Deconstructing acrophobia: Physiological and psychological precursors to developing a fear of heights

Abstract

Background:
Acrophobia is one of the most prevalent phobias, affecting as many as 1 in 20 individuals. Of course, heights often evoke fear in the general population too, and this suggests that acrophobia might actually represent the hypersensitive manifestation of an everyday, rational fear. In this study, we assessed the role of sensory and cognitive variables in acrophobia.

Methods:
Forty-five participants (Mean age 25.07 years, 71% female) were assessed using a booklet with self-reports as well as several behavioral measures. The data analysis consisted in multivariate linear regression using fear of heights as the outcome variable.

Results:
The regression analyses found that visual field dependence (measured with the rod and frame test), postural control (measured with the Sharpened Romberg Test), space and motion discomfort (measured with the Situational Characteristics Questionnaire), and bodily symptoms (measured with the Bodily Sensation Questionnaire) all serve as strong predictors for fear of heights (Adjusted r(2)=.697, P<.0001). Trait anxiety (measured with the State Trait Anxiety Inventory Form Y-2) was not related with fear of heights, suggesting that this higher order vulnerability factor is not necessary for explaining this particular specific phobia in a large number of individuals.

Conclusion:
The findings reveal that fear of heights is an expression of a largely sensory phenomena, which can produce strong feelings of discomfort and fear in the otherwise calm individuals. We propose a theory that embraces all these factors and provides new insight into the aetiology and treatment of this prevalent and debilitating fear.

Full text

Research Article
DEPRESSION AND ANXIETY 27 : 864– 870 (2010)
DECONSTRUCTING ACROPHOBIA: PHYSIOLOGICAL AND
PSYCHOLOGICAL PRECURSORS TO DEVELOPING A FEAR
OF HEIGHTS
Carlos M. Coelho, Ph.D.
1

and Guy Wallis, Ph.D.
1,2
Background: Acrophobia is one of the most prevalent phobias, affecting as many
as 1 in 20 individuals. Of course, heights often evoke fear in the general population
too, and this suggests that acrophobia might actually represent the hypersensitive
manifestation of an everyday, rational fear. In this study, we assessed the role of
sensory and cognitive variables in Acrophobia. Methods: Forty-five participants
(Mean age 25.07 years, 71% female) were assessed using a booklet with self-
reports as well as several behavioral measures. The data analysis consisted in
multivariate linear regression using fear of heights as the outcome variable.
Results: The regression analyses found that visual field dependence (measured
with the rod and frame test), postural control (measured with the Sharpened
Romberg Test), space and motion discomfort (measured with the Situational
Characteristics Questionnaire), and bodily symptoms (measured with the Bodily
Sensation Questionnaire) all serve as strong predictors for fear of heights
(Adjusted r
2
5.697, Po.0001). Trait anxiety (measured with the State Trait
Anxiety Inventory Form Y-2) was not related with fear of heights, suggesting that
this higher order vulnerability factor is not necessary for explaining this particular
specific phobia in a large number of individuals. Conclusion: The findings reveal
that fear of heights is an expression of a largely sensory phenomena, which can
produce strong feelings of discomfort and fear in the otherwise calm individuals.
We propose a theory that embraces all these factors and provides new insight into
the aetiology and treatment of this prevalent and debilitating fear. Depression and
Anxiety 27:864–870, 2010.
rr
2010 Wiley-Liss, Inc.
Key words: acrophobia; visual field dependence; posture; space and motion
discomfort; bodily symptoms; anxiety
INTRODUCTION
Acrophobia is a specific phobia characterized by an
extreme fear of heights, affecting as many as 1 in 20
adults.
[1]
Acrophobic behavior involves the avoidance of
various height-related circumstances, including stairs,
terraces, apartments, and offices located in high build-
ings, and sometimes bridges and elevators. Given the
striking breadth of aversive situations and stimuli
avoided, it is not surprising that individuals with
acrophobia feel extremely impaired and restricted in
their movements.
[2]
Moreover, due to the pervasive and
chronic avoidance of a wide range of situations that form
part of everyday living, this disorder carries an increas-
ing social impact. Current taxonomic models postulate
the existence of hierarchical dimensions falling under Published online 8 April 2010 in Wiley Online Library (wileyonline
library.com).
DOI 10.1002/da.20698
Received for publication 13 November 2009; Revised 27 February
2010; Accepted 7 March 2010
The authors disclose the following financial relationships within
the past 3 years: Contract grant sponsor: Portuguese Foundation
for Science and Technology; Contract grant number: SFRH/BPD/
26922/2006; Contract grant sponsor: Human Frontier Program;
Contract grant number: RGP 03/2006.

