Autonomic dysregulation and the Window of Tolerance Model of the effects of complex emotional trauma

Abstract

This paper reviews the Window of Tolerance model of the long-term effects of the severe emotional trauma associated with childhood abuse, a model which can also be applied to adult trauma of sufficient severity to cause post-traumatic stress disorder, chronic dysthymic disorders and chronic anxiety disorders. Dysfunctional behaviours such as deliberate self-harm and substance abuse are seen as efforts to regulate an autonomic nervous system which is readily triggered into extreme states by reminders of the original traumatic events. While midbrain areas such as the periaqueductal gray mediate instant defence responses to traumatic events and their memory triggers it is proposed that ascending monoaminergic tracts are implicated in longer-term changes in mood and arousal. An imbalance of ascending dopaminergic tracts may drive rapid fluctuations in level of arousal and in the associated mood, drive and motivation. Animal models of depression frequently use traumatic experiences of pain, isolation or social defeat to induce changes in mesolimbic and mesocortical dopamine systems which may alter prefrontal cortical control of midbrain defence responses. A focus on the pharmacology of the Window of Tolerance could provide advances in drug treatments for promoting emotional regulation in those who are suffering from the chronic sequelae of traumatic experiences.

Full text

Perspective
Autonomic dysregulation and the Window
of Tolerance model of the effects of complex
emotional trauma Journal of Psychopharmacology
0(00) 1–9
!The Author(s) 2010
Reprints and permissions:
sagepub.co.uk/journalsPermissions.nav
DOI: 10.1177/0269881109354930
jop.sagepub.com
FM Corrigan
1
, JJ Fisher
2
and DJ Nutt
3
Abstract
This paper reviews the Window of Tolerance model of the long-term effects of the severe emotional trauma associated with childhood abuse, a model
which can also be applied to adult trauma of sufficient severity to cause post-traumatic stress disorder, chronic dysthymic disorders and chronic anxiety
disorders. Dysfunctional behaviours such as deliberate self-harm and substance abuse are seen as efforts to regulate an autonomic nervous system
which is readily triggered into extreme states by reminders of the original traumatic events. While midbrain areas such as the periaqueductal gray
mediate instant defence responses to traumatic events and their memory triggers it is proposed that ascending monoaminergic tracts are implicated in
longer-term changes in mood and arousal. An imbalance of ascending dopaminergic tracts may drive rapid fluctuations in level of arousal and in the
associated mood, drive and motivation. Animal models of depression frequently use traumatic experiences of pain, isolation or social defeat to induce
changes in mesolimbic and mesocortical dopamine systems which may alter prefrontal cortical control of midbrain defence responses. A focus on the
pharmacology of the Window of Tolerance could provide advances in drug treatments for promoting emotional regulation in those who are suffering
from the chronic sequelae of traumatic experiences.
Keywords
Ascending dopamine systems, autonomic dysregulation, dysfunctional behaviours, trauma, window of tolerance
The psychophysiology of traumatic experience
It has been established that exposure to threat or trauma
stimulates the autonomic nervous system (ANS), resulting
in sympathetic hyperarousal and parasympathetic
(dorsal-vagal-mediated) hypoarousal states accompanying
animal defence survival responses such as fight, flight, sub-
mission and freeze (LeDoux, 2002; Ogden et al., 2006; Porges,
2003; Van der Kolk, 1996a, 1996b). Following cessation of
the threat, many victims continue to suffer from autonomic
sensitivity to stimuli directly or indirectly related to the trau-
matic events. Thus, threatening and traumatic experiences
result in a bewildering array of cognitive, emotional and phys-
iological symptoms: emotions of fear, shame and rage; numb-
ing of feelings and body sensations; overactivity of the stress
response system; and painful, negative beliefs about the self
that serve to intensify the distressing feelings and body
responses. With a dysregulated nervous system that cannot
modulate either heightened emotional states or states of
depression and numbing, patients often report difficulty in
tolerating emotional and physiological arousal without
becoming overwhelmed, as well as problems in recovering
from experiences of intense activation or depression (Ogden
et al., 2006; Van der Kolk, 1996a). In addition, they report
strong somatic responses in which the body tends to become
frozen, collapsed or driven: action becomes either impossible
or impulsive. Non-threatening situational cues often activate
sympathetic nervous system (SNS) activity and fight–
flight responses, while dangerous situations instead elicit
parasympathetic non-responsiveness or submission–compli-
ance responses.
Physiological arousal and the Window of
Tolerance
One model for understanding and explaining the fluctuations
in clinical features that can occur unpredictably and rapidly in
the disorders that arise from the effects of severe trauma
(Ogden et al., 2006) is the ‘Window of Tolerance’ model of
autonomic arousal (Siegel, 1999). Siegel (1999) proposes that
between the extremes of sympathetic hyperarousal and para-
sympathetic hypoarousal is a ‘window’ or range of optimal
arousal states in which emotions can be experienced as toler-
able and experience can be integrated (Figure 1). Although it
has not yet been validated experimentally, the pleomorphic
complex trauma disorders that are the result of childhood
abuse become more comprehensible within this framework
and there is evidence that different emotions are accompanied
by distinct patterns of ANS response with sympathetic
1
Argyll & Bute Hospital, Lochgilphead, Argyll, UK.
2
Center for Integrative Healing, Watertown, Massachusetts, USA.
3
Department of Neuropsychopharmacology and Molecular Imaging,
Imperial College, London, UK.
Corresponding author:
FM Corrigan, Argyll & Bute Hospital, Lochgilphead, Argyll PA31 8LD, UK.
Email: frank.corrigan@nhs.net
J Psychopharmacol OnlineFirst, published on January 21, 2010 as doi:10.1177/0269881109354930

activation accompanied by varying degrees of parasympa-
thetic deactivation (Rainville et al., 2006). Other studies
have observed that trauma-related symptoms and emotions
may be accompanied by more extreme autonomic effects
(Lanius, 2005; Perry et al., 1995). In a study of individuals
with borderline personality disorder, a diagnosis that has
been correlated with childhood trauma (Herman et al.,
1989; Ogata et al., 1990; Zanarini et al., 1989) subjects
demonstrated sympathetically-driven active defence responses
to social interaction with a stranger, in contrast with controls
who were more able to readily engage the ‘vagal brake’ (the
ventral vagal complex) for rapid attunement with the other
person (Austin et al., 2007). While emotional dysregulation
driving maladaptive efforts to diminish distress has provided
a useful model for treating borderline personality disorder
with dialectical behaviour therapy (Linehan, 1993), and can
be adapted to many situations in which harmful actions occur
in response to distress, the autonomic dysregulation model
allows additional hypotheses about the associated psychophy-
siology of involuntary somatic responses to triggers that
evoke some component of a trauma memory.
