Anxiety-Control Strategies Is There Room For Neutralization in Successful Exposure Treatment

Anxiety-control strategies 1
Running Head: ANXIETY-CONTROL STRATEGIES

Anxiety-Control Strategies: Is there Room for Neutralization in Successful Exposure Treatment?

Chris L. Parrish, Adam S. Radomsky*, and Michel J . Dugas
Department of Psychology, Concordia University

*Corresponding address: Department of Psychology, Concordia University, 7141 Sherbrooke
St. West, Montreal, QC, H4B 1R6, CANADA.
Abstract
Cognitive-behavioral theory suggests that anxiety-control strategies such as neutralization,
distraction and various forms of safety behavior have the potential to diminish the effectiveness
of and/or interfere with exposure treatment. Yet, it is common practice when treating individuals
with anxiety disorders to employ various anxiety-control strategies as a means of assisting
clients/patients with difficult exposure situations. Questions surrounding the issue of which
anxiety-control strategies help vs. hinder exposure-based treatments (and under which
circumstances) have been a topic of much investigation and continue to be a focus of theoretical
debate. The present article reviews several key studies which collectively shed some light on
this debate. The evidence suggests that clients anxiety-control strategies may be less likely to
become counter-productive when: (i) they promote increases in self-efficacy, (ii) they do not
demand excessive attentional resources, (iii) they enable greater approach behavior and
integration of corrective information (via disconfirmatory experiences), and (iv) they do not
promote misattributions of safety. Theoretical and clinical implications of these findings are
discussed, and future directions for research in this area are suggested.

Keywords: anxiety; neutralization; safety behaviour; distraction; exposure.
Anxiety-Control Strategies: Is there Room for Neutralization in Successful Exposure
Treatment?
Clinically anxious individuals use a number of strategies to control unpleasant thoughts,
images and emotions. Common examples of these strategies include direct (behavioral)
avoidance, thought suppression, overt compulsions, and various forms of subtle avoidance (e.g.,
distraction, mental rituals, safety-seeking behaviors, etc.). For the purposes of this review, the
terms anxiety-control strategies and anxiety-neutralizing behavior will be used
interchangeably to refer to these collective acts, given their proposed common function (i.e., of
controlling - or neutralizing - anxiety).
Cognitive-behavioral theories of anxiety disorders (e.g., Beck, Emery & Greenberg,
1985; Clark, 1999; Salkovskis, 1996) suggest that avoidance and other forms of anxiety-
neutralizing behaviors are counter-productive, such that they provide temporary relief from fear
and discomfort, yet maintain anxiety in the long run. Nevertheless, many (subtle) avoidant
strategies are utilized in clinical practice as a means of easing clients into anxiety-provoking
treatment situations. Indeed, Craske, Street and Barlow (1989) note that distraction is used
often, both by clients as a method of coping with high levels of anticipatory anxiety, and by
therapists in their instruction to clients of ways to approach feared situations (p.664).
Salkovskis, Clark and Gelder (1996) make the theoretical distinction between adaptive coping
strategies (e.g., rationalization, avoidance of real threats), which are employed to manage
anxiety, and maladaptive safety behaviors (e.g., neutralization, carrying safety aids, avoiding
perceived danger, etc.), which are intended to prevent the occurrence of feared catastrophes. It is
hypothesized that maladaptive safety behaviors prevent the unambiguous disconfirmation of
negative beliefs, thus maintaining the perceived validity of these beliefs and related fears
(Salkovskis, 1996). However, in clinical practice it is often difficult for therapists to discern
whether their clients coping efforts may be counter-productive, and evidence examining this
issue is mixed (Thwaites & Freeston, 2005). Thus, it is important to consider the following
question: Under what circumstances (if any) can clients anxiety-neutralizing behavior facilitate
exposure treatment for anxiety disorders, and what are the mechanisms involved?
The current discussion presents a selective review of findings that pertain to this question.
Although there is currently no clear consensus regarding the defining features of many of the
constructs under review, we compare studies that have examined the effects of similar anxiety-
control strategies on exposure-driven fear reduction. Because the counter-productive effects of
direct (behavioral) avoidance are well-established, research on this issue is not reviewed here.
Likewise, the consequences of thought suppression (e.g., Purdon, 1999, 2004; Purdon, Rowa, &
Antony, 2005; Rassin, Merckelbach, & Muris, 2000) and overt compulsions (e.g., Rachman,
2002; Salkovskis, 1999) have recently been reviewed elsewhere, and thus, are not covered here.