Correspondence to: Carlos M. Coelho, Perception and Motor
Systems Laboratory, Connell Building, University of Queensland,
QLD 4072, Australia. E-mail: ccoelho@hms.uq.edu.au
1
Perception and Motor Control Laboratory, University of
Queensland, Brisbane, Queensland, Australia
2
Queensland Brain Institute, University of Queensland,
Brisbane, Queensland, Australia
rr
2010 Wiley-Liss, Inc.

the umbrella of the anxiety disorders.
[3]
Successive
models have labeled these general traits differently, but
as Zinbarg and Barlow
[4]
point out, the conceptual and
empirical overlap remains strong: all these models imply
the overarching involvement of trait anxiety based
largely on the comorbidity of fear subtypes.
Earlier research indicates that anxiety disorders
associated with more vague and/or unpredictable
triggers (e.g., panic, general anxiety disorder) correlate
more with trait anxiety than do more circumscribed
disorders, such as specific phobias,
[5]
but that neuroti-
cism represents a general vulnerability factor for all
forms of anxiety.
[6]
Despite commonalities between
these disorders, the heterogeneity of specific phobias is
formally recognized in the DSM-IV,
[7]
which identifies
four subtypes: animal, natural environment (e.g.,
heights and water), situational (e.g., airplanes and
elevators), and blood-injection-injury.
In the last few years, studies have begun to reveal a
link between postural control and anxiety disorders.
[8]
Unfortunately, the tendency to regard these disorders
as related has lead to patients with differing specific
disorders being lumped together, making it difficult to
ascertain the selective effects of factors such as balance
control. There are, in fact, several good reasons for
thinking that acrophobics may struggle with postural
control. One concerns the related phenomenon of
heights vertigo: a warning signal created by loss of
postural control when the distance between the
observer and visible stationary objects becomes too
large.
[9]
The explanation for this everyday loss of
postural control is that the increased distance between
the observer and the nearest available visual targets
reduces motion parallax cues.
[10]
The lack of these
important depth cues leads to a perceptual conflict, as
the vestibular and somatosensory receptors sense a
body shift not detected by the visual system.
[11]
This, in
turn, has an intense influence on postural responses.
[12]
The resolution of this conflict is achieved by increasing
postural sway, thereby reactivating visual control.
[13]
Despite the parallels to acrophobia, authors in the
heights vertigo literature are at pains to dissociate the
two.
[9]
Nonetheless, it seems possible that acrophobia
could be regarded as an extreme response to this
normal warning signal.
Expanding the notion of a role for postural control in
acrophobia, one possibility is that some individuals are
more reliant on visual cues to control posture than
others. This would make them particularly vulnerable to
heights vertigo.
[14]
In fact, it has been known for some
time that some individuals rely heavily on vision to
regulate their posture and balance.
[15]
Recent findings do
suggest that individuals with acrophobia have poor
postural control, especially in the absence of strong
visual cues.
[16]
What we also know is that persons with
acrophobia have high levels of space and motion
discomfort (SMD), a construct related to physical
symptoms elicited by inadequate visual or kinesthetic
information for normal spatial orientation.
[17]
What is
more, there is evidence that acrophobics report a
heightened sensitivity to physical symptoms, such as
dizziness, feeling short of breath, or heart palpitations.
[18]
The unique contribution of this article is to explore
the role of the abovementioned factors,
[19]
which have
been studied before in disconnected articles concerning
heights and fear of heights. Our hypothesis is that
visual field dependence, postural stability, space and
motion discomfort, and bodily symptoms combine to
elicit the fear and that they will, therefore, predict a
person’s level of acrophobia independently of any
underlying trait anxiety.
MATERIALS/METHODS
PARTICIPANTS
Forty-five participants, recruited from the university population,
took part in the study (Mean age 25.07 years, 71% female) and
received a nominal fee for their participation. Participants with past
or current history of psychiatric disorder were excluded, as were
participants with a known history of vestibular and balance deficits or
other reported neurological or orthopedic disorders that affect
postural control. The 45 selected participants had normal or
corrected to normal vision. All participants were informed of the
testing protocol and provided full informed written consent for
research participation. At the time of initial testing, subjects were
naı¨ve to the experimental hypothesis. The study was approved by the
Committee of Ethics of the University of Queensland.
RATING SCALES
The four questionnaires used in the experiment were:
(1) The Acrophobia Questionnaire
[20]
describes 20 situations frequently
mentioned by acrophobics as anxiety provoking (e.g., standing
next to an open window on an elevated floor). This instrument
shows good internal consistency and test–retest reliabilities,
[20]
and served as a complement for fear of heights, also measured
with a Behavioral Avoidance Test (BAT) explained after the rating
scales section of this paper.
(2) The Bodily Sensation Questionnaire (BSQ)
[21]
contains 17 items
which list body sensations that are usually reported as being
associated with fear and that clients report to be disturbing.
Example items include heart palpitations, dizziness, and feelings of
short breath. Participants report how frequently they experienced
these symptoms on a 5-point Likert scale. The BSQ has very good
internal consistency, stability, and concurrent validity.
[22]
(3) The Situational Characteristics Questionnaire (SitQ)
[17]
was de-
signed to measure SMD. The SitQ is recommended for
quantifying SMD in patients with anxiety and/or balance
disorders. The scores evaluate situations characterized by: (a)
long visual distances (e.g., fields, wide streets, courtyards); (b)
confusing or complex visual stimuli (e.g., fluorescent lighting or
moving stripes); (c) combinations of unusual or complex visual,
proprioceptive and vestibular cues (e.g., widescreen movies,
walking down supermarket aisles); (d) excessive vestibular
stimulation (e.g., dancing); and (e) movements involving neck
extension and reorientation with respect to gravity (e.g., closing
eyes in shower, looking up at tall buildings).
[23]
(4) The State Trait Anxiety Inventory Form Y-2
[24]
is a 40-item
questionnaire (20 for state and 20 for trait anxiety). For the
purpose of this study, we used the trait anxiety scale listing several
865Research Article: Deconstructing Acrophobia
Depression and Anxiety

