Content uploaded by Georg H. Eifert
Author content
All content in this area was uploaded by Georg H. Eifert
Content may be subject to copyright.
Evaluating differential predictions of emotional
reactivity during repeated 20% carbon dioxide-
enriched air challenge
Michael J. Zvolensky
Brown University School of Medicine, Providence, RI, USA
Matthew T. Feldner and Georg H. Eifert
West Virginia University, Morgantown, WV, USA
Sherry H. Stewart
Dalhousie University, Canada
The present study explored psychological predictors of response to a series of three
25 second inhalations of 20
%
carbon dioxide-enriched air in 60 nonclinical par-
ticipants. Multiple regression analyses indicated that only anxiety sensitivity
physical concerns predicted self-reported fear, whereas both physical anxiety
sensitivity concerns and behavioural inhibition sensitivity independently predicted
affective ratings of emotional arousal. In contrast, the psychological concerns
anxiety sensitivity dimension predi cted ratings of emotional displeas ure (valence),
and both psychological anxie ty sensitivity concerns and behavioural inhibition
sensitivity independently predicted emotional dyscontrol. No variables sig-
nificantly predicted heart rate. These data are in accord with current models of
emotional reactivity that hi ghlight the role of cognitive variables in the production
of anxious and fearful responding to somatic perturbation, and help further clarify
the particular predictors of anxiety-related responding to biological challenge.
Experimental psychopathology research increasingly has relied on the use of
biological challenge procedures for investigating the parameters of anxious and
fearful responding during interoceptive perturbation. Of the biological chal-
lenges, inhalations of carbon dioxide-enriched air (CO
2
) have been among the
most frequently employed (Rapee, 1995; Sanderson, & Wetzler, 1990;
COGNITION AND EMOTION, 2001, 15 (6), 767–786
Correspondenc e should be addressed to Michael J. Zvolensky, who is now at the University of
Vermont, Department of Psychology, John Dewey Hall, Burlington, VT, 05405-0134 USA (Email:
m_zvolen@dewey.uvm.edu).
This research was support ed, in part, from a research grant from Society of Science for Clinical
Psychology and a scholarship from American Psychological Association awarded to the first author.
#
2001 Psychology Press Ltd
http://www.tandf.co.uk/journals/pp/02699931.html DOI:10.1080/02699930143000284
Zvolensky & Eifert, 2000). Unlike traditional forms of aversive tactile stimu-
lation that produce primarily pain (e.g., cold pressor tasks), CO
2
produces
abrupt, escalating bodily sensations that define, in part, anxiety and fear states
that characterise panic and related di sorders. Using such provocation pro-
cedures, it has been possible to examine the extent to which specific biological,
behavioural, and cognitive process variables predict anxious and fearful
responding when indi viduals experience bodily peturbation (McNally, 1994,
1999).
Anxiety sensitivity and prediction of emotional
reactivity during states of bodily arousal
Research examining the influence of cognitive processes in the prediction of
anxious and fearful responding to panic provocation procedures has foun d that
concerns about the negative consequences of bodily sensations (i.e., ‘‘anxiety
sensitivity’’) predi ct challenge response (McNally & Eke, 1996; Rapee &
Medero, 1994; Rassovsky, Kushner, Schwarze, & Wangensteen, 2000). Anxiety
sensitivity refers to the fear of anxiety-related symptoms that are based on
beliefs that such sensations have negative somatic, social, or psychol ogical
consequences (Taylor, 1995). For example, if a person perceives bodily sensa-
tions that are associated wi th autonom ic arousal as a sign of imminent harm, this
‘‘high anxiety sensitive’’ individual will likely experience el evated levels of
anxiety. Research suggests that the Anxiety Sensitivity Index (ASI; Reiss,
Peterson, Gursky, & McNally, 1986) has three lower order fac tors that all load
on a single higher order factor. The lower order factors represent physical,
psychological, and social concerns, and the higher order factor represents the
global anxiety sensitivity construct (Stewart, Ta ylor, & Baker, 1997; Zinbarg,
Mohlman, & Hong, 1999).
There i s emerging evidence that the greater the degree of specificity between
pre-existing fear about bodily sensations and symptoms experienced, the better
self-reported anxiety-related responding can be predi cted (Eifert, Zvolensky,
Sorrell, Hopko, & Lejuez, 1999; Rachman & Taylor, 1993). For instance,
Schmidt (1999) recently found that cardiopulmonary fears, a subcomponent of
anxiety sensi tivity physical concerns, was the only dimension t hat predicted
reported anxiety and bodily sensations during 35
%
CO
2
inhalation in panic
disorder patients. These findings are consistent with the idea that differen t
dimensions of anxiety sensitivity may be uniquely related to theoretically
similar aspects of negative emotional responding (Eifert, Zvolensky, & Lejuez,
2000). Still, the majority of existing research in this arena has utilised unimethod
approaches i n the a ssessment of anxi ety and fear states (i.e., self-reported
assessment devices), precluding direct examination of the multi dimensional
nature of emotional reactivity. This limitation impedes clarification as to how
specific cognitive process variables affect multiple levels of emotional
768
ZVOLENSKY ET AL.
responding. As issues such as response concordance-discordance across systems
provides important, if not critical, insight into the nature and pa tterning of
anxiety and fear sta tes (see Zinbarg, 1998), a natural step for future research is to
isolate how specific c ognitive processes affe ct emotional responding with a
multimethod assessment approach.
Another important task of research in this arena pertains to examining how
heretofore unexamined theoretically relevant cognitive process variables affect
the nature and patterning of responding to bodily perturbation. One such cog-
nitive variable is behavi oural inhibition sensitivity (BIS), which has direct
relevance to understanding anxiety-related responding t o states of physical
stress.
Behavioural inhibition sensitivity
Central to theoretical accounts of anxiety psychopathology, behavioral inhibi-
tion sensitivity denotes a hypervigiliance to potentially harmful sensory events
(Gray & McNaughton, 1996). Behavioural inhi bition is a multidimensional
construct defined by a sensitivity to signals of punishment, unfamiliarity, and
nonreward (Cloninger, 1987). Although theoretical notions may vary between
researchers, the re is a general consensus that behavioural inhibition is char-
acterised by escape/avoidance, elevated somatic activity, distressed verbal and
affective expression, as well as a general uneasiness in situations construed as
potentially threatening (Davidson, Ekman, Saron, Senulis, & Friesen, 1990;
Kagan, 1989). Following Gray (1987, 1990), it is hypothesised that the
behavioural inhibition system controls the experience of anxiety, and contributes
to the inhibition of behaviour t hat may lead to negative outcomes, pa rticularly in
aversive and novel contexts. The behavioural activation system, on the other
hand, is responsible for positive emotional experiences and primarily sensitive
to signals of reward.
1
At a neuorophysiologica l level of analysis, the BIS is
comprised of the septohippocampal system, its monoa minergic afferents
extending from the brainstem, and neocortical projections in the frontal lobe
(Gray & McNaughton , 1996). Accordingly, t he BIS involves physiological and
neurotransmitter processes and higher cortical functions, including cognitive-
affective reactions to a wide variety of stimuli.