Trauma, negative cognitions and affective
states outside the Window of Tolerance
When a traumatic event is overwhelming and inescapable, the
associated high arousal emotional state may not be easily
modulated, leading to inhibition of cortical activity
(LeDoux, 2002; Ogden et al., 2006), a loss of cognitive
witnessing, and perhaps even to states of speechless rage or
blind terror (Van der Kolk, 1996b). Alternatively, the dorsal
vagal branch of the parasympathetic nervous system (PNS)
may become activated (Porges, 1995) leading to a state of
total submission characterized by numbing, collapse,
dissociation or even ‘feigned death’ (Porges, 2003), a state
from which the prospect of dying can seem a welcome release.
The PNS-dominant state of near death is accompanied by
flaccidity of the muscles and is therefore distinguishable
from the tonic immobilization state (‘scared stiff’ or cata-
tonic) (Moskowitz, 2004), in which high physiological arousal
is combined with an inability to engage in voluntary move-
ment of the body, a peritraumatic orienting and defence
response postulated to be secondary to simultaneous SNS
and PNS co-activation (Ogden et al., 2006). Peritraumatic
tonic immobility has been associated with severe psychologi-
cal sequelae of depression and anxiety years after the child-
hood sexual abuse in which it was experienced, when fear and
helplessness were combined with an inability to escape from
the horrifying situation (Bovin et al., 2008; Heidt et al., 2005).
More recently the occurrence of tonic immobility in male
victims of urban violence has been associated with a poor
response to drug therapy (Fiszman et al., 2008).
This Window of Tolerance model (Ogden et al., 2006;
Siegel, 1999) describes all these animal defensive states
(Figure 1): the hyperarousal connected to sympathetic activa-
tion, the autonomic hypoarousal characteristic of parasym-
pathetically mediated dorsal vagal responses, and the ‘deer in
the headlights’ frozen flight response connected to high SNS
and PNS co-activation. In addition, the Window of Tolerance
as an explanatory conceptualization also prescribes a treat-
ment approach: that of regulating autonomic arousal within a
Window of Tolerance in which affect and cognition can be
tolerated so that patients can both think and feel (Fisher and
Ogden, 2007). When patients with trauma-related disorders,
such as post-traumatic stress disorder (PTSD) and borderline
personality disorder, develop greater ability to self-regulate
autonomic arousal, symptoms tend to diminish or ameliorate,
and they are able to engage more effectively in
well-established treatments for trauma. In addition, use of a
Sympathetic-dominant Hyperarousal:
Emotionally flooded, reactive, impulsive, hypervigilant, fearful, angry.
Intrusive imagery and affects, racing thoughts
Flashbacks, nightmares, high-risk behaviour
Efforts to reduce this state may include suicide planning, self harm,
compulsive cleaning, abuse of alcohol or opiates
Parasympathetic-dominant Hypoarousal:
Flat affect, numb, “empty” or “dead”
Cognitively dissociated, inability to think
Collapsed, disabled defensive responses
Helpless and hopeless
Efforts to reduce may include suicide planning, self-harm, compulsive
Window of Tolerance
Optimal arousal zone, encompassing both intense emotion
and states of calm or relaxation, in which emotions can be
tolerated and information integrated
Freeze
Mute, terrified, frozen
defence responses.
High arousal coupled
with physical immobility*
Figure 1. Autonomic arousal in the wake of trauma: sympathetic hyperarousal and parasympathetic hypoarousal states drive emotional and autonomic
dysregulation. Also shown is the frozen fear state in which both extremes may be present. States of optimal arousal and emotional regulation are
relatively rare or difficult to maintain. Some examples are included of how dysfunctional behaviours can be utilised in the service of emotional
regulation: both to reduce the intensity of the high arousal states and the depth of the low arousal states. Adapted from Ogden et al (2006) and Siegel
(1999).
2Journal of Psychopharmacology 0(00)

model that emphasizes the role of the ANS in perpetuating
post-traumatic symptoms addresses both physiological hyper-
arousal symptoms (high-risk behaviour, suicidal impulses,
self-injury) and hypoarousal symptoms (shame, dysthymia,
depression, loss of energy, self-loathing) (Figure 1).
Emotions, defence responses and the mid-
brain periaqueductal grey
One hypothesis offering an explanatory model for these
effects of traumatic experiences on arousal levels involves
the midbrain periaqueductal grey (PAG), which is a key struc-
ture for mediating the physiological effects of defence
responses. Dorsolateral columns are involved in active, sym-
pathetically driven fight or flight behaviours while ventrolat-
eral areas are engaged in parasympathetically dominant
freeze reactions (Bandler and Shipley, 1994; Watt, 2000).
Recently it has been proposed that the dorsomedial column
has specificity for the modulation of aversive, avoidant states
with corticotrophin-releasing factor (CRF) exerting an exci-
tatory influence on this region (Borelli and Brandao, 2007). It
will be of interest to see whether more precise techniques for
imaging the brainstem reveal an association of the dorsome-
dial column with trauma-related avoidant responses of dis-
gust and shame as activation of the dorsomedial column leads
to conditioned place aversion (Zanoveli et al., 2007). The
Window of Tolerance model displayed in Figure 1 sees
freeze responses as the result of co-activation of sympathetic
and parasympathetic components of the ANS and this, in
PAG terms, would mean co-activation of dorsolateral and
ventrolateral columns. However, dorsomedial PAG stimula-
tion for conditioned place aversion is associated with
increased open field freezing, possibly of the attentive,
information-gathering type (Zanoveli et al., 2007), and there
are two other types of freezing (stimulus bound and
non-stimulus bound) produced by activation of dorsal aspects
of PAG (Brandao et al., 2008). Thus it is possible that there
are several different types of freeze response depending on
which dorsal PAG columns are co-activated with the
low-arousal, opioid-mediated ventrolateral area responsible
for conditioned freeze with its bradycardia, hypotension
and energy-conservation functions. The PAG activation of
other brainstem structures mediates the autonomic changes
required by emotions and defence responses and the resulting
changes in the physiological state of the body are fed back
through the thalamus to insular and anterior cingulate corti-
ces (Craig, 2005). Distressing emotions associated with the
sympathetic arousal of trauma, such as fear, anger and dis-
gust, are also dependent on activation of the right anterior
insular cortex. The concomitant opponent deactivation of the
left insular cortex (Craig, 2005) may lead to loss of feelings of
comfort and safety and difficulties in articulating the experi-
ence. During treatment with a safe, empathic therapist
encouraging non-judgemental mindfulness, and through
training in skills for soothing self-regulation (e.g. Fay,
2007), the balance of insula activation could move to
left-dominance.