statements about the general state of being and levels of trait
anxiety, including items such as ‘‘I am happy’’ (item 30) or ‘‘I feel
inadequate’’ (item 35). Participants report how these sentences
are appropriate in describing them on a four-point Likert scale
(1 5not at all to 4 5very much). The total score varies from a
minimum of 20 to a maximum of 80 points.
BEHAVIORAL MEASURES
(1) The BAT is frequently used as a measure of fear of heights.
[25]
In our study, participants were requested to ascend a 15 m high
external stairway which had 70 steps, divided into six landings
(including the ground floor), each separated by 14 steps (Fig. 1).
Participants were invited to climb seven steps at a time, spending
about 15 sec on each landing during which they explored their
surroundings. After this brief exploration period, they evaluated their
level of anxiety (Units of Subjective Distress—USD)
[26]
on a scale
from 1 to 10. While ascending the first set of steps, participants were
accompanied by the researcher in order to explain the exercise.
Participants were explicitly instructed not to hold onto the railing
during testing and were required to look down over the railing at
each landing before filling out their SUDS score. The test finished on
attaining the fifth floor or earlier if the participant reported being
unable to ascend further.
(2) The Rod and Frame Test
[27]
provides a measure of error in the
perception of verticality in degrees. It measures a participant’s ability
to align a rod to vertical within a titled frame.
[28]
The Rod and Frame
visual scene was rendered by a Silicon Graphics Onyx 3200 (SGI Inc.,
Fremont, CA) equipped with an InfiniteReality III graphics. The
scene was projected onto a wall using a BARCO 808S (Kortrijk,
Belgium) analogue projector. The projected image was 2.7 m high
and 3.2 m wide (300 mm above the floor), placed at approximately
2 m from the participant, subtended approximately 78 681of arc.
The image frame rate was 72 Hz and the update rate of the
projection, 24 Hz. Image resolution was set at 1,280 1,024 pixels.
To remove extraneous visual cues to verticality, the test was
conducted in the dark. Participants performed 20 repetitions of the
task. We followed the procedures outlined by Isableu and co-
workers,
[15]
tilting both rod and frame at 181where the effect has
been found to be maximal (Fig. 2). When the outer frame is tilted,
errors in the alignment of the rod result and these errors can be
measured as deviation from true vertical. The resulting error
produces a figure indicative of a participant’s visual field dependence.
(3) The Sharpened Romberg Test
[29]
is a postural task that assesses
balance. It is commonly used in Diving Medicine.
[30]
This test adds
sensitivity to the classical Romberg test
[31]
in routine clinical settings.
Participants are instructed to stand quietly with arms crossed in front
of their chest for a single trial of 2 min, or up to six trials if none
accomplished the summed time of 120 sec before making a visible
step to the side. The final measurement is either 120sec or the
average of the six trials, if none reached 120 sec. Participants stood
barefoot with eyes closed and arms crossed, with the tip of one foot
close to the heel of the other (tandem position), such that both feet
occupied a lateral area not greater than the width of one foot. The
participant was required to use the same foot position for each
consecutive condition. Although the Romberg test is now fairly
standardized, there is some variability in examination technique and
interpretation across expert neurologic examiners.
[32]
In particular,
there is variability in how much postural instability is required for a
positive test (e.g., increased sway only, a step to the side, or a fall);
whether sway is critical at the ankles or the hips; whether footwear
should be worn or removed; whether hands should be held at the side
or extended forward or laterally; etc.
[32]
For the purpose of our study,
we chose to use the most sensitive version of the test (barefoot with
arms crossed) because participants reported no obvious dysfunction
of the proprioceptive, cerebellar, or vestibular pathways.
Figure 1. The behavioral avoidance test (BAT) used in these
studies to behaviorally assess fear of heights. The figures on the
left-hand side indicate height in meters.
Figure 2. Participants were invited to adjust the orientation of
the rod (inside the frame), such that the rod seemed to align with
vertical. The visual field dependence offset error is depicted in
the figure as a
E
.
866 Coelho and Wallis
Depression and Anxiety