To date, researchers largely have focused scientific efforts on the physiolo-
gical processes involved with the BIS, and t he behavioural manifestations of BIS
in children (Davidson et al., 1990; Harmon-Jones & Allen, 1997; Kagan,
Reznick, & Snidman, 1988; Reznick et al., 1986; Sobotka, Davidson, & Senulis,
1992; Sutt on & Davidson, 1997). Collectively, these findings indicate that
particular neural substrates located in the right frontal areas of the cerebral
1
We do not further discuss the BAS at this juncture because this system is not directly relevant to
the focus of the present paper.
PREDICTION OF ANXIETY AND FEAR
769
cortex are related to the experience of anxie ty based negative affect, and that a
behavioural inhibition re sponse style predicts anxiety-related responding in
future circumstances. Although such evidence attests to the physiological and
behavioural components of t he BIS construct, it also is reasonable to e xpect that
BIS should be apparent at a cognitive level of analysis. That is, cognitive pro-
cesses arise from the same fundame ntal brain structure as do physiological and
behavioural proce sses (Gray & McNaughton, 1996) and therefore should
similarly be relevant to the prediction of negative emotional responding to a
greater or lesser extent. Yet, at this stage of research development, research
addressing cognitive-based aspects of BIS has unfortunately bee n less forth-
coming.
Behavioural inhibition sensitivity and bodily sensations. Carver and White
(1994) have developed the BIS/BAS scale to assess dispositional sensitivities to
Gray’s two ge neral motivational systems. Research indicates that the scale can
detect varying sensitivities in the presumed emotion regulatory systems. For
instance, sc ores on the BIS scale have been found t o predict relative changes in
anxiety that occur in response to a cold pressor task (Carver & White, 1994,
study 3). Other research using the BIS/BAS scales indicates relations betwe en
the BAS-BIS difference score and electrophysiological asymmetr y is restricted
largely to prefrontal scalp regions (Sutton & Davidson, 1997), data consistent
with behavioural inhibition theory (Gray, 1990).
It thus far remains unclear to what extent BIS, at a cognitive level of analysis,
relates to anxious and fearful responding to bodil y sensations. This is unfortu-
nate given that the available, albeit limited, evidence suggests behavioural
inhibition is associated with heightened physiological activity and emotional
distress (Garcia-Coll, Kagan, & Reznick, 1984). Additionally, th e presence of
panic attacks and panic disorder in parents is associated with a greater prob-
ability of BIS in first order offspring (Rosenbaum et al., 1988). It is important to
recognise that these findings also are consistent with the role of interoceptive
attentional processes in the generation and mainte nance of anxiety and fear. For
instance, contemporary models of panic and related disorders emphasise that
individuals expect somatic events to be threatening and that there is an exag-
gerated vigilance on internal sensations (Bouton, Mineka, & Barlow, 2001).
Similar to behavioural inhibition theory, such conceptualisations posit that
heightened anxious and fearful responding occurs in respons e to bodily sensa-
tions perceived as dangerous, contributing to increased attention on somatic
events (Ehlers & Breuer, 1992; M cNally, 1998; Schmidt, Lerew, & T rakowski,
1997; Schmidt & Trakowski, 1999). It is plausible, then, a cognitive-based BIS
might represen t one process variable that may produce elevated anxious
responding to bodily sensations. Rather than promoting anxiety-related
responding solely because bodily peturbation is believed to have negative
consequences (i.e., anxiety sensitivity), a cognitive-based BIS could theoreti-
770
ZVOLENSKY ET AL.
cally promote anxious and fearful responding for a number of reasons (e.g. ,
uncontrollable, unpredictable, unfamiliar , negatively valenced).
The pre sent investigatio n
The present study was designed to evaluate t wo interrelated hypotheses con-
cerning the prediction of anxious and fearful responding to bodily sensations
using repeated administrations of 20
%
CO
2
. The first aim was to evaluate
multiple anxiety sensitivity dimensions’ (i.e., physical, psychological, and social
concerns) ability to predict anxi ety and fear in response to CO
2
challenge
relative to other theoretically relevant variables (e.g., trait anxiety). The second
aim was to provide an initial test as to whether a cognitive-based BIS also would
independently predict particular aspects of anxious and fearful responding to
recurrent interoceptive pe rturbation. We studied CO
2
-challenge naive, non-
clinical persons in order to rule out the possibility that any observed differences
would be due to psychopathology, and at the same time, increase the chance the
provocation would function as a theoretically relevant unfamiliar experience.
We utilised a multimethod assessment protocol to provide a comprehensive
evaluation of emotional responding. Consistent with existing research (Schmidt,
1999), we expected that concerns about the negative consequences of psycho-
logical and physical sensations commonly produced by the CO
2
(i.e., anxiety
sensitivity psychological and physical concerns, respecti vely) would be more
predictive of provocation-induce d anxiet y than other concerns not related to
those symptoms (i.e., anxiety sensitivity social concerns) and trait anxiety. It
also was hypothesised that BIS would be significantly predi ctive of emotional
displeasure, arousal, and dyscontrol as highlighted in various contemporary
perspectives of behavioural inhibition (Gray & M cNaughton, 1996).
METHOD
Participants
A total of 60 part icipants (34 females) were recruited from a larger screening
sampl e at West Virgi nia University. The ethnic distribution was 90
%
(n = 54)
Caucasian, 6.7
%
(n = 4) Afric an-American, and 3.3
%
(n = 2) Hispanic. Parti-
cipants were excluded from participation if they reported having heart di sease,
epilepsy or a seizure disorder, hypertension, lung disorders (e.g., emphysema),
being pregnant, or having previously participated in a CO
2
-challenge study.
Additionally, participants were excluded if they reported a personal history of
psychopathology , including panic attacks and use of psychotropic medication.
The exclusionary criteria were assessed prior to partic ipation with a semi-
structured interview routinely employed in our laboratory (e.g., Zvolensky,
Lejuez, & Eifert, 1998). Overall, the exclusionary criteria served to decrease the
probability of any complications arising from CO
2
inhalation and to rule out
PREDICTION OF ANXIETY AND FEAR
771
related relevant psychopathological states. All eligibility criteria were evaluated
by advanced graduate students in clinical psychology who were trained an d
supervised in the use of the semi-structured CO
2
-screening interview.
In total, 83 persons originally were contacted by phone to assess eligibility
status. Of these individuals, eight were excluded from participation due to their
use of cardiovascular-reactive pharmacological a gents (i.e., decongest ants ,
asthm a medication). Fourteen additional individuals were excluded for reporti ng
having previously experienced a panic attack (n = 9), having been treated for a
psychological dysfunction (n = 2), or both (n = 3). Finally, one person was
excluded due to previous participation in another CO
2
investigation. All part i-
cipants completed a written consent prior to the study. At the conclusion of the
investigation, participants were debriefed and compensated with extra course
credit for their efforts.
Measures
Pre-experimental psychological assessment. The Behavioral Inhibition/
Activation scale (BIS/BAS; Carver & White, 1994) is a 22-item self-report
instrument designed to assess BIS and BAS dispositional sensit ivities.
Participants indicate on a 4-point Likert-type scale (0 = strongly disagree to 4
= strongly agree) the degree to which they agree about a particular statement
(e.g., ‘‘If I think something unpleasant is going to happe n I usually get pretty
worked up’’). The sca le produces three separate scores, one for BIS sensitivity
(7 items), and three BAS subscales (i.e., reward responsiveness, 5 items; drive, 4
items; and fun seeking, 4 items). Scores for each of the subscales are derived by
summing all responses for scale i tems (reverse scoring when appropriate).