Motor aspects of defence responses will require subcortical
circuits through the PAG similar to those described by
McHaffie et al. (2005). The behaviours associated with fight
or flight have sufficient similarity in different people to suggest
that there are intrinsic tendencies to activate particular muscle
groups when the situation requires specific actions to ensure
survival. Grillner et al. (2005) describe a toolbox of motor
programs which is kept under tonic inhibition by gamma
aminobutyric acid (GABA)ergic pallidal neurons (including
those of the substantia nigra pars reticulata) until a selection
is made in the striatum. A failure of selection from two dis-
inhibited programs would lead to the state of frozen flight or
frozen fight accompanying the ANS co-activation.
The neurochemistry of PAG activation is complex, invol-
ving, to select a few examples, 5-hydroxytryptamine (5HT)-
activated inputs from the basolateral amygdala (Martinez
et al., 2007), cholinergic stimulation of the central amygdala
(Leite-Panissi et al., 2003), GABA-A modulated 5HT activa-
tion of the ventrolateral orbital prefrontal cortex (Huo et al.,
2008) and intrinsic acetylcholine (Kroes et al., 2007), GABA
(De Menezes et al., 2006; Reimer et al., 2008), cytokines
(Bhatt et al., 2008), opioids (Kishimoto et al., 2001), canna-
binoids (Terzian et al., 2008), CRF (Carvalho-Netto et al.,
2007), glutamate and 5HT (Moraes et al., 2008).
Affect regulation and the Window of Tolerance
During multiple or prolonged exposures to childhood trau-
matic events (including sexual abuse, physical abuse, expo-
sure to domestic violence, and trauma related to medical
treatments or accidents), hyper- and hypoarousal states
occur not only in response to threatening events but also to
their anticipation. After repeated SNS/PNS activation during
formative years, the adult in later life may have difficulties
in regulating affect (Ogden et al., 2006; Schore, 2003;
Van der Kolk, 1996a), resulting in prolonged states of fear
and anger (SNS-dominance) or despair and depression
(PNS-dominance). In the affectively flattened hypoaroused
state, the trauma survivor may feel that life is not worth
living, that nothing matters, that death would be a relief.
There is a lack of energy and enthusiasm, and a reported
absence of distress may coexist with the matter-of-fact
suicidal thinking. This presentation may mimic, or coexist
with, a major depressive episode, and prescription of antide-
pressants is the obvious choice for the clinician. If antidepres-
sant use leads to an increase in anxiety and arousal, there can
be a switch to a hyperaroused state (in which the energy for
a suicide attempt is available) or a rapid alteration between
the extremes. If suicidal thinking is initially a way of soothing
the empty nihilism and despair of the hypoaroused state
(Herman, 1992), antidepressant-related autonomic activation
may result in the ideas taking on fresh urgency as a way to
end intense suffering. Similarly, a temporary intensification of
the hypoarousal state would increase suicidal thinking, which,
if followed by increased energy, could result in impulsive
acting out. The converse may also occur when suicidal think-
ing is initially a way of soothing intense activation, unbear-
able impulses and overwhelming emotions. Increased focus
and autonomic stability as a result of increased serotonin
reuptake inhibition may result in more ability to plan and
implement actions.
Corrigan et al. 3

Ascending tracts and levels of arousal
Although the acute physiological effects of traumatic events
and post-traumatic triggers can be attributed to the SNS/PNS
accompaniments of defence responses, the sustained effects on
cognition and emotion require an alternative explanation for
the alteration of the ‘torque’ (Watt, 2000) of the thalamocor-
tical mantle. Key areas are the intralaminar nuclei, the
nucleus reticularis thalami, the superior colliculi, the core
monoamine nuclei in the brainstem, and the midbrain retic-
ular formation, with PAG outputs having modulatory influ-
ences on these significant structures (Watt, 2000). Ascending
dopaminergic tracts such as the mesocortical, nigrostriatal
and mesolimbic systems are good candidates for this role.
The innervation from the midbrain ventral tegmental area
(VTA) reaches not only the nucleus accumbens and ventral
pallidum but also areas involved in the memory of emotional
experiences and learning of fear responses, such as the baso-
lateral nucleus of the amygdala (McGaugh, 2004). Also, the
mesodiencephalic areas involved in the generation of emo-
tions and defence responses have bilateral projections to the
thalamus that are separate from the nigrostriatal, mesolimbic
and mesocortical dopamine systems (Sanchez-Gonzalez et al.,
2005) and have more diverse origins. The PAG projections to
the mediodorsal nucleus, the principal thalamic input to the
orbitomedial prefrontal cortex, are likely to be significant for
affectively loaded mood states and behaviours but a circuit
involving the laterodorsal nucleus of the thalamus, the baso-
lateral amygdala and the dorsal PAG has also been proposed
for panic disorder (Zanoveli et al., 2007).
Long-term administration of antidepressant drugs attenu-
ates the footshock stress response of mesocortical dopamine
neurons in the rat (Dazzi et al., 2001). The atypical antipsy-
chotics olanzapine and clozapine have a similar effect (Dazzi
et al., 2004), which may help to explain why drugs such as
quetiapine and olanzapine are often clinically useful for
widening the Window of Tolerance in complex trauma
patients. The persistent state of behavioural depression fol-
lowing a prolonged period of repeated immersion stress which
is accompanied by reduced dopamine transmission in the rat
prefrontal cortex (Mizoguchi et al., 2008) may serve as a
model for the hypoarousal state.
Stimulation of dopamine D2 receptors in the prefrontal
cortex reduces, whereas N-methyl D-aspartate (NMDA)
antagonism in the prefrontal cortex increases, the release of
dopamine in the nucleus accumbens and the spontaneous
motor activity of the rat (Del Arco and Mora, 2008), support-
ing the hypothesis that an imbalance of dopamine transmis-
sion in mesocortical and mesolimbic systems would have
effects on motor activation and arousal.