ANALYSIS
Statistical analyses were conducted through Student’s t-test for
comparison of two normally distributed independent samples and
multivariable linear regression. Subjective discomfort for heights, as
recorded by the participants, was established as the outcome variable,
whereas predictor variables were visual field dependence, Romberg
test (nonvisual postural control), and the questionnaires relating to
situations evoking disorientation, body symptoms, and trait anxiety.
Earlier to the multivariate model, correlations between the outcome
(USD) and predictive variables were observed using Pearson’s. The
accuracy of the model was assessed through outliers and residuals
diagnoses and generalization, as well as the goodness of fit.
RESULTS
Table 1 summarizes the results of the comparison
between participants low and high in discomfort for
heights on the BAT. There were significant differences
between groups on all variables. This analysis was
repeated using trait anxiety to divide participants (STAI
cut-off point o40). Participants with high trait anxiety
also scored more highly in disorientation and body
symptoms, but there were no significant differences
between groups in balance (w
2
(1) 51.5, P4.05) and
visual field dependence (t(43) 5.09, P4.05).
Correlations analysis, as displayed in Table 2, in-
dicated that the designated outcome variable, USD,
correlated moderately with SitQ (r5.70, Po.0001)
and BSQ (r5.62, Po.0001), and low with STAI-Tr
(r5.33, Po.05). USD was measured from 0 to 10 at
the highest level of the staircase climbed, with a
proportional adjustment for individuals unable to scale
the entire staircase (two additional points for each set
of stairs).
To assess the predictive power of the various factors,
a linear regression analysis was performed in which
three separate hierarchical models were generated.
Visual field dependence and posture were entered in
Model 1, disorientation and body symptoms in
Model 2, and trait anxiety in Model 3. Visual field
dependence, posture, disorientation, and body symp-
toms were all unique predictors of subjective discom-
fort. Trait anxiety not only failed to contribute
significantly to the regression but actually decreased
the predictive power of the model. Table 3 displays the
standardized regression coefficients and adjusted r
2
for
each model. Model 1 accounted for 43% of the
variance in subjective discomfort in heights
TABLE 1. Discomfort in heights in participants who
scored low and high (N545)
Low (n522) High (n523) P
VFD (vision) 4.78 (2.85) 7.04 (2.90) .012