Factor analysis indicates the scal e taps the hypothesised psychological
dimensions (Carver & White, study 1). Additionally, the BIS/BAS scale has
demonstrated adequate internal consistency (e.g., BIS scale alpha: .74) and
adequate convergent and discriminant validity (Carver & White, study 2).
The ASI (Reiss et al., 1986) is a 16-item questionnaire in which respondents
indicate on a 5-point Likert-type scale (0 = very little to 4 = very much) the
degree to which they are concerned about the possible negative consequences of
anxiety symptoms. The ASI score is derived by summing all responses, with
total scores ranging from 0 to 64. The ASI has good internal consistency and
test-retest reliability (Peterson & Re iss, 1992). Persons with panic disorder score
significantly higher on the ASI than persons with other types of anxi ety dis-
orders (e.g., generalised anxiety disorder) despite these same individua ls
reporting similar levels of trait anxiety (Taylor, Koch, & McNally, 1992).
Research indicates that factor structure of the ASI is comprised of three lower
order factors (i.e., physical, 9 items, psychological, 5 items; an d social concerns,
2 items) that load on a single higher order (global anxiety sensitivity) factor
(Stein, Jang, & Livesley, 1999; Stewart et al., 1997; Zinbarg et al., 1999;
772
ZVOLENSKY ET AL.
Zvolensky, McNeil, Porter, & Stewart, in press). In the present st udy, we
computed ASI lower-order scores similar to other investigations in this area
(e.g., Stewart et al., 1997).
The Trait Anxiety Inventory Form-Y (STAI; Spielberger, Gorsuch, Lushene,
Vagg, & Jacobs, 1983) is a face valid 20-item scale designed to index trait
(chronic) anxiety. Participants respond on a 4-point Likert-type scale (1 = no
anxiety to 4 = extreme anxiety), with total scores for each subscale ranging fro m
20 to 80. Scores are computed by summin g responses, reverse scoring when
appropriate. The STAI is a reliable and valid measure of trait anxiety that i s
commonly employed in anxiety research with clinical and nonclinical popula-
tions (Spielberger et al., 1983).
Experimental cognitive and affective assessment of response to the CO
2
challenge. To assess fear levels, a Subjective Units of Di stress scale (SUDs;
Wolpe, 1958) was used to index self-reported fear, ranging from zero (no fear)
to eight (extreme fear).
To examine the affective response to the challenge, we employed the Self-
Assessment Manikin (SAM; Lang, 1980). The SAM is a nonverbal pictorial
assessment methodology using a human-like stimulus. According to factor-
analytic research, the SAM permits the rapid assessment of three fundamental
dimensions of affective meaning in emotional experience: Valence (scale 1),
Arousal (scale 2), and Dominance/Control (scale 3; Lang, 1984). SAM ratings
are made along five figure s for each scale (Lang, Bradley, & Cuthbert, 1990).
Participants select their current level of affective state al ong the dimensions of
valence, arousal, and dominance/control on a 9-point scale. Specifically, parti-
cipants can place an ‘‘X’’ over any of the five figures in each scale, or between
any two figures. Tot al scores range from zero to nine for each of the three
dimensions. The SAM has sound psychometric properti es and has previously
been successfully employed to m easure affective responses (Bradley & Lang,
1994), including anxiety studies involving nonclinical and clinical participants
(McNeil, Vrana , Melamed, Cuthbert, & Lang, 1993). Consistent with previous
research (B radley & Lang, 1994), participants were instructed to rate their
personal reaction using t he pictorial dimensions (Lang, 1980).
Experimental physiological assessment. Heart rate has arguably been one of
the most studied indices of peripheral physiological activity in both the
behavioural inhibition and anxiety literatures (e.g., Kagan, 1989). Consistent
with such previous work, we employed a Coulbourn Modular recording system
to assess heart rate responding online at a sample rate of 10 samples per second
(
§
5 volts). Channels were calibrated online prior to sampling. Heart rate was
sampled in beats/per minute (bpm) using a digital Coulbourn tachometer fed
through a S75-01 bioamplifier and assessed via Medi-Trace pregelled Ag/AgCl
electrodes. Heart rate placement followed standard bilateral positioning on
PREDICTION OF ANXIETY AND FEAR
773
either side of the participant’s rib cage. Prior to data analysis, all physiological
data were screened for outliers due to sampling error (e.g., participant
movement).
Apparatus
Experimental sessions were completed in a 2 m
£
6 m sound attenuated room in
the Department of Psychology a t West Virgi nia University. The room containe d
a chair, a desk with a Pentium microcomputer, SVGA colour monitor, mouse,
and keyboard. An intercom allowed participants to communicate freely with the
experimenter in the adjacent room. The experimenter room contained 20
%
CO
2
compressed air enclosed in a 40 cylinder gas tank, a one-way mirror, and a
Coulbourn Modular recording device read through a Pentium microcomputer.
Arousal-inducing stimulus. T he arousal-inducing stimulus was 20
%
CO
2
-
enriched air (20
%
CO
2
, 21
%
O
2
, 59
%
NO
2
). Participants were equipped with a
continuous positive pressure Downs C-Pap Mask (Vital Signs Inc., Model No.
9000). The CO
2
was store d in a 101 cm cylinder and fed through a 5 cm
£
5 cm
hole via aerosol tubing from the experimenter room to a positive-pressure downs
C-pap mask worn by t he participant. An automated apparatus was used for CO
2
delivery (see Lejuez, Forsyth, & Eifert, 1998, for a comprehensive description).
A one-way mirror allowed the experimenters to directly observe all session
events. In total, there were three 25s 20
%
CO
2
inhalations . The intertrial interval
for CO
2
inhalations varied between 120 s and 420 s (M = 300 s). With the
average intertrial interval being 300 s, there was sufficient time for any
emotional effects from a prior inhalation to dissipate (Zvolensky et al., 1998).
Previous research has found that 20
%
CO
2
reliably produces recurrent episodes
of bodily arousal (Forsyth & Eifert, 1998 ; Lejuez, O’Donnell, Wirth, Zvolensky,
& Eifert, 1998; Zvol ensky, Eifert, Lejuez, & McNeil, 1999).
Procedure
Eligible participants were scheduled to come to the psychophysiologica l
laboratory. To prevent the possibility of physiological contamination-related
effects, these participants were asked t o avoid physical exercise for 1 hour prior
to their scheduled appointment (Manuck, Kamarck, Kasprowicz, & Waldstein,
1993). On arrival, participants completed a consent form explaining that the
procedure would involve breathin g 20
%
CO
2
-enriched air. All participants were
informed that t hey had a right to discontinue participation at a ny time during the
investigation without penalty, and that inhalation of the CO
2
-enriched air would
produce sensations associated with bodily arousal. Participants then completed
the BIS/BAS, ASI, and STAI (randomly ordered). Once pre-experimental
measures were completed, participants were seated in the experimental room,
and then heart rate recording electrodes and CO
2
mask were attached. Partici-
774
ZVOLENSKY ET AL.
pants then sat quietly for a 10 min habituation period plus a baseline heart rate
recording interval at the end. Durin g this habituation/heart rate baseline
assessment, the experimenter left the experimental room. After the habi tuation/
heart rate baseline period, participants completed a pre-expe rimental SUDs
rating to index anticipatory anxiety .