Dysfunctional behaviours and the Window
of Tolerance
Illicit drugs that activate the mesolimbic dopamine (ML-DA)
system and provide reward through activation of the shell of
the nucleus accumbens include cocaine and amphetamine
(Alcaro et al., 2007). One explanatory hypothesis for the
development of substance abuse, self-destructive behaviour
and other dysfunctional behaviours in patients with a history
of trauma is that these substances and the associated compul-
sive activities seek to regulate low- and high-arousal states.
The use of alcohol and cannabis to reduce high activation
states is more difficult to explain if the only relevant neuro-
biology relates to the rewarding stimulation in the nucleus
accumbens shell. However, alcohol and benzodiazepines
would reduce arousal through a GABAergic inhibition of
the ML-DA system, and GABA interneurons are responsible
for different degrees of activation of the PAG columns
(Watt, 2000).
Suicide planning is used by trauma patients to both acti-
vate and deactivate arousal, depending on the baseline state
(Perry and Szalavitz, 2007). This soothing or activating
scenario-construction is likely to require imagery circuitry in
the visual cortex, precuneus and posterior cingulate cortex
but also important may be the dopaminergic projections to
the prefrontal cortex which, according to Alcaro et al. (2007),
facilitate information processing without the affective aspects
of Seeking, potentially an increase in arousal without positive
affect. Given that these mesocortical projections can inhibit
dopamine release in the nucleus accumbens (Alcaro et al.,
2007) so the prefrontal areas will be critical to the direction
of the modulation.
Bathing, washing and cleaning are often used by complex
trauma survivors to avoid the dysphoria associated with
hyper- or hypoarousal states. The association between child-
hood trauma and adult obsessive-compulsive symptoms
has recently been documented from a large study of college
students (Mathews et al., 2008) and an affective orbitofron-
tal–striatal circuit has been implicated in obsessive-
compulsive disorder (Menzies et al., 2008). A wide range of
obsessive-compulsive behaviour functions to either increase
cortical activity (as in counting and checking) or regulate arou-
sal (as in bathing, washing and cleaning rituals).
The valence of the hyperarousal state is also subject to
fluctuations and dysfunctional behaviours may be attempts
to convert a fearful, negatively valenced hyperarousal to a
more positive appetitive state. Reynolds and Berridge (2008)
have demonstrated that the shell of the rat nucleus accumbens
is highly sensitive to environmental influences such as bright
light and loud music, and the behaviours generated by dis-
ruption of glutamate transmission there are readily converted
from appetitive to fearful. There is preliminary evidence from
functional magnetic resonance imaging (fMRI) studies of
altered processing of rewards in the nucleus accumbens and
prefrontal cortex in PTSD (Sailer et al., 2008).
Rapid oscillation between extremes of arousal state and
frantic efforts to achieve regulation, even by way of dysfunc-
tional behaviours, can lead to a ‘biphasic rollercoaster’ of
experience which is chaotic and disturbing to the sufferer
(Figure 2).
Trauma and animal models of depression
Animal models of depression that have highlighted the impor-
tance of dopaminergic tracts have relied on the traumas of
pain, isolation or defeat. Rearing rat pups in social isolation
induces hyperfunction of ML-DA systems and hypofunction
of mesocortical dopamine (Fone and Porkess, 2008). Similar
4Journal of Psychopharmacology 0(00)

observations are made when tail pressure is used as the stres-
sor: dopamine depletion in the medial prefrontal cortex
potentiates the stress-induced increase in extracellular dopa-
mine in the nucleus accumbens shell (King et al., 1997).
Rats that repeatedly demonstrated defensive or submissive
behaviour when placed in the cage of an aggressive resident
rat were found to have increased dopamine levels in the
nucleus accumbens and prefrontal cortex when they were
again exposed to the threat of social defeat (Tidey and
Miczek, 1996), providing support for the hypothesis that
the hypervigilant, hyperaroused state is associated with
increased extracellular dopamine in the nucleus accumbens
and prefrontal cortex. However, rats that were returned to
the safety of their home cage (not in sight of the aggressor)
were found to have higher dopamine concentrations while
being behaviourally predominantly inactive, suggesting that
they were in a state of aroused vigilance rather than in a
post-traumatic hypoarousal condition. The aversion to
social contact of socially defeated mice is normalized by
chronic administration of antidepressant (Berton et al.,
2006), supporting the use of this animal trauma response
model in modelling depression and in the discovery of new
antidepressants.
Thus, trauma can lead to increased mesolimbic dopamine
and reduced or increased mesocortical dopamine. Also, the
dopaminergic tracts from PAG through the dorsomedial tha-
lamus to prefrontal cortex which have been demonstrated in
the macaque monkey (Sanchez-Gonzalez et al., 2005) would
warrant further study in the traumatized human.
Prefrontal cortex and the autonomic nervous
system
The orbitomedial prefrontal cortex has projections, presum-
ably regulatory, to the mesodiencephalic areas necessary for
the generation of core emotions and defence responses
(Ongur and Price, 2000). Animal studies suggest that in
some species parasympathetic responses, with bradycardia
and respiratory inhibition, are elicited from the anterior cin-
gulate cortex while sympathetic responses are elicited from
more ventral regions of subcallosal and orbital prefrontal
cortex (Buchanan and Powell, 1993). Sympathetic responses
can also be elicited by stimulation of midline and mediodorsal
thalamic nuclei and these structures may therefore be able to
counteract parasympathetic changes mediated by the anterior
cingulate and orbital prefrontal cortex (Buchanan and
Powell, 1993). In the rat there are direct projections from
the anterior cingulate and orbitofrontal cortex to the dorsal
motor nucleus of the vagus, the solitary nucleus and the
nucleus ambiguus of the medulla. In the rabbit there are
also projections to the mediodorsal, midline and intralaminar
nuclei of the thalamus. Given that a direct pathway exists
from areas 32 and 25 to the autonomic regulatory nuclei in
the medulla it has been proposed that prefrontal cortex and
thalamic activity is associated not with reflex homeostatic
changes in autonomic activity but with the more complex
changes involved in associative learning. The anterior cingu-
late cortex, the orbitofrontal cortex and the mediodorsal
nucleus of the thalamus all have reciprocal connections with
the basolateral nucleus of the amygdala (Buchanan and
Powell, 1993).