Romberg (balance) 23%
a
65%
a
.005
b
SitQ 23 (19) 54 (28) .000

BSQ 31 (11) 45 (10) .000

STAI-Tr 36 (12) 44 (11) .029

STAI-Tr, State Trait Anxiety Inventory; SitQ, Situational Character-
istics Questionnaire; BSQ, Body Symptoms Questionnaire; VFD,
Visual Field Dependence. Significance levels

Po.05,

Po.01, two-
tailed independent samples t-test.
a
Percentage of participants who were unstable.
b
w
2
test; USD cut off point r5.
TABLE 2. Pearson’s correlations between subscales
VFD BSQ SitQ USD
STAI .04 .26 .47

.33

VFD .12 .25 .54

BSQ .53

.70

SitQ .62

STAI, State Trait Anxiety Inventory; SitQ, Situational Characteristics
Questionnaire; BSQ, Body Symptoms Questionnaire; VFD,
Visual Field Dependence. Significance levels

Po.05,

Po.01,

Po.0001, two-tailed correlations.
TABLE 3. Hierarchical regression analysis for subjective discomfort in heights (N545)
Variables bSig Adjusted R
2
R
2
change Fdf P
Model 1 .432 .457 17.702 2, 42 o.0001
VFD .338 .005
Romberg .540 .000
Model 2 .697 .267 26.272 4, 40 o.0001
VFD .311 .001
Romberg .339 .001
SitQ .299 .006
BSQ .329 .004
Model 3 .689 .000 20.503 5, 39 o.0001
VFD .312 .001
Romberg .339 .001
SitQ .292 .019
BSQ .329 .005
STAI TR .013 .902
STAI, State Trait Anxiety Inventory; SitQ, Disorientation Sensitivity Questionnaire; BSQ, Body Symptoms Questionnaire; VFD, Visual Field
Dependence.
867Research Article: Deconstructing Acrophobia
Depression and Anxiety