During the session, participants sat alone in the experimental room. The
investigato r could view the participant from the one-way mirror and bidirec-
tional communication was possible via the intercom. All participants received
three 25 s inhalation of 20
%
CO
2
-enriched air. Previous research i ndicates that
this concentration and duration of gas administration produces elevated self-
reported anxiety and bodily arousal (Zvolensky & Eifert, 2000). Participants
were not informed about the onset, duration, or offset of CO
2
administration. At
30 s post inhalation, participants complete d the SUDs and SAM. In previous
research, we have found the 30 s post inhalation assessment allows for the
evaluation of anxiety-related responding during a time period when CO
2
-enri-
ched air is having an anxiety-provoking effect (e.g., Zvolensky et al., 1999).
Following completion of the fina l gas administration, participants sat for a 5 min
rest period in order to allow challenge-based arousal to di ssipate. After the final
rest period, the heart rate recording electrodes and CO
2
mask were removed, and
the participant was debriefed.
RESULTS
Validity of 20
%
CO
2
challenge in produci ng anxiety related responding. To
assess the validity of the challenge in producing anxiety-related distress, we first
examined the extent of self-reported and physiological reactivity to each CO
2
administration. Consistent with other researcher s examining emotional reactivity
(e.g., Larkin & Zayfert, 1996), we calculated the mean baseline-to-challenge
difference in SUDs for each CO
2
inhalation. In a similar way, experimental
physiological responsiveness was assessed using change scores because this
index is a sensitive measure of physiologica l reactivity. Specifically,
physiological responsiveness was assessed by subtracting each inhalation period
from the 60 s baseline time period. Exam ining physiological activity in this
manner permitted evaluation of somatic reactivity, while taking into considera-
tion pre-experimental physiological activity. For all analyses, to control for
family-wise erro r rate, the alpha level for these analyses was set at .01 (Cohen,
1988).
In regard to SUDs ratings, paired-samples t-tests revealed statistically sig-
nificant baseline-to-challenge change across inhalations, such that grea ter
anxiety was reported post-relative to pre-challenge. The CO
2
administrations
produced significantly e levated mean SUDs rating s from baseline (M = 2.1, SD
= 1.0) to the CO
2
assessment (M = 4.7, SD = 1.7) across all three trials. The
corresponding mean (change) differences were 2.6 (1.5) for inhalation one, t(59)
PREDICTION OF ANXIETY AND FEAR
775
= 10.5, p
<
.001, 2.4 (1.4), for inhalation two, t(59) = 10.1, p
<
.001, and 2.4
(1.5) for inhalation three, t(59) = 10.0, p
<
.001.
We used the same data-analytic procedure for heart rate. Baseline responding
was assessed using data from the last 60 s of the 10 min baseline resting period,
and challenge heart rate was a ssessed using data for the 25 s during CO
2
inhalation period. The CO
2
administrations produced significantly elevated
mean heart rate responding from baseline (M = 79.5, SD = 7.7 ) to CO
2
inhalation (M = 89.1, SD = 9.2) across all three trials. In particular, the corre-
sponding mean (change) differences were 9.0 (10.9) for inhalation one, t(59) =
11.1, p
<
.001, 11 (11.8) fo r inhalation two, t(59) = 11.1, p
<
.001, and 9.0 (11.8)
for inhalation three, t(59) = 9.8, p
<
.001. Taken together, there was relatively
convincing evidence that the provocation procedure reliably produced elevated
heart rate and self-reported anxiety.
Descriptive data and zero order relations between theoretically relevant
variables. For comparison purposes, we first computed means and standard
deviations for the rel evant predictor variables. As can be seen in Table 1, the
TABLE 1
Descriptive data and zero-order relations between predictor and criterion variables
(
n
= 60)
Mean or n
(SD or
%
)
SUDs SAM
valence
SAM
arousal
SAM
control
Heart
rate
Demographic /Behavioural variables
Gender (female/male) 34 (57
%
) .49** .43** .41** .26* .24*
Age 21.1 (2.3) .08 .05
7
.05 .09 .04
Smoking status (yes/no) 17 (28
%
) .19 .20 .17 .17 .09
Psychologica l variables
BIS 21.9 (3.3) .24* .22** .62** .37** .16
ASI (global score) 22.6 (15.1) .29* .26* .45** .37** .13
ASI (physical concerns) 12.3 (10.6) .28* .23 .42** .29* .11
ASI (psychological concerns) 5.4 (4.3) .26* .31* .39** .49** .11
ASI (social concerns) 5.0 (1.5) .18 .04 .19 .19 .06
STAI-Trait 40.5 (10.8) .19 .19 .21* .16 .08
Pre-experimental SUDs rating 2.1 (1.0) .35** .09 .14 .05 .11
Note: W e collapsed across CO
2
trials because the same pattern of zero order relations was present
across all three gas administrations. Both gender and smoking were dummy coded (1 and 2). The
number for gender reflects females, and the number for smoking status reflects a positive response to
using smoking tobacco. * p
<
.05; ** p
<
.01.
ASI, Anxiety Sensitivity Index (Reiss et al., 1986); BIS/BAS, Behavioral Inhibition and
Activation Scale (Carver & White, 1994); STAI, State-Trait Anxiety Inventory (Spielberger et al.,
1983).
* p
<
.05; ** p
<
.01.
776
ZVOLENSKY ET AL.
scores for the self-report measure s were similar to those found in other
investigations involving nonclinical participants (Carver & White, 1994;
Peterson & Reiss, 1992; Spielberger et al., 1983; Stewart et al., 1997).
To better understand how behavioural inhibition sensitivity relates to other
theoretically relevant psychological variables, we computed zero order relations
between each of the primary (psychological) predictor variables (see Table 2).
As expected, the scores on the BIS correlated fairly strongly with the anxiety
sensitivity and t rait anxiety variables, such that greater behavioural inhibition
sensitivity was associated with greater sc ores on these other anxiety-related
measures. This pattern of results suggests that the BIS is related to other theo-
retically relevant psychological vari ables, but not to such a degree that it is
indistinguishable from them (range of shared variance: 13–37
%
). Thus, base d on
these data, it appears the BIS may share in common some features with anxiety
sensitivity, but does not denote the same construct. As has been foun d in other
investigations, anxiety sensitivity, and trait anxiety were si gnificantly positivel y
intercorrelated (Asmundson, Norton, Lanthier, & Cox, 1996; McNally & Eke,
1996), indicating some degree of overlap (Lilienfeld, Turner, & Jacob, 1993),
but not to such a degree that these constructs were indistinguishable (range of
shared variance: 3–35
%
; McNally, 1994, pp. 106–118). Finally, consistent with
theoretical perspectives that suggest two primary systems are involved with the
regulation of behaviour, BAS factors showed little relation with BIS score s or
with the other anxiety-related variables (range of shared variance: 1–3
%
).
We then computed zero order correlations between the predictor variables
and each of the dependent variables to explore the relations among these
theoretically relevant factors. For ease of presentation, we collapsed across CO
2
TABLE 2
Zero order relations between psychological varia bles (
n
= 60)
Instrument 1 2 3 4 5 6 7 8 9
1. BIS – .61** .61** .56** .35** .59**
7
.04 .03 .18
2. ASI (total score) – – .98** .90** .42** .58** .08 .13 .19
3. ASI (physical) – – – .83** .37** .57** .13 .13 .21
4. ASI (psychological) – – – – .24 .56**
7
.08 .12 .02
5. ASI (social) – – – – – .17 .08 .05 .23
6. S TAI-trait – – – – – – .16 .12
7
.04
7. BAS (drive) – – – – – – – .11 .28*
8. BAS (fun) – – – – – – – – .27*
9. BAS (reward) – – – – – – – – –
Note: BIS/BAS, Behavioral Inhibition and Activation Scale (Carver & White, 1994); ASI,
Anxiety Sensitivity Index (Reiss et al., 1986); STAI, State-Trait Anxiety Inventory-Trait (Spielberger
et al., 1983).