The present review focuses on three broad areas of investigation: (i) the effects of distraction on
fear reduction both during and after exposure (ii) the effects of safety behavior on anxiety and
fear-related cognitions, and (iii) the effects of neutralization on subsequent anxiety/discomfort
and urges to neutralize. Theoretical and clinical implications of these findings are discussed, and
promising directions for further research are suggested.
Theoretical Background
Mechanisms of Fear Reduction
While anxiety disorders are among the most effectively treated forms of
psychopathology, the mechanisms by which anxiety and fear reduction occur during exposure
treatment are not yet fully understood (Hofmann, 2008; Oliver & Page, 2003; McNally, 2007;
Schmid-Leuz, Elsesser, Lohrmann, J hren & Sartory, 2007). Traditionally, behavior theorists
have relied on habituation models (e.g., Groves & Thompson, 1970) to explain these processes.
These models suggest that conditioned fear responses are subject to extinction with repeated and
prolonged exposures to feared stimuli, similar to processes involved in habituation to novelty
(Agras, 1965; Mowrer, 1939; Watts, 1971). Exposure duration, stimulus intensity and attention
to phobic cues are hypothesized to be key moderators of fear extinction (Watts, 1971, 1974;
Watts, Trezise, & Sharrock, 1986). Accordingly, it is predicted that events or behaviors that
interfere with these essential components of exposure should compromise the amount of fear
reduction achieved, as well as increase the probability that the individual will experience a return
of fear upon subsequent exposures (Watts, 1974). For example, distraction and other forms of
cognitive and/or behavioral avoidance during exposure are hypothesized to negatively impact
upon fear reduction (Rodriguez & Craske, 1993).
The concept of emotional processing (Rachman, 1980; Foa & Kozak; 1986; Foa,
Huppert, & Cahill, 2006) was later proposed in an attempt to explain the mechanisms of
exposure-driven fear reduction from an information processing (i.e., cognitive) perspective. Foa
and Kozaks (1986) theory, which elaborates on the earlier work of Lang (1977, 1984) and
Rachman (1980), proposes that feared stimuli and their meanings are represented in memory as
fear structures or prototypes that consist of associations between fearful emotions,
cognitions and behavior. These fear structures can be accessed upon exposure to the
corresponding feared stimuli. However, it is hypothesized that for lasting fear reductions to
occur: (i) the fear structure must be fully activated in memory (as evidenced by heightened
emotional arousal and self-reports of fear), and (ii) internal representations of the feared stimulus
must be modified through corrective information that highlights the innocuous nature of the
stimulus and thereby decreases harm expectancy (see also Hofmann, 2008). It was originally
proposed that these necessary conditions serve to weaken associations between elements of the
fear structure and allow extinction of the fear response to occur. However, Foa et al. (2006) have
noted that recent developments in animal research (e.g., Bouton, 2004; Myers and Davis, 2002;
Rescorla, 1996) suggest that learned safety information creates a new set of associations between
fear cues and safety (i.e., a safety structure) which inhibits the fear response, rather than
replacing and/or modifying the original fear structure. Notwithstanding this conceptual revision,
the emotional processing model predicts that behavioral and cognitive avoidance strategies (e.g.,
safety behaviour, distraction, etc.) that interfere with the activation and/or modification of fear
structures (or learning of new safety structures?) during exposure should hinder fear reduction
and promote the return of fear.
In contrast, Banduras (1977) social learning theory states that activities that diminish
emotional arousal and enhance an individuals perceived sense of personal mastery and control
(components of self-efficacy) in anxiety-provoking situations should facilitate fear reduction.
For example, distraction and other subtle avoidance strategies may help to diminish anxiety in
fearful situations, thus allowing phobic individuals to approach feared stimuli for longer periods
and to gain a greater sense of mastery over their fears (Rodriguez & Craske, 1993). Therefore,
Banduras social learning framework grants the possibility for anxiety-reducing coping
mechanisms to facilitate fear reduction under certain circumstances.
Finally, behavioral neuroscientists have begun to uncover biological mechanisms that
may influence exposure-driven fear reduction. For example, it has been established that N-
methyl-d-aspartate (NMDA) receptor activity in the amygdala plays an important role in
diminishing fear in both animals and humans through exposure (Davis, 2002; Walker, Ressler,
Lu, & Davis, 2002). Consistent with this interpretation, recent investigations involving
individuals suffering with Social Phobia (Hofmann et al., 2006) and Acrophobia (i.e., fear of
heights) (Ressler et al., 2004) have found that administering the NMDA agonist d-cycloserine
(DCS) shortly before exposure reliably enhances treatment benefits (i.e., reduction of fear and
anxiety symptoms) in both the short- and long-term. Given that DCS administration only
hastens extinction when used in conjunction with exposure (Walker et al., 2002), it has been
hypothesized that NMDA receptor activity may serve to consolidate learning of corrective
information (i.e., safety in the presence of feared cues). In addition, based on animal research
(Milad & Quirk, 2002) which has found a negative correlation between activity in the medial
prefrontal cortex (mPFC) and return of fear, McNally (2007) has hypothesized that any
intervention that can boost activity in the mPFC during exposure to fear provoking stimuli may
yield therapeutic benefits. However, it is not currently known how various anxiety-control
strategies may affect these cortical processes, and further investigation is required to determine
their effects.