(Po.0001), Model 2 accounted for 69.7% of the
variance, and Model 3 explained 68.9%. Trait anxiety
failed to attain statistical significance. The inspection
of the plot for the regressed standardized residuals
indicated the absence of outliers, which was also
observed by the analysis of standardized residuals and
Mahalanobis distances. Generalization assumptions of
multicollinearity, homoscedasticity, normality of er-
rors’ distribution, and independence of errors were met
and goodness of fit was ascertained (F(5, 39) 520.987,
Po.0001).
In summary, the regression analyses found visual
field dependence, postural control, space and motion
discomfort, and bodily symptoms all serve as strong
predictors for fear of heights (Adjusted r
2
5.697,
Po.0001). Trait anxiety was not related with fear of
heights, suggesting that this higher order vulnerability
factor is not necessary for explaining this particular
specific phobia in a large number of individuals.
DISCUSSION
This study has tested a range of factors thought to be
related to fear of heights. Our aim has been to generate
a unified model of how these factors interrelate and
contribute to the disorder. On the basis of our results
and analyses, we propose that the likelihood of
developing a fear of heights is greatest in visual field-
dependent participants with poor nonvisual postural
control and high sensitivity to body symptoms. Under
a series of subsections, we discuss how each of these
factors contributes to the ultimate response.
VISUAL FIELD DEPENDENCE
Participants with acrophobia showed higher visual
field dependence errors. More specifically, they were
given to occasionally producing large errors, signifi-
cantly higher than the largest errors of non-fearful
participants. This result points to an overreliance on
visual information in acrophobics. This, in turn, offers
the first compelling evidence that this reliance is
independent of the phobia itself, because the rod and
frame stimuli do not evoke any sense of depth or
height, unlike the flow stimuli used in an earlier
study.
[33]
The natural lack of motion cues at heights
causes the visual input to differ from that gained
though natural postural sway at ground level. This, in
turn, triggers postural vertigo.
[9,34]
POSTURAL INSTABILITY
At heights, reduced visual feedback creates postural
destabilization. Individuals attempt to increase postural
sway to reactivate visual balance feedback,
[13]
but their
poor postural control
[16]
causes further instability.
Recent studies
[14,35]
showed that fear of heights is
probably not due to a vestibular malfunctioning.
Nonetheless, vestibular sensations or ‘‘symptoms’’ can
arise with discordant or incongruent information
among the sensory channels. In these circumstances,
vestibular sensations are experienced even by indivi-
duals with normal vestibular function.
[23]
Nonetheless, the relation between self-motion and
fear of heights is still a matter for further study, for it is
also possible that when acrophobics freeze at heights,
they reduce their motion parallax cues to depth,
leading to an overestimation of the distance, thereby
increasing fear.
[36–38]
SPACE AND MOTION DISCOMFORT
Participants with acrophobia also have higher scores
of SMD. This implies that people with acrophobia are
sensitive to inadequate visual or kinesthetic informa-
tion for normal spatial orientation.
[17]
SMD seems to
arise with discordant or incongruent information
among visual, kinesthetic, and vestibular sensory
channels.
[39]
Our results also support the claim that
patients with anxiety and SMD rely more heavily on
visual cues for postural control.
[8]
Jacob and co-
workers
[14]
also found that disorientation sensitivity
was related to postural instability, and it is likely that
postural instability precedes body symptoms, in a
process similar to that seen in motion sickness.
[40]
BODY SYMPTOMS
Postural control relies on multisensory processing
and motor responses that seem to be automatic and
occur without conscious awareness.
[12]
An individual’s
interpretation of their own body symptoms may serve
to differentiate between a sensation of arousal or fear.
Like Davey and co-workers,
[18]
we found a high
correlation between measures of acrophobia and a
tendency to interpret internal body sensations of
anxiety as threatening. These findings led Davey and
co-workers
[18]
to suggest that cognitive biases increas-
ingly lead the individual to interpret bodily sensations
as threatening, and that this may be a mechanism
through which chronic acrophobia is acquired (Fig. 3).
SUMMARY
There are currently two major approaches to the
conceptualization of phobias. The first, built around
cognitive models, assumes that individuals with phobia
learn to regard feelings of anxiety as evidence of actual
danger.
[41]
One criticism of such models is that they
can seem somewhat circular: ‘‘I’m anxious because I
fear, and I fear because I’m anxious.’’ The alternative
approach holds that phobias are not learnt, but are
manifestations of general aversive responses to evolu-
tionarily prepared threat stimuli. This approach has
been shown to have some explanatory power, but
likewise has been criticised for falling into a circular
argument when explaining the relative prevalence of
specific phobia subtypes: ‘‘The fear is common because
it is prepared, and is prepared because it is common.’’
The main problem with either approach is that it is
868 Coelho and Wallis
Depression and Anxiety