Functional brain imaging in the human confirms the role
of both the anterior cingulate cortex and the ventral prefron-
tal cortex in control of autonomic responses. Stimulation of
the sympathetic system with exercise or arithmetic is accom-
panied by activation in the dorsal anterior cingulate cortex
(Critchley et al., 2000), while parasympathetic modulation is
accompanied by activation in the ventromedial prefrontal
cortex (Nagai et al., 2004; Wong et al., 2007). The vagal
brake (Porges, 1995) may have its hardwiring in the human
brain in the ventromedial prefrontal cortex rather than the
anterior cingulate cortex as would be expected from the
animal studies referred to above. It is not yet possible to attri-
bute these prefrontal activations unequivocally to specific
autonomic effects or to identified neurotransmitter systems.
However, Bergmann (2008), echoing and developing the
work of Schore (1994), has proposed that post-traumatic
hyperarousal states are mediated by a dopaminergic circuit
involving a ventral sympathetic area of orbitofrontal cortex
that has reciprocal dopaminergic connections with the ventral
tegmental area and projections to the nucleus accumbens.
This system is often involved in positively valenced states of
appetitive behaviour, motivational reward and active coping,
but could be negatively valenced under conditions of threat
(Reynolds and Berridge, 2008). The post-traumatic hypoar-
ousal state, in contrast, is considered by Bergmann (2008) to
be mediated by a noradrenergic circuit involving lateral
regions of orbitofrontal cortex that have reciprocal connec-
tions with, amongst others, parasympathetic autonomic areas
Flashback
terror Eating
for
comfort
I’m fat
disgusting
a failure
Isolation
predisposes
to memory
Flashback
rage
Shame
Self-directed
rage
Alcohol
intoxication
Withdrawal
nightmares
BIPHASIC
ROLLERCOASTER
Figure 2. An example of how dysfunctional behaviours can be efforts to
regulate distress by a person striving to be within the Window of
Tolerance. Eating for comfort to soothe the terror of a flashback is ini-
tially helpful but then leads to feelings of shame, self-loathing and
worthlessness. This gives rise to suicidal thinking but also to withdrawal
and social isolation which make intrusive memories more disturbing. The
next flashback is accompanied by rage then by shame at the rage, leading
to alcohol intoxication to promote oblivion. Withdrawal from the alcohol
leads to a high arousal state in which triggers to further flashbacks with
dominant fear or rage will occur, continuing the cycle.
Corrigan et al. 5

of the lateral hypothalamus and arousal-regulating neurons in
brainstem vagal centres. This lateral tegmental limbic fore-
brain–midbrain circuit down-regulates negative affect by acti-
vating the onset of a parasympathetic inhibitory state which
can be positively valenced with soothing and calming or neg-
atively valenced with apathy and shame.
Brain imaging in complex trauma and in
depression
Preliminary findings suggest that patients with complex
trauma and dissociative disorders have smaller hippocampal
volumes than controls and that the degree of reduction in
hippocampal volume is correlated both with the clinical sever-
ity of the disorder and the extent of early trauma (Ehling
et al., 2008). Similar findings were reported in a study of
women with borderline personality disorder and a history
of early life trauma (Driessen et al., 2000). Despite these find-
ings, imaging studies of hippocampal volume in depression
are often conducted with no recording of the trauma history
of the participants and this is acknowledged to be a possible
confounding variable in efforts to link volume changes with
depressive subtype (Greenberg et al., 2008). Brain research by
Lanius et al. (cited in Ogden et al., 2006) demonstrated
reduced thalamic activation in subjects with histories of
trauma relative to controls but this has not yet been studied
in relation to arousal level.
The immediacy of hopelessness and chronic
depression
In some cases, depressive episodes appear to be linked to
years of bullying, neglect or other abuse. Psychodynamic the-
orists considered internalized anger to be capable of transfor-
mation into depressed mood but they provided no account of
how this might be mediated physiologically. If there are
repeated situations in which a fight response is obstructed,
thoughts of hopelessness, helplessness and worthlessness
accompanying the parasympathetic down-regulation of the
fight response would be associated with low mood – and a
mental state that is biologically determined for a brief and
immediate response to attack can then become chronic.
During an attack by a predator there is a need for a focus
on immediate response rather than on long-term planning.
The function of hopelessness, the belief that ‘this is never
going to get better’, may be the disconnection of the situa-
tionally redundant fight and flight responses. This outlook
conserves energy, through avoidance, hiding or other appro-
priate behavioural responses, and also facilitates submission
if there is a clear imbalance of power.
The physiological experience of helplessness (loss of
energy and tone, collapse, slowed cognitive processing),
accompanied by the cognition ‘there is nothing I can do:
no action of mine will be helpful’ promotes resignation
and energy conservation. Like hopelessness and worthless-
ness (‘I deserve this because I’m bad’) it reinforces submis-
sion responses and passive compliance or acceptance. Thus,
symptoms of depression and hopelessness reflect the
experience of incomplete and/or obstructed fight responses
automatically followed by parasympathetically mediated
‘total submission’ (Porges, 1995) responses. Energy conser-
vation allows individuals to sustain such states for lengthy
periods of time. Subsequently, the patient’s subjective expe-
rience is one of depression, hopelessness and worthlessness,
which is often exacerbated in the context of anger or
self-assertion.
Energy conservation can be seen in the DSM-IV (APA,
1994) criteria for major depressive episode of reduced energy
and fatigue. Low mood is described as feeling sad or empty,
and feelings of worthlessness or inappropriate guilt may be
present. Feelings of hopelessness feature in the DSM-IV cri-
teria for dysthymic disorder, along with other features of a
low arousal state such as low energy and low self-esteem.
Concurrent experience of anxiety and depression then
would not be a question of co-morbidity but of activated
fear simultaneous with the incomplete fight or flight
responses. When these are constantly present, rather than
available briefly for situations of danger, the short-term per-
spective remains and obviates the possibility of optimism.
How can there be enjoyment, interest, energy or pleasure
when the outlook rises from a state of frozen submission?
This result of social defeat stress, as in the rat model of
Berton et al. (2006), is a chronic depressive condition reflect-
ing the survival need for dorsal vagal and parasympathetic
dominance.