* p
<
.05; ** p
<
.01.
PREDICTION OF ANXIETY AND FEAR
777
trials because the observed relations did not vary in a meaningful manner across
inhalations (see Table 1). That is, there was no e vidence of statistically sig-
nificant habituation or sensitisation patterns wi thin this experimental session, a
findin g consistent with previous research involving exposure to aversive
stimulation (Forsyth & Lejuez, 1999; Van den Hout, van der Molen, Griez,
Lousberg, & Nansen, 1987). Furthe rmore, this methodol ogical tactic reduced the
overall number of analyses, limiting the probability of Type I error.
BIS scores showed a significant relation with self-reported anxiety and
affective ratings, such that greater behavioural inhibition sensitivity was asso-
ciated with greater levels of emotional distress. A generally similar pattern of
results was apparent for the relation between ASI global score, ASI psycholo-
gical concerns, and ASI physical concerns dimensions and the SUDs and SAM
ratings. The ASI social concerns dimension did not show a significant relation
with any of the dependent measures. The STAI significantly c orrelated only
with the SAM arousal dimension, indicating trait anxiety may be particularly
likely to be related to perceived bodily activi ty. Pre-experimental SUDs ratings
correlated significantly with SUDs scores in response to the CO
2
challenge, but
not with other aspects of emotional response to the challenge. Overall, none of
the pre-experimental psychological variables were significantly related t o heart
rate, suggesting a greater degree of discordance between psychological and
physiological domains relative to psychological and affective domains .
Prediction of challenge responding. Hierarchical multiple regression
analyses were performed with each of the primary dependent measures.
Independent variables were divided into three levels in the hierarchy: (a) gender
and age were at level one, (b) pre-experimental SUDs ratings at level two, and
(c) BIS, anxiety sensitivity first order factors, and STAI at level three. We did
not inc lude ASI total scores in the equat ions because it was conceptually
redundant with the inclusion of its lower order scores.
2
First- and second-level
variables were entered into the analysis, and the stepwise method was used for
third-level variables. As before, we collapsed across CO
2
trials because the
observed relations did not vary in a meaningful manner across inhalations.
3
Overall, the hierarchical multiple regression analysis predicting SUDs scores
indicated that the predictor vari ables together explained 39
%
of the variance in
2
However, to empirically address t his issue, we computed separate regression analyses with ASI
total score without the inclusion of the lower order ASI scores. The total ASI scores significantly
predicted response to challenge in the same manner as th e lower-order scores, which suggest the
relative ‘‘predi ctive power’’ of the index could be due to specific subdimensions.
3
We also computed separate regression analyses for each inhalation (i.e., did not collapse across
trials). The results from these analyses did not differ in either the pattern nor size of observed effect
relative to those that are reported.
778
ZVOLENSKY ET AL.
self-reporte d fea r, F(3, 56) = 16.6, p
<
.001 (adjusted R
2
= 36
%
). In step 1, only
gender was a significant predictor of self-reported anxiety, explaining 23
%
of
the variance (adjusted R
2
= 21
%
; p
<
.001); females responded with greater fear
than males. In step 2, pre -experimental SUDs ratings predicted an additional 7
%
of the variance (adjusted R
2
= 6
%
; p
<
.02, Sr = .19). When the block of
psychological predictor variables were entered in step 3, only the ASI physical
concerns met criteria for entry into the equation, predicting an additional 10
%
of
the variance (adjusted R
2
= 9
%
; p
<
.01, Sr = .38).
In regard to the SAM ratings of self-perceived arousal, regression analysis
indicated that the predictor variables together explained 53
%
of the variance,
F(3, 56) = 16.4, p
<
.001 (adju sted R
2
= 52
%
). In step 1, gender was the only
significant predictor, explaining 15
%
of the variance (adjusted R
2
= 13
%
; p
<
.001); again, females reported greater arousal than males. In step 2, pre-
experimental SUDs ratings did not significantly predicted SAM arousal ratings
(adjusted R
2
= 3
%
, n.s.). In step 3, an additional 38
%
(adjusted R
2
= 36
%
) was
predicted by the psychological variables. Only BIS scores (p
<
.01, Sr = .39) and
ASI physical concerns scores (p
<
.01, Sr = .35) met criteria for entry into the
equation, indicating that they served as significant independent psychological
predictors of SAM arousal ratings.
For the SAM ratings of emotional valence, the predictor variables together
explained 34
%
of the variance, F(3, 56) = 10.7, p
<
.001 (adjusted R
2
= 30
%
). In
step 1, only gender was a significant predictor, explaining 14
%
of the variance
(adjusted R
2
= 12
%
; p
<
.01). In step 2, pre-experimental SUDs ratings did not
significantly predict additional variance (adjusted R
2
= 2
%
, n.s.). When the
block of psychological predictor variables were entered in step 3, only the AS I
cognitive concerns met criteria for ent ry into the equation, predicting an addi-
tional 17
%
of the variance (adjusted R
2
= 16
%
; p
<
.001, Sr = .45).
In terms of the SAM ratings of emotional control, the predictor variables
together explained 41
%
of the variance, F(3, 56) = 9.2, p
<
.001 (adjusted R
2
=
39
%
). In steps 1 and 2, there were no significant predictors (combined level 1
and 2 adjusted R
2
= 10
%
, n.s.). When t he block of psychological predictor
variables were entered in step 3, both the ASI cognitive concerns (p
<
.01, Sr =
.53) and B IS (p
<
.01, Sr = .32) significantly predicted control/dominance
ratings, predicting a n additional 19
%
and 10
%
of the variance, respectively.
Although we examined heart rate in the same manner as the other dependent
measures, we found no significant predictors at all.
DISCUSSION
Contemporary research suggests that persons who respond with heightened
levels of anxiety-related responding to bodily perturbation fear the consequences
of somatic sensa tions, experiencing them as dangerous and personally threa-
tening (Holloway & McNally, 1987). Moreover, there is emerging evidence that
PREDICTION OF ANXIETY AND FEAR
779
the relative degree of congruity between dispositionally based anxiety processes
and induced-interoceptive sensations is a critical aspect of such responding, such
that that elevated anxiety i s primarily produced when trigged by specific fear
cues that are most similar and salient to the relevant fear domain (Schmidt,
Lerew, & Jackson, 1999; Zvolensky, Goodie, McNeil, Sperry, & Sorrell, 2001).
Although these findings have helped to refin e our understanding of the cognitive
processes involved with fear production, it is unknown whether other theoreti-
cally relevant psychological processes also are related anxious and fearful
responding. Considering contemporary theories of behavioural inhibition (Clo-
ninger, 1987; Gray, 1990; Kagan, 1989), cognitive-based BIS theoretically also
should be predictive of challenge-induced anxiety and fear.