largely hung up on the question of what is learnt and
what inherited, leaving the precise causes of phobia
unknown. As LeDoux points out, ‘‘Nature and nurture
are partners in our emotional life. The trick is to figure
out what their unique contributions are.’’
[42]
Our
results provide an answer to this question by specifying
the anatomy of the fear as a combination of factors tied
to a specific situation, thereby clarifying the relative
contributions of nature and nurture. What our model
also does is help clarify where acrophobia fits into the
larger scheme of specific phobias and emotions in
general.
[19,43]
Future research, using these parameters,
could be improved by the assessment of sway and
physiological measures during BAT,
[44]
particularly
through the use of virtual reality.
On the basis of the results described in this paper we
have revealed the specific sensory and cognitive biases
that contribute to the development of the fear of
heights, and in so doing have helped identify a need to
reassess its link to the other anxiety disorders. It is hoped
that our model will form the starting point for new and
effective clinical interventions. To that end, work in our
lab is now focused on treatments using postural training
and desensitization to bodily symptoms.
Acknowledgments. The authors are grateful for
the technical support and assistance provided by Welber
Marinovic and Daniela Gonc-alves as well as the
valuable feedback provided by Ottmar Lipp, Aymar de
Rugy, Campbell Reid, and Steven Cloete. The work
was supported by a grant from the Portuguese
Foundation for Science and Technology to Carlos
Coelho (SFRH/BPD/26922/2006), and by a Human
Frontier Program Grant (RGP 03/2006) to Guy Wallis.
REFERENCES
1. Depla M, ten Have ML, van Balkom A, de Graaf R. Specific fears
and phobias in the general population. Results from the
Netherlands Mental Health Survey and Incidence Study (NEM-
ESIS). Soc Psychiatry Psychiatr Epidemiol 2008;43:200–208.
2. Menzies RG. Height Phobia. In: Davey GLC, editor. Phobias.
A Handbook of Theory, Research and Treatment. Chichester:
Wiley; 1997:129–138.
3. Watson D. Dimensions underlying the anxiety disorders: a
hierarchical perspective. Curr Opin Psychiatry 1999;12:181–186.
4. Zinbarg RE, Barlow DH. The structure of anxiety and the
anxiety disorders: a hierarchical model. J Abnorm Psychol 1996;
105:181–193.
5. Mineka S, Watson D, Clark LA. Comorbidity of anxiety and
unipolar mood disorders. Annu Rev Psychol 1998;49:377–412.
6. Longley SL, Watson D, Noyes R, Yoder K. Panic and phobic
anxiety: associations among neuroticism, physiological hyperar-
ousal, anxiety sensitivity, and three phobias. J Anxiety Disord
2006;20:718–739.
7. APA. Diagnostic and Statistical Manual of Mental Disorders. 4th
ed. Washington, DC: American Psychiatric Association; 2000.
8. Redfern MS, Furman JM, Jacob RG. Visually induced postural
sway in anxiety disorders. J Anxiety Disord 2007;21:704–716.
9. Brandt T, Arnold F, Bles W, Kapteyn TS. The mechanism of
physiological height vertigo. I. Theoretical approach and
psychophysics. Acta Otolaryngol 1980;89:513–523.
10. Guerraz M, Guerraz M, Sakellari V, Burchill P, Bronstein AM.
Influence of motion parallax in the control of spontaneous body
sway. Exp Brain Res 2000;131:244–252.
11. Nadler JW, Angelaki DE, DeAngelis GC. A neural representa-
tion of depth from motion parallax in macaque visual cortex.
Nature 2008;452:642–710.
12. Bronstein AM, Buckwell D. Automatic control of postural sway
by visual motion parallax. Exp Brain Res 1997;113:243–248.
13. Gatev P, Thomas S, Kepple T, Hallett M. Feedforward ankle
strategy of balance during quiet stance in adults. J Physiol (Lond)
1999;514:915–928.
14. Jacob RG, Redfern MS, Furman JM. Space and motion
discomfort and abnormal balance control in patients with anxiety
disorders. J Neurol Neurosurg Psychiatry 2009;80:74–78.
15. Isableu B, Ohlmann T, Cremieux J, Amblard B. How dynamic
visual field dependence-independence interacts with the visual
contribution to postural control. Hum Mov Sci 1998;17:
367–391.
16. Boffino CC, de Sa CSC, Gorenstein C, et al. Fear of heights:
cognitive performance and postural control. Eur Arch Psy
Clin N 2009;259:114–119.
17. Jacob RG, Woody SR, Clark DB, Lilienfeld SO. Discomfort
with space and motion: a possible marker of vestibular
dysfunction assessed by the situational characteristics question-
naire. J Psychopathol Behav Assess 1993;15:299–324.
18. Davey GCL, Menzies R, Gallardo B. Height phobia and biases in
the interpretation of bodily sensations: some links between
acrophobia and agoraphobia. Behav Res Ther 1997;35:997–1001.
19. Coelho CM, Waters AM, Hine T, Wallis G. The use of virtual
reality in acrophobia research and treatment. J Anxiety Disord
2009;23:563–574.
Figure 3. When exposed to heights (a) the lack of visual cues (b)
creates a natural increase in postural sway (c) aimed at
reactivating visual control. In visual field-dependent individuals
(d) with poor nonvisual postural control, the lack of visual cues
will result in postural instability (e) and a feeling of being off-
balance resulting in a moderate level of fear. For those
individuals who tend to feel disoriented and sick through such
instability (f) more fear will result. However, some individuals
may experience a thrill or excitement through disorientation
rather than a debilitating fear. It is only those who experience
discomfort in the presence of dizziness and disorientation (g)
who will go on to experience a strong fear response.
869Research Article: Deconstructing Acrophobia
Depression and Anxiety