Because trauma-related stimuli continue to evoke sympa-
thetically mediated hyperarousal responses and the ‘vagal
brake’ to inhibit or truncate fight or flight responses,
trauma patients often experience characteristic fluctuations
of state described in the ‘distress dipole’ model. If fight or
flight responses chronically coexist with a hypoarousal state
that down-regulates to truncate active defensive responses,
patients may become trapped in the ‘distress dipole’ and
become unable to access any positive affective state. Some
examples are given in Figure 3.
Summary: psychopharmacology and the
Window of Tolerance
We are arguing that while arousal states outside the window
of tolerance are initially associated with SNS or PNS activa-
tion, the psychological effects that persist beyond the acute
change in heart rate and respiratory rate appear to be
mediated – at least in part – by ascending dopaminergic
tracts from the midbrain to the thalamus and the mesocorti-
cal, nigrostriatal and mesolimbic dopaminergic systems.
Evidence from animal studies supports the view that trauma
induces excessive dopamine release in mesolimbic systems,
but the role of reduced dopaminergic activity in mesocortical
tracts is less clear. It may be an imbalance between mesocor-
tical and mesolimbic dopamine transmission that facilitates
rapid switching between high and low arousal states in those
with a history of severe trauma. The nucleus accumbens is
sensitive to environmental factors (Reynolds and Berridge,
2008) and the experience of the arousal state as positively
or negatively valenced may relate to whether it is in fearful/
defensive or appetitive mode.
6Journal of Psychopharmacology 0(00)

High arousal states that are aversive can be modulated by
suicide planning, starvation, abuse of alcohol and cannabis,
bathing, grooming and compulsive cleaning, and by
self-harm. Low arousal states can be modified by compulsive
activities involving grooming and cleaning, suicide planning,
risk-taking (such as driving too fast), self-harm, and abuse of
alcohol, amphetamine, ecstasy or cocaine. Treatment
approaches need to involve patient education about these
states and their role in symptom perpetuation and exacerba-
tion. In addition to pharmacological approaches, the substi-
tution of more effective and less harmful behaviours to
regulate autonomic states must be emphasized. For both
hyperarousal and hypoarousal, mindfull curiosity is regulat-
ing, as is psychoeducation. In high arousal, soothing strate-
gies may be useful, as well as exercise, yoga and other physical
activities requiring focused attention. Low arousal benefits
from gentle activity, movement, humour and increased corti-
cal activity. In suicidal patients with both low and high arou-
sal patterns, treatment with antidepressants and atypical
antipsychotics needs to be monitored carefully from the per-
spective of autonomic arousal and the ‘Window of Tolerance’
to maximize the psychopharmacological benefits in this pop-
ulation and reduce the possibility of an increase in suicidal
behaviour.
References
Alcaro A, Huber R, and Panksepp J (2007) Behavioral functions
of the mesolimbic dopaminergic system: an affective
DISTRESS DIPOLES
Sympathetic
hyperarousal
FLIGHT
Fear and hypervigilance: ‘I
can’t be trapped. I can’t get
too close or committed. I
have to be ready to run.’ ‘If
I can’t run, I can escape
with drink or drugs or
eatin
g
disorders.’
FIGHT
Anger and self-blame:
‘I shouldn’t have let it
happen. It will not happen
again. I won’t trust. I won’t
be controlled.’ FIGHT
Anger and hypervigilance: ‘It will
happen again, but now I’m
ready. I will push people away –
I’ll test their credibility. And if
necessary, I’ll kill myself, or I’ll
kill them.’
Parasympathetic hypoarousal
ATTACH
‘I can’t take care of myself
– I’m in danger, and I need
help. Please love me and
care for me and protect
me.’ ‘Please don’t leave .
. . ‘
SUBMIT
Safety lies in compliance. .
‘I can’t say No. I always
have to give in.’ ‘If I go
numb and compliant, I
won’t be hurt again.’ ‘It
isn’t safe to assert myself
or ask for help – it’s
dangerous to fight.’
SUBMIT
Shame and depression increase
compliance and thus safety: ‘I deserved
it, or it wouldn’t have happened.’
‘I’m giving up because I don’t deserve
more.’ ‘I’m too tired to fight.’ ‘There’s no
point – it’s hopeless.’ ‘I’m weak and
worthless.’
WINDOW OF TOLERANCE
Figure 3. Some examples of states in which the ‘rollercoastering’ is so rapid that the high and low arousal states effectively coexist. The negative
cognitions are in matching pairs, each with an opposite arousal state, giving rise to an extremely unpleasant dynamic dysphoric tension.
Corrigan et al. 7

neuroethological perspective. Brain Res Rev doi: 10.1016/j.brain-
resrev.2007.07.014.
American Psychiatric Association (APA). (1994) Diagnostic Criteria
from DSM-IV – Quick Reference. Washington DC: American
Psychiatric Association.
Austin MA, Riniolo TC, Porges SW (2007) Borderline personality
disorder and emotion regulation: insights from the polyvagal
theory. Brain Cogn 65: 69–76.
Bandler R, Shipley MT (1994) Columnar organization in the mid-
brain periaqueductal gray: modules for emotional expression?
Trends Neurol Sci 17: 379–389.
Bergmann U (2008) She’s come undone: a neurobiological explora-
tion of dissociative disorders. In: Forgash C, Copeley M (eds)
Healing the Heart of Trauma and Dissociation: with EMDR and
Ego State Therapy. New York: Springer, 61–90.
Berton O, McClung CA, DiLeone RJ, et al. (2006) Essential role of
BDNF in the mesolimbic dopamine pathway in social defeat
stress. Science 311: 864–868.
Bhatt S, Bhatt R, Zalcman SS, and Siegel A (2008) Role of IL-1beta
and 5-HT2 receptors in midbrain periaqueductal gray (PAG) in
potentiating defensive rage behavior in cats. Brain Behav Immun
22: 224–233.
Borelli KG and Brandao ML (2007) Effects of ovine injections into
the dorsomedial, dorsolateral and lateral columns of the peria-
queductal gray: a functional role for the dorsomedial column.
Horm Behav, doi: 10.1016/j.yhbeh.2007.08.013.
Bovin MJ, Jager-Hyman S, Gold SD, Marx BP, Sloan DM (2008)
Tonic immobility mediates the influence of peritraumatic fear and
perceived inescapability on posttraumatic stress symptom severity
among sexual assault survivors. J Traum Stress 21: 402–409.