Factors Involved in the Maintenance of Fear and Anxiety
Salkovskis (1996) describes the neurotic paradox as the observation that clinically
anxious individuals fears persist despite experiencing repeated safe encounters with anxiety-
provoking situations. Behavior theorists have attempted to explain the persistence of these fears
with the concept of sensitization due to premature termination of exposure (Battersby, 2000).
However, such accounts fail to explain instances in which fear re-emerges following prolonged
exposure. Cognitive theories (Clark, 1999; Salkovskis, 1996; Salkovskis et al., 1996) provide a
simple explanation for this paradox; it is proposed that phobic individuals use of subtle safety-
seeking behaviors during exposure maintains fear and anxiety by preventing the disconfirmation
of catastrophic cognitions. This effect is clearly illustrated in the following hypothetical
example: A client who believes that they will faint if their anxiety escalates beyond a certain
threshold may sit down or lean on a wall for support whenever they feel a sudden emergence of
anxiety symptoms. When their anxiety passes and they do not faint, they are likely to consider
the situation a near miss and to attribute the non-occurrence of the feared event (i.e., fainting)
to their preventive efforts, thereby reinforcing their maladaptive beliefs, predictions and behavior
(Salkovskis, 1996). As such, it is proposed that individuals who routinely employ safety
behaviors in anxiety-provoking situations are more likely to experience a return of fear upon
subsequent encounters with the feared situation(s). It is also hypothesized that some anxiety-
control strategies have the unintended effect of increasing the individuals anxious response
(e.g., sweating, blushing, trembling, etc.), thus initiating a vicious cycle of anxious symptoms
and behavior (Clark, 1999; Salkovskis, 1996). Accordingly, proponents of the cognitive theory
suggest that exposure treatment can only be maximally effective if these subtle avoidance
behaviors are eliminated.
Another phenomenon that may contribute to the long-term maintenance of fear and
anxiety is the over-prediction of fear. The over-prediction of fear simply refers to an
overestimation of how frightened one will be when encountering an anxiety-provoking situation.
Rachman and colleagues (Rachman & Bichard, 1988; Rachman & Lopatka, 1986a, 1986b)
propose that over-prediction of fear typically follows aversive experiences in which fear was
under-predicted (e.g., an unexpected panic episode). The extreme fear associated with these
unanticipated panic episodes is hypothesized to subsequently increase anticipatory anxiety and
fear predictions in situations that resemble the original experience. According to Rachman and
Bichard (1988):
The tendency to over-predict fear remains relatively unchanged unless and until
disconfirmations occur. In most clinically significant fears that are accompanied or
followed by strongly avoidant behavior, the possibilities of experiencing such
disconfirmations are limited. In this way, their over-predictions of fear can be preserved
from disconfirmation, and continue relatively unchanged (p.308).
Thus, again a vicious cycle is created in which the over-prediction of fear encourages the use of
counter-productive avoidant strategies (including subtle avoidance), which, in turn, maintain
over-predictions.
Background
Distraction vs. Focused Attention During Exposure
Borkovec and Grayson (1980) suggested that the objective presentation of stimuli does
not guarantee functional exposure to those stimuli (p.118, emphasis added). They further
implied that the amount of fear reduction achieved as a result of exposure should be less in cases
where information processing is compromised (e.g., due to distraction).
The first study to explicitly test these claims was conducted by Grayson, Foa and
Steketee (1982), who manipulated participants focus of attention during in vivo exposure and
examined the effects on fear reduction. In this study, individuals diagnosed with Obsessive-
Compulsive Disorder (OCD) were exposed to a highly-feared contaminant for 90 minutes on
each of two consecutive days. Each participant completed the exposure under conditions of both
distracted and focused attention, in a counterbalanced design. In the distracted exposure
condition, participants held the contaminated object with one hand, while playing a video game
with the other hand. In the focused exposure condition, the experimenter had participants talk
about the contaminant they were holding and the discomfort it aroused. Consistent with the
notion that distraction prevents functional exposure, participants who underwent the distracted
exposure condition on the first day demonstrated significantly less between-session habituation
than those who were instructed to focus on the feared stimulus during the first exposure.