20. Baker BL, Cohen DC, Saunders JT. Self-directed desensitization
for acrophobia. Behav Res Ther 1973;11:79–89.
21. Chambless DL, Caputo GC, Bright P, et al. Assessment of fear of
fear in agoraphobics—the body sensations questionnaire and the
agoraphobic cognitions questionnaire. J Consult Clin Psychol
1984;52:1090–1097.
22. Fischer J, Corcoran K. Measures for Clinical Practice:
A Sourcebook. 2nd ed. New York: Free Press; 1994.
23. Jacob RG, Redfern MS, Furman JM. Optic flow-induced sway in
anxiety disorders associated with space and motion discomfort.
J Anxiety Disord 1995;9:411–425.
24. Spielberger CD. Manual for the State-Trait Anxiety Inventory
(Form Y) (‘‘Self-Evaluation Questionnaire’’). California: Con-
sulting Psychologists Press; 1983.
25. Abelson JL, Curtis GC. Cardiac and neuroendocrine responses to
exposure therapy in height phobics: desynchrony within the
‘‘physiological response system’’. Behav Res Ther 1989;27:561–567.
26. Wolpe J, Lang PJ. A fear survey schedule for use in behavior-
therapy. Behav Res Ther 1964;2:27–30.
27. Whitkin HA, Lewis HB, Hertzman M, et al. Personality
Through Perception. New York: Harper; 1954.
28. Bray A, Subanandan A, Isableu B, et al. We are most aware of our
place in the world when about to fall. Curr Biol 2004;14:
R609–R610.
29. Fregly AR, Graybiel A. An ataxia test battery not requiring rails.
Aerospace Med 1968;39:277–282.
30. Lee C-T. Sharpening the sharpened romberg. SPUMS J
1998;28:125–132.
31. Romberg M. Textbook of Nervous Diseases of Man. Berlin:
Duncker; 1851.
32. Lanska DJ, Goetz CG. Romberg’s sign—development,
adoption, and adaptation in the 19th century. Neurology
2000;55:1201–1206.
33. Huweler R, Kandil FI, Alpers GW, Gerlach AL. The impact of
visual flow stimulation on anxiety, dizziness, and body sway in
individuals with and without fear of heights. Behav Res Ther
2009;47:345–352.
34. Bles W, Kapteyn TS, Brandt T, Arnold F. The mechanism
of physiological height vertigo. II. Posturography. Acta
Oto-Laryngologica 1980;89:534–540.
35. Alpers GW, Adolph D. Exposure to heights in a theme park: fear,
dizziness, and body sway. J Anxiety Disord 2008;22:591–601.
36. Clerkin EM, Cody MW, Stefanucci JK, et al. Imagery and fear
influence height perception. J Anxiety Disord 2009;23:381–386.
37. Stefanucci JK, Proffitt DR. The roles of altitude and fear in the
perception of height. J Exp Psychol Hum Percept Perform
2009;35:424–438.
38. Stefanucci JK, Storbeck J. Don’t look down: emotional arousal
elevates height perception. J Exp Psychol Gen 2009;138:
131–145.
39. Brandt T, Daroff RB. Physical therapy for benign paroxismal
positional vertigo. Arch Otolaryngol Head Neck Surg 1980;
106:484–485.
40. Stoffregen TA, Smart LJ. Postural instability precedes motion
sickness. Brain Res Bull 1998;47:437–448.
41. Arntz A, Rauner M, Van Den Hout M. ‘‘If I feel anxious, there
must be danger’’: ex-consequentia reasoning in inferring danger
in anxiety disorders. Behav Res Ther 1995;33:917–925.
42. LeDoux JE. The Emotional Brain. New York: Simon &
Schuster; 1996.
43. Coelho CM, Purkis H. The origins of specific phobias:
influential theories and current perspectives. Rev Gen Psychol
2009;13:335–348.
44. Simeonov PI, Hsiao HW, Dotson BW, Ammons DE. Height
effects in real and virtual environments. Hum Factors 2005;
47:430–438.
870 Coelho and Wallis
Depression and Anxiety