The present investigation tested whether specific anxi ety sensi tivity dimen-
sions and BIS differentially predict various aspects of anxiety-related respond-
ing to repeated 20
%
CO
2
challenge. Consistent with existing research (Eifert et
al., 1999; McNally & Eke, 1996), ASI physical conc erns was the only psy-
chological variable that significantly predicted post-challenge self-reported fear
after taking into consideration relevant demographic variables and anticipatory
anxiety. In a related way , both the ASI physical concerns dimension and BIS
predicted affective ratings of somatic arousal. From a content -specificity per-
spective, these results are consistent with the hypothesis that the greater level of
correspondence between the particular (pre-existing) anxiety sensitivity concern
and feared event, the better fear responding can be predicted (Cox, 1996). At the
same time, considering affective ratings of arousal, the results indicate that a
general sensitivity to challenge-induced arousal (i.e., behavioural inhibition
sensitivity) also is independently i nfluential in the prediction of arousal-based
emotional distress. This finding is consistent with evidence that suggests
behavioural inhibi tion is associated wit h heightened emotional distress to
physiological activity (Garcia-Coll et al., 1984; Rosenbaum et al., 1988). As
CO
2
challenge produces elevated visceral arousal, a defining symptom of
specific (interoceptive) fears (Lang, Levin, Miller, & Kozak, 1983), it seems
reasonable to conclude that cognitive concerns that correspond to a sensitivity to
such unfamilia r and potentia lly threatening stimulation is important in the pre-
diction of self-reported fear and a rousal-based affective ratings.
Inspection of the results for the other affective response components indi-
cated a different pattern of findings. Specifically, the ASI psychological con-
cerns dimension was significantly predictive of affective ratings of emotional
displeasure and dyscontrol, whereas BIS was si gnificantly and independently
predictive of only affective ratings of limited e motional cont rol. No other
psychological variables remained in the equations after taking into consideration
anticipatory anxiety and relevant dem ographic variables. These findings are
consistent with perspective that although 20
%
CO
2
inhalation e vokes bodily
perturbation, it also can activate concomitant cognitive reactions, particularly in
those in those persons concerne d about negative psychological consequences
780
ZVOLENSKY ET AL.
(e.g., losing control, going crazy) of such sensations and/or with a general
sensitivity to potentially aversive events. Schmidt et al. (1999) recently found a
similar pattern of results in regard to the ASI dimensions in a large nonclinical
sampl e of military c adets during a 5 week basic cadet training. Specifically, the
psychological concerns subscale of the ASI was the only lower order anxiety
sensitivity dimension that predicted panic attacks after c ontrolling for a personal
history of panic attacks an d trait anxiety. These da ta converge with evidence that
suggests there may be particular cognitive features that promote anxiety-related
responding to bodily arousal (Lang et al., 1983, 1990; Schmidt, 1999).
Similar to other researchers (Asmundson, Norton, Wilson, & Sandler, 1994),
we found no predictors for heart rate during the provocation. These findings are
interesting given that theoretical ac counts of emotion indicate emotions are
expressed by partially independent response systems (Barlow, 1988; Lang, 1980).
The results indicate greater levels of response concordance between cognitive
variables and affective responding to evocative stimulation, and greater levels of
response discordance between cognitive variables and physiological responding
to evocativ e stimulation. Such relative response concordance-discordanc e is a
well-established finding in the emotion reactivity literature generally and in
biological challenge studie s spec ifically (see Zvolensky & Eifert, 2000, for a
discussion of this issue). Indeed, Forsyth, Eifert, and Canna (2000) have found
that there is marked individual variation in terms of how one experiences panic.
For instance, strong cognitive-affectiv e reactions have been observed in the
absence of physiological arousal or pathology, and conversely, marked somatic
changes can occur in the absence of cogni tive-affective reactions of anxiety and
fear (e.g., Kushner & Beitman, 1990). One the one hand, these findings can be
interpreted as evidence in support of theoretical views that emotional experiences
are best understood as partially independent response systems rather than a more
unified construct (Bouton et al., 2001). On the other hand, it is plausible that
variance pertaining to the different me thodologies used to assess these different
content are as determines, at least in part, this observation (Cone, 1998). To
continue to unravel the interrelation and patterning between various response
systems, researchers may find it helpful to include other physiological processes
involved with the defensive system presumed to be active duri ng anxiety and fear
states (LeDoux, 1990). As one example, resea rchers could incorporate the direct
assessment of central arousal (e.g., EEG) and parasympa thetic activity (e.g., vagal
tone) as well as other indices of autonomic activity (e.g., skin conductance) in
biological challenge studies.
There are a number interpretative-related caveats and directions for future
research that warrant consideration. First, all of the psychological variables were
assessed using self-repor t instruments. Accordingly, it is possible the observed
findings were, at least in part, due to shared me thod variance. Although such a
methodological strategy was useful at this stage of research development, it
likely will be useful to incorporate experimental cognitive methodologies (e.g.,
PREDICTION OF ANXIETY AND FEAR
781
dot probe, Stroop) into future research in this area, particularly those tha t tap
automatic types of processing (e.g., interpretative biases for threat). Second, we
used the 16-item ASI in the present study even t hough it does not offer an
expansive amount of assessment information for all possible sensitivity domains
For example, cardiopulmonary concerns ar e only one part of physical concerns
on the 16-item ASI an d the se particular physical concerns were found to be most
relevant to anxiety-related responding to CO
2
challenge by Schmidt (1999). Our
decision to employ the 16-item ASI at this junc ture was based upon its well-
documented psychometric properties, and the fact that it has been the most
frequently used of the available anxiety sensitivity measures (Taylor, 1995).
Still, future researchers may find it useful to employ the recently developed
Revised Anxiety Sensitivity Index (Taylor & Cox, 1998) which includes a larger
number of items and more detailed assessment of various types of physical
concerns (cardiovascular vs. respiratory vs. neurological vs. gastrointestinal)
which could be useful in CO
2
-challenge work. For instance, in the 16-item
version of the ASI the social subscale is comprised of only two items, weak-
ening its psychometric properties (e.g., internal consistency). Finally, as with all
quasi-experimental designs, causal relations cannot be unambiguously i nferred,
leaving the results open to a number of interpretations. One way to better
address this issue would be to provide a prospective assessment of how parti-
cular psychological variables relate to anxious and fearful responding to arousal-
related bodily sensations in the natural environment.
In summary, the results of the present study, in conjunction with other
experimental psychopathology evidence , help to identify how theoretically
relevant variables produce heightened levels of anxious an d fearful responding
to bodily sensations. Our results indicate that at a cognitive level of analysis,
particular anxiety sensitivity domains and BIS predict differential aspects of
anxious and fearful responding to recurrent 20
%
CO
2
challenge. These results
converge and uniquely extend the existing evidence that psychological processes
are not merely consequences of panic-related psychopathology, but functi on as
important predictors of anxious responding.
Manuscript received 12 April 2000
Revised manuscript received 24 February 2001
REFERENCES
Asmundson, G.J.G., Norton, G.R., Lanthier, N.J., & Cox, B.J. (1996). Fear of anxiety: Do current
measure s assess unique aspects of t he construct? Personality and Individual Differences, 20,
607–612.
Asmundson, G.J.G., Norton, G.R., Wilson, K.G., & Sandler, L.S. (1994). Subjective symptoms and
cardiac reactivity to brief hyperventilation in individuals with high anxiety sensitivity. Behaviour
Research and Therapy, 32, 237–241.