Brandao ML, Zanoveli JM, Ruiz-Martinez RC, Oliveira LC,
Landeira-Fernandez J (2008) Different patterns of freezing beha-
viour organized in the periaqueductal grey of rats: association
with different types of anxiety. Behav Brain Res 188: 1–13.
Buchanan SL, Powell DA (1993) Cingulothalamic and prefrontal
control of autonomic function. In: Vogt BA, Gabriel M (eds)
Neurobiology of Cingulate Cortex and Limbic Thalamus. Boston:
Birkhauser, 381–414.
Carvalho-Netto EF, Litvin Y, Nunes-de-Souza RL, Blanchard DC,
Blanchard RJ (2007) Effects of intra-PAG infusion of ovine CRF
on defensive behaviours in Swiss-Webster mice. Behav Brain Res
176: 222–229.
Craig AD (2005) Forebrain emotional asymmetry: a neuroanatomi-
cal basis? Trends Cogn Sci 9: 566–571.
Critchley HD, Corfield DR, Chandler MP, Mathias CJ, Dolan RJ
(2000) Cerebral correlates of autonomic cardiovascular arousal: a
functional neuroimaging investigation in humans. J Physiol 523:
259–270.
Dazzi L, Serra M, Spiga F, Pisu MG, Jentsch JD, Biggio G (2001)
Prevention of the stress-induced increase in frontal cortical dopa-
mine efflux of freely moving rats by long-term treatment with
antidepressant drugs. Eur Neuropsychopharm 11: 343–349.
Dazzi L, Seu E, Cherchi G, Biggio G (2004) Inhibition of
stress-induced dopamine output in the rat prefrontal cortex by
chronic treatment with olanzapine. Biol Psychiatry 55: 477–483.
Del Arco A, Mora F (2008) Prefrontal cortex-nucleus accumbens
interaction: in vivo modulation by dopamine and glutamate in
the prefrontal cortex. Pharmacol Biochem Behav 90: 226–235.
De Menezes RCA, Zaretsky DV, Fontes MAP, DiMicco JA (2006)
Microinjection of muscimol into caudal periaqueductal gray
lowers body temperature and attenuates increases in temperature
and activity evoked from the dorsomedial hypothalamus. Brain
Res 1092: 129–137.
Driessen M, Herrmann J, Stahl K, et al. (2000) Magnetic resonance
imaging volumes of the hippocampus and the amygdala in women
with borderline personality disorder and early traumatization.
Arch Gen Psychiatry 57: 1115–1122.
Ehling T, Nijenhuis ERS, Krikke AP (2008) Volume of discrete brain
structures in complex dissociative disorders: preliminary findings.
Prog Brain Res 167: 307–310.
Fay D (2007) Becoming Safely Embodied. Somerville, MA: Heart Full
Life Publishing.
Fisher J, Ogden P (2007) Sensorimotor psychotherapy. In: Courtois
CA, Ford JD (eds) Treating Complex Post-traumatic Disorders.
New York: Guilford Press, 312–328.
Fiszman A, Mendlowicz MV, Marques-Portella C, et al. (2008)
Peritraumatic tonic immobility predicts a poor response to phar-
macological treatment in victims of urban violence with PTSD.
J Affect Disord 107: 193–197.
Fone KCF, Porkess MV (2008) Behavioural and neurochemical
effects of post-weaning social isolation in rodents – relevance to
developmental neuropsychiatric disorders. Neurosci Biobehav Rev
37: 1087–1102.
Greenberg DL, Payne ME, MacFall JR, Steffens DC, Krishnan RR
(2008) Hippocampal volumes and depression subtypes. Psych
Res: Neuroimag 163: 126–132.
Grillner S, Hellgren J, Menard A, Saitoh K, Wilkstrom MA (2005)
Mechanisms for selection of basoc motor programs – roles for the
striatum and pallidum. Trends Neurosci 28: 364–370.
Heidt JM, Marx BP, Forsyth JP (2005) Tonic immobility and chid-
hood sexual abuse: a preliminary report evaluating the sequela of
rape-induced paralysis. Behav Res Therap 43: 1157–1171.
Herman JL (1992) Trauma and Recovery. New York: W.W. Norton
& Sons.
Herman JL, Perry JC, van der Kolk BA (1989) Childhood trauma in
borderline personality disorder. Am J Psychiatry 146: 490–495.
Huo F-Q, Qu C-L, Li Y-Q, Tang J-S, Jia H (2008) GABAergic mod-
ulation is involved in the ventrolateral orbital cortex 5-HT1A
receptor activation-induced antinociception in the rat. Pain 139:
398–405.
King D, Zigmond MJ, Finlay JM (1997) Effects of dopamine deple-
tion in the medial prefrontal cortex on the stress-induced increase
in extracellular dopamine in the nucleus accumbens core and
shell. Neuroscience 77: 141–153.
Kishimoto K, Koyama S, Akaike N (2001) Synergistic mu-opioid
and 5-HT1A presynaptic inhibition of GABA release in rat peri-
aqueductal gray neurons. Neuropharmacology 41: 529–538.
Kroes RA, Burgdorf J, Otto NJ, Panksepp J, Moskal JR (2007)
Social defeat, a paradigm of depression in rats that elicits
22-kHz vocalizations, preferentially activates the cholinergic sig-
nalling pathway in the periaqueductal gray. Behav Brain Res 182:
290–300.
Lanius UF (2005) EMDR processing with dissociative clients:
adjunctive use of opioid antagonists. In: Shapiro R (ed.) EMDR
Solutions: Pathways to Healing. New York: Norton.
LeDoux J (2002) The Synaptic Self: How our Brains Become Who
We Are. New York: Guilford Press.
Leite-Panissi CRA, Coimbra NC, Menescal-de-Oliveira L (2003) The
cholinergic stimulation of the central amygdala modifying the
tonic immobility response and antinociception in guinea pigs
depends on the ventrolateral periaqueductal gray. Brain Res
Bull 60: 167–178.
Linehan MM (1993) Skills Training Manual for Treating Borderline
Personality Disorder. New York: Guilford Press.
Martinez RCR, de Oliveira AR, and Brandao ML (2007)
Serotonergic mechanisms in the basolateral amygdala differen-
tially regulate the conditioned and unconditioned fear organized
in the periaqueductal gray. Eur Neuropsychopharmacol, doi:
10.1016/j.euroneuro.2007.02.015.
