 2003a. The concept of allostasis in biology and
biomedicine. Horm. Behav. 43 (1), 2 – 15.
McEwen, B.S., Wingfield, J.C., 2003b. Response to commentaries on the
concept of allostasis. Horm. Behav. 43 (1), 28 – 30.
Sawchenko, P.E., Li, H.Y., Ericsson, A., 2000. Circuits and mechanisms
governing hypothalamic responses to stress: a tale of two paradigms. Prog.
Brain Res. 122, 61 – 78.
Schnorpfeil, P., Noll, A., Schulze, R., Ehlert, U., Frey, K., Fischer, J.E., 2003.
Allostatic load and work conditions. Soc. Sci. Med. 57 (4), 647 – 656.
Schulkin, J., McEwen, B.S., Gold, P.W., 1994. Allostasis, amygdala, and
anticipatory angst. Neurosci. Biobehav. Rev. 18 (3), 385 – 396.
Seeman, T.E., Singer, B.H., Rowe, J.W., Horwitz, R.I., McEwen, B.S., 1997.
Price of adaptation-allostatic load and its health consequences. MacArthur
studies of successful aging. Arch. Intern. Med. 157 (19), 2259 – 2268.
Selye, H., 1936. A syndrome produced by diverse nocuous agents. Nature 138,
32.
Selye, H., 1976. Stress in Health and Disease. Butterworths, Boston.
Selye, H., 1980. Selye’s Guide to Stress Research. Van Nostrand Reinhold Co.,
New York.
Sterling, P., Eyer, J., 1988. Allostasis: a new paradigm to explain arousal
pathology. In: Fisher, S., Reason, J.T. (Eds.), Handbook of Life Stress,
Cognition, and Health. Wiley, Chichester, p. 750.