782
ZVOLENSKY ET AL.
Barlow, D.H. (1988). Anxiety and its disorders. New York: Guilford Press.
Bouton, M.E., Mineka, S., & Barlow, D.H. ( 2001) . A modern learning theory perspective on the
etiology of panic disorder. Psychological Review, 108, 4–32.
Bradley, M.M., & Lang, P.J. (1994). Measuring emotion: The self-assessment manikin and the
semantic differential. Journal of Behavior Therapy and Experimental Psychiatry, 25, 49–59.
Carver, C.S., & White, T.L. (1994). Behavioral inhibition, behavioral activation, and affective
responses to impending reward and punishment: The BIS/BAS scales. Journal of Personality and
Social Psychology, 67, 319–333.
Cloninger, C.R. (1987). A systematic method for clinical description and classification of personality
variants. Archives of General Psychiatry, 44, 573–588.
Cohen, J. (1988). Statistical power analysis (2nd ed.). Hillsdale, NJ: Erlbaum.
Cone, J.D. (1998). Hierarchica l views of anxiety: What do they profit us? Behavior Therapy, 29,
325–332.
Cox, B.J. (1996). The nature and assessment of catastrophic thoughts in panic disorder. Behaviour
Research and Therapy, 34, 363–374.
Davidson, R.J., Ekman, P., Saron, C.D., Senulis, J.A., & Friesen, W.V. (1990). Approach-withdrawa l
and cerebral asymmetry: Emotional expression and brain physiology: I. Journal of Personality
and Social Psychology, 58, 330–341.
Eifert, G.H., Zvolensky, M.J., & Lejuez, C.W. (2000). Heart-focused anxiety and chest pain: A
review. Clinical Psychology : Science and Practice, 7, 403–417.
Eifert, G.H., Zvolensky, M.J., Sorrell, J.T., Hopko, D.R., & Lejuez, C.W. (1999). Predictors of self-
reported anxiety and panic symptoms: An evaluation of anxiety sensitivity, suffocation fear,
heart-focuse d anxiety, and breath-holding duration. Journal of Psychopatholog y and Behavioral
Assessment, 21, 293–305.
Ehlers, A., & Breuer, P. (1992). Increased cardiac awareness in panic disorder. Journal of Abnormal
Psychology , 101 , 371–382.
Forsyth, J.P., & Eifert, G.H. (1998). Intensity of systemic alarms in content-specific fear con-
ditioning: Comparing 20
%
versus 13
%
CO
2
-enriched air as a UCS. Journal of Abnormal Psy-
chology, 107, 291–304.
Forsyth, J.P., Eifert, G.H., & Canna, M.A. (2000). Evoking analogue subtypes of panic attacks in a
nonclinical population using carbon dioxide-enriche d air. Behaviour Research and Therapy, 38,
559–572.
Forsyth, J.P., & Lejuez, C.W. (1999). Anxiogenic effects of repeated administrations of 20
%
CO
2
-
enriched air: Stability withi n sessions and habituation across time. Paper present ed at the
annual meeting of the Association for Advancement of Behavior Therapy convention,
Toronto, Canada.
Garcia-Coll, C., Kagan, J., & Reznick, J.S. (1984). Behavioral inhibition in young children. Child
Development, 55, 1005–1019.
Gray, J.A. (1990). Brain systems that mediate both emotion and cognition. Cognition and Emotion, 4,
269–288.
Gray, J.A. (1987). The psychology of fear and stress. Cambridge, UK: Cambridge University Press.
Gray, J.A., & McNaughton, N. (1996). The neuropsychology of anxiety: A reprise. In D.A. Hope
(Ed.), Nebraska Symposium on Motivation: Perspectives on anxiety, panic, and fear (Vol. 43, pp.
61–134). Lincoln, NE: University of Nebraska Press.
Harmon-Jones, E., & Allen, J.J. (1997). Behavioral activation sensitivity and resting frontal EEG
asymmetry: Covariation of putative indicators related to risk for moo d disorders. Journal of
Abnormal Psychology, 106, 159–163.
Holloway, W., & McNally, R.J. (1987). Effects of anxiety sensitivity on the response to hyper-
ventilation. Journal of Abnormal Psychology, 96, 330–334.
Kagan, J. (1989). Temperamental contributions to social behavior. American Psychologist, 44, 668–
674.
PREDICTION OF ANXIETY AND FEAR
783
Kagan, J., Reznick , J.S., & Snidman, N. (1988). Biological bases of childhood shyness. Science, 240,
167–171.
Kushner, M.G., & Beitman, B.D. (1990). Panic attacks without fear: An overview. Behaviour
Research and Therapy, 28, 469–479.
Lang, P.J. (1980). Behavioral treatment and bio-behavioral assessment: Computer applications. In
J.B. Sidowski, J.H. Johnson, & T.A. Williams (Eds.), Technology in the health care delivery
system (pp. 119–137). Norwood, NJ: Ablex.
Lang, P.J. (1984). Cognition in emotion: Concept and action. In C.E. Izard, J. Kagan, & R.B. Zajonc
(Eds.), Emotions, cognitions, and behavior (pp. 192–228). New York: Cambridge University
Press.
Lang, P.J., Bradley , M.M., & Cuthbert, B.N. (1990). Emotion, attention, and the startle reflex.
Psychological Review, 97, 377–398.
Lang, P.J., Levin, D.N., Miller, G.A., & Kozak, M. J. (1983). Fear behavior, fear imagery and the
psychophysiology of emotion: The problem of affective response integration. Journal of
Abnormal Psychology, 92, 276–306.
Larkin, K.T., & Zayfert, C. (1996). Anger management training with mild essential hypertensive
patients. Journal of Behavioral Medicine, 19, 415–433.
LeDoux, J.E. (1990). Information flow from sensation to emotion plasticity in the neural computation
of stimulus values. In M. Gabriel & J. Moore (Eds.), Learning and computationa l neuroscience:
Foundations of adaptive networks (pp. 3–52). Cambridge, MA: Bradford Books/MIT Press.
Lejuez, C.W., Forsyth, J.P., & Eifert, G.H. (1998). Devi ces and methods for administering carbon
dioxide-enriche d air in experimenta l and clinica l settings. Journal of Behavior Therapy and
Experimental Psychiatry, 29, 239–248.
Lejuez, C.W., O’Donell, J., Wirth, O., Zvolensky, M.J., & Eifert, G.H. (1998). Avoidance of carbon
dioxide-enriche d air in humans. Journal of the Experimental Analysis of Behavior, 70, 79–86.
Lilienfeld, S.O., Turner, S.M., & Jacob, R.G. (1993). Anxiety sensitivity: An examination of theo-
retical and met hodologica l issues. Advances in Behaviour Research and Therapy, 15, 147–183.
Manuck, S.B., Kamarck, T.W., Kasprowicz, A.S., & Waldstein, S.R. (1993). Stability and patterning
of behaviorally evoked cardiovascular reactivity. In J. Blascovich & E.S. Katkin (Eds.), Cardi-
ovascular reactivity to psychological stress and disease: An examination of the evidence (pp. 83–
108). Washington, DC: American Psychological Association.
McNally, R.J. (1999). On the experimental i nduction of panic. Behavior Therapy, 30, 333–341.
McNally, R.J. (1998). Information-processin g abnormalities in anxiety disorders: Implications for
cognitive neuroscience . Cognition and Emotion, 12, 479–495.