Mathews CA, Kaur N, Stein MB (2008) Childhood trauma and
obsessive-compulsive symptoms. Depress Anxiety 25: 742–751.
8Journal of Psychopharmacology 0(00)

McGaugh JL (2004) The amygdala modulates the consolidation of
memories of emotionally arousing experiences. Ann Rev Neurosci
27: 1–28.
McHaffie JG, Stanford TR, Stein BE, Coizet V, Redgrave P (2005)
Subcortical loops through the basal ganglia. Trends Neurosci 28:
401–407.
Menzies L, Chamberlain SR, Laird AR, Thelen SM, Sahakian BJ,
Bullmore ET (2008) Integrating evidence from neuroimaging and
neuropsychological studies of obsessive-compulsive disorder: the
orbitofrontal-striatal model revisited. Neurosci Biobehav Rev 32:
525–549.
Mizoguchi K, Shoji H, Ikeda R, Tanaka Y, Tabira T (2008)
Persistent depressive state after chronic stress in rats is accompa-
nied by HPA axis dysregulation and reduced prefrontal dopami-
nergic neurotransmission. Pharmacol Biochem Behav 91: 170–175.
Moraes CLK, Bertoglio LJ, Carobrez AP (2008) Interplay between
glutamate and serotonin within the dorsal periaqueductal gray
modulates anxiety-related behaviour of rats exposed to the ele-
vated plus-maze. Behav Brain Res 194: 181–186.
Moskowitz AK (2004) ‘‘Scared stiff’’: catatonia as an
evolutionary-based fear response. Psychol Rev 111: 984–1002.
Nagai Y, Critchley HD, Featherstone E, Trimble MR, Dolan RJ
(2004) Activity in ventromedial prefrontal cortex covaries with
sympathetic skin conductance level: a physiological account of a
‘‘default mode’’ of brain function. NeuroImage 22: 243–251.
Ogata SN, Silk KR, Goodrich S, Lohr NE, Westen D, Hill EM
(1990) Childhood sexual and physical abuse in adult patients
with borderline personality disorder. Am J Psychiatry 147:
1008–1013.
Ogden P, Minton K, Pain C (2006) Trauma and the body: a sensor-
imotor approach to psychotherapy. New York: Norton.
Ongur D, Price JL (2000) The organization of networks within the
orbital and medial prefrontal cortex of rats, monkeys and
humans. Cereb Cortex 10: 206–219.
Perry B and Szalavitz M (2007) The boy who was raised as a dog: and
other stories from a child psychiatrist’s notebook – what trauma-
tized children can teach us about life, loss and healing. New York:
Basic Books.
Perry BD, Pollard RA, Baker WI, Sturges C, Vigilante D, and
Blakley TL (1995) Continuous heartrate monitoring in maltreated
children. In: Annual meeting of the American Academy of Child
and Adolescent Psychiatry, New Research (abstract).
Porges S (1995) Orienting in a defensive world: mammalian modifi-
cations of our evolutionary heritage. A polyvagal theory.
Psychophysiology 32: 301–318.
Porges S (2003) The polyvagal theory: phylogenetic contributions to
social behaviour. Physiol Behav 79: 503–513.
Rainville P, Bechara A, Naqvi N, Damasio AR (2006) Basic emo-
tions are associated with distinct patterns of cardiorespiratory
activity. Int J Psychophysiol 61: 5–18.
Reimer AE, Oliveira AR, Brandao ML (2008) Selective involvement
of GABAergic mechanisms of the dorsal periaqueductal gray and
inferior colliculus on the memory of the contextual fear as
assessed by the fear potentiated startle test. Brain Res Bull 76:
545–550.
Reynolds SM, Berridge KC (2008) Emotional environments retune
the valence of appetitive versus fearful functions in nucleus
accumbens. Nature Neurosci 11: 423–425.
Sailer U, Robinson S, Fischmeister FPS, et al. (2008) Altered reward
processing in the nucleus accumbens and mesial prefrontal cortex
of patients with posttraumatic stress disorder. Neuropsychologia
46: 2836–2844.
Sanchez-Gonzalez MA, Garcia-Cabezas MA, Rico B, Cavada C
(2005) The primate thalamus is a key target for brain dopamine.
J Neurosci 25: 6076–6083.
Schore AN (1994) Affect regulation and the origin of the self.
New Jersey: Lawrence Erlbaum Associates.
Schore AN (2003) Affect Dysregulation and Disorders of the Self.
New York: WW Norton.
Siegel DJ (1999) The Developing Mind. New York: Guilford.
Terzian ALB, Aguiar DC, Guimaraes FS, and Moreira FA (2008)
Modulation of anxiety-like behaviour by Transient Receptor
Potential Vanilloid Type 1 (TRPV1) channels located in the dor-
solateral periaqueductal gray. Eur Neuropsychopharmacol, doi:
10.1016/j.euroneuro.2008.11.004.
Tidey JW, Miczek KA (1996) Social defeat stress selectively alters
mesocorticolimbic dopamine release: an in vivo microdialysis
study. Brain Res 721: 140–149.
Van der Kolk BA (1996a) The complexity of adaptation to trauma:
self-regulation, stimulus discrimination, and characterological
development. Traumatic Stress: the Effects of Overwhelming
Experience on Mind, Body, and Society. New York: Guilford
Press, 182–213.
Van der Kolk BA (1996b) The body keeps the score: approaches to
the psychobiology of posttraumatic stress disorder. In: Traumatic
stress: the Effects of Overwhelming Experience on Mind, Body, and
Society. New York: Guilford Press, 214–241.
Watt DF (2000) The centrencephalon and thalamocortical integra-
tion: neglected contributions of periaqueductal gray. Conscious
Emotion 1: 91–114.
Wong SW, Masse N, Kimmerly DS, Menon RS, Shoemaker JK
(2007) Ventral medial prefrontal cortex and cardiovagal control
in conscious humans. NeuroImage 35: 698–708.
Zanarini MC, Gunderson JG, Marino MF, Schwartz EO,
Frankenburg FR (1989) Childhood experiences of borderline
patients. Compr Psychiatry 30: 18–25.
Zanoveli JM, Ferreira-Netto C, Brandao ML (2007) Conditioned
place aversion organized in the dorsal periaqueductal gray
recruits the laterodorsal nucleus of the thalamus and the basolat-
eral amygdala. Exp Neurol 208: 127–136.
Corrigan et al. 9