McNally, R.J. (1994). Panic disorder: A critical analysis , New York: Guilford Press.
McNally, R.J., & Eke, M. (1996). Anxiety sensitivity, suffocation fear, and breath-holding duration
as predictors of response to carbon dioxide challenge. Journal of Abnormal Psychology, 105,
146–149.
McNeil, D.W., Vrana, S.R., Melamed, B.G., Cuthbert, B.N., & Lang, P.J. (1993). Emotional imagery
in simple and social phobia: Fear versus anxiety. Journal of Abnormal Psychology, 102, 212– 225.
Peterson, R.A., & Reiss, S. (1992). Anxi ety sensitivity index manual (2nd ed.). Worthington, OH:
International Diagnostic Systems.
Rachman, S., & Taylor, S. (1993). Analyses of claustrophobia . Journal of Anxiety Disorders, 7, 281–
291.
Rapee, R.M. (1995). Psychologica l factors influencing the affective response to biological challenge
procedure s in panic disorder. Journal of Anxiety Disorders, 9, 59–74.
Rapee, R., & Medero, L. (1994). Fear of physical sensations and trait anxiety as mediators of the
response to hyperventilation in nonclinical subjects. Journal of Abnormal Psychology, 4, 693–699.
Rassovsky, Y., Kushner, M.G., Schwarze, N.J., & Wangensteen, O.D. (2000). Psychological and
physiological predictors of response to carbo n dioxide challenge in individuals with panic dis-
order. Journal of Abnormal Psychology, 108, 616–623.
784
ZVOLENSKY ET AL.
Reiss, S., Peterson, R.A., Gursky, M., & McNally, R.J. (1986). Anxiety, sensitivity, anxiety
frequency , and the prediction of fearfulness. Behaviour Research and Therapy, 24, 1–8.
Reznick, J.S., Kagan, J., Snidman, N., Gersten, M., Baak, M.G., & Rosenberg, A. (1986). Inhibited
and uninhibited children: A follow-up study. Child Development, 57, 660–680.
Rosenbaum , J.F., Biederman, J., Gersten , M., Hirshfeld, D.R., Meminger, S.R., Herman, J.B., Kagan,
J., Reznick, J.S., & Snidman, N. (1988). Behavioral inhibition in children of parents with panic
disorder and agoraphobia . Archives of General Psychiatry, 45, 463–470.
Salkovskis, P.M., & Clark, D.M. (1989). Affective responses to hyperventilation: A test of the
cognitive model of panic. Behaviour Research and Therapy, 28, 51–62.
Sanderson, W.C., & Wetzler, S. (1990). Five percent carbon dioxide challenge: Valid analogue and
marker of panic disorder? Biological Psychiatry, 27, 689–701.
Schmidt, N.B. (1999). Examination of differential anxiety sensitivities in panic disorder: A test of
anxiety sensitivity subdomains in predicting fearful responding to a 35
%
CO
2
challenge. Cog-
nitive Therapy and Research, 23, 3–19.
Schmidt, N.B., Lerew, D.R., & Jackson, R.J. (1999). Prospective evaluation of anxiety sensitivity in
the pathogenesi s of panic: Replication and extension. Journal of Abnormal Psychology, 108,
532–537.
Schmidt, N.B., Lerew, D.R., & Trakowski, J.H. (1997). Body vigilance in panic disorder: Evaluating
attention to bodily perturbations. Journal of Consulting and Clinical Psychology, 65, 214–220.
Schmidt, N.B., & Trakowski, J.H. (1999). Attentional focus and fearful responding in patients with
panic disorder during 35
%
CO
2
challenge. Behavior Therapy, 30, 623–640.
Sobotka, S.S., Davidson, R.J., & Senulis, J.A . (1992). Anterior brain electrical asymmetries in
response to reward and punishment. Electrocencephalograph y and Clinical Neurophysiology, 83,
236–247.
Spielberger, C.D., Gorsuch, R.L., Lushene, R., Vagg, P.R., & Jacobs, G.A. (1983). Manual for the
state-trait anxiety inventory (Form Y). Palo Alto, CA: Consulting Psychologists Press.
Stein, M.B., Jang, K.L., & Livesley, W.J. (1999). Heritability of anxiety sensitivity: A twin study.
American Journal of Psychiatry, 156, 246–251.
Stewart, S.H., Taylor, S., & Baker, J.M. (1997). Gender differences in dimensions of anxiety sen-
sitivity. Journal of Anxiety Disorders, 11, 179–200.
Sutton, S.K., & Davidson, R.J. (1997). Prefrontal brain asymmetry: A biological substrate of the
behavioral approach and inhibition systems. Psychological Science, 8, 204–210.
Taylor, S. (1995) . Anxiety sensitivity: Theoretical perspectives and recent findings. Behaviour
Research and Therapy, 33, 243–258.
Taylor, S., & Cox, B.J. (1998). Anxiety sensitivity: Multiple dimensions and hierarchic structure.
Behaviour Research and Therapy, 36, 37–51.
Taylor, S., Koch, W.J., & McNally, R.J. (1992). How does anxiety sensitivity vary across the anxiety
disorders? Journal of Anxiety Disorders, 6, 249–259.
Van den Hout, M.A., van der Molen, G.M., Griez, E., Lousberg, H., & Nansen, A. (1987). Reduction
of CO
2
-induced anxiety in patients with panic attacks after repeated CO
2
exposure. American
Journal of Psychiatry, 144, 788–791.
Wolpe, J. (1958). Psychotherapy by reciprocal inhibition. Stanford, CA: Stanford University
Press.
Zinbarg, R.E. (1998). Concordance and synchrony in measures of anxiety and panic reconsidered: A
hierarchical model of anxiety and panic. Behavior Therapy, 29, 301–323.
Zinbarg, R.E., Mohlman, J., & Hong, N.N. (1999). Dimensions of anxiety sensitivity. In S. Taylor
(Ed.), Anxiety sensitivity: Theory, research, and treatment of the fear of anxiety (pp . 83–114).
Mahwah, NJ: Erlbaum.
Zvolensky, M.J., & Eifert, G.H. (2000) . A review of psychol ogica l factors/processes affecting
anxious respondi ng during voluntary hyperventilation and inhalations of carbon dioxide-enriched
air. Clinical Psychology Review, 21, 375–400.
PREDICTION OF ANXIETY AND FEAR
785
Zvolensky, M.J., Eifert, G.H., Lejuez, C.W., & McNeil, D.W. (1999). The effects of offset control
over 20
%
carbon dioxide-enriched air on anxious responding. Journal of Abnormal Psychology,
108, 624–632.
Zvolensky, M.J., Goodie, J.L., McNeil, D.W., Sperry, J.A., & Sorrell, J.T. (2001). Anxiety sensitivity
in the prediction of pain-related fear and anxiety in a heterogeneous chronic pain population.
Behaviour Research and Therapy, 39, 683–696.
Zvolensky, M.J., Lejuez, C.W., & Eifert, G.H. (1998). The role of control in anxi ous responding: An
experimental test using repeated administrations of 20
%
CO
2
-enriched air. Behavior Therapy, 29,
193–209.
Zvolensky, M.J., McNeil, D.W., Porter, C.A., & Stewart, S.H. (in press). Assessment of anxiety
sensitivity in young American Indians and Alaska Natives. Behaviour Research and Therapy.
786
ZVOLENSKY ET AL.