The same team of researchers attempted to replicate this study using a between-
participants design (Grayson, Foa, & Steketee, 1986). Although an identical protocol was used,
their findings were at odds with those of the original experiment. Both groups failed to
demonstrate between-session anxiety reduction, and contrary to expectation, participants who
completed exposure while distracted reported greater within-session anxiety reduction than
participants in the focused condition. In fact, the only finding that suggested an advantage for
focused exposure was that participants heart rate gradually decreased during exposure in the
focused condition, while high levels of physiological arousal were maintained throughout the
exposure in the distracted group.
Craske, Street, J ayaraman and Barlow (1991) studied individuals with snake and spider
phobias to determine: (i) how distraction during in vivo exposure affects phobics experience of
fear in the short term, and (ii) whether phobic individuals who are not given any specific
instructions during exposure will demonstrate a natural tendency to use cognitive avoidance (i.e.,
distraction) to cope with fearful encounters. Using a repeated measures design, student
participants were exposed to either a live snake or spider (depending on their primary fear) under
three conditions: natural exposure, focused exposure, and distracted exposure. Distracted
exposure involved listening for key words presented in an audio-taped message and indicating
each time the key words were played by placing a check mark on a sheet of paper, while
maintaining visual focus on the feared animal. In contrast, the focused exposure condition
involved listening to an audio-taped passage that included instructions to maintain both visual
and attentional focus on the feared stimulus (e.g., by examining different aspects of the animal
closely). The order of focused and distracted exposure conditions was counter-balanced, and in
all cases, was preceded by the natural exposure condition and followed by a baseline assessment
period. It was found that participants experienced less subjective fear during distracted exposure
than during focused exposure, regardless of the order of conditions, while heart rate remained
stable across all conditions. Also, subjective fear ratings in the natural exposure condition most
closely resembled those provided by participants in the distracted condition. The authors
concluded that distraction may inhibit the immediate elicitation of fear in anxiety-provoking
situations, and that phobic individuals may naturally tend to counteract attentional biases towards
threat with cognitive avoidance.
In a study of individuals with Claustrophobia, Kamphius and Telch (2000) tested several
predictions derived from emotional processing theory. Participants were randomly assigned to
one of four exposure conditions: (i) guided threat reappraisal (GTR), (ii) cognitive load distractor
task (CL), (iii) GTR +CL, or (iv) exposure only (EO). In the GTR condition, participants were
told to attend to evidence concerning the validity of their core fears, while participants in the CL
condition performed a demanding dual-process distractor task. Participants in the GTR +CL
condition were given both sets of instructions, but were told to prioritize the distractor task.
Overall, participants completed 30 minutes of in vivo exposure, which was broken down into
blocks lasting a maximum of 5 minutes each. A number of fear indices (e.g., subjective anxiety
ratings, peak fear, heart rate variability, etc.) were collected throughout the procedure and
participants were classified according to end-state functioning at post-treatment and follow-up (2
weeks) assessments. Results were generally consistent with predictions derived from emotional
processing theory. Although participants in all four conditions demonstrated significant within-
trial habituation, those in the GTR conditions reported significantly greater reductions in
subjective anxiety from pre- to post-exposure than participants in the CL condition. Also, a
statistical trend suggested that participants in the two distraction groups showed greater return of
fear at 2-week follow-up. Thus, in general, results showed that participants engagement in a
cognitive load task during exposure hindered the amount of fear reduction achieved, whereas
instructions to test negative beliefs led to greater symptom reduction, indicating that cognitive
factors play a role in exposure-based fear reduction.
More recently, Telch and colleagues (2004) asked a group of individuals reporting
extremeclaustrophobic fear to complete a total of 30 minutes of in vivo exposure under one of
four conditions: (i) increased threat, (ii) neutral, (iii) cognitive load (CL), or (iv) exposure only
(EO). The increased threat condition consisted of attending to fear-relevant threat words (e.g.,
trapped, suffocate) and forming images of these words, while participants in the neutral
condition performed the same task with neutral words (e.g., banana). Meanwhile, participants in
the CL condition performed a demanding cognitive task (i.e., the Seashore Rhythm Test;
Halstead, 1947), which required a great deal of their attentional resources. Consistent with
prediction, participants in the CL condition demonstrated less symptom improvement at post-
treatment than participants in the EO condition. However, contrary to what the emotional
processing model (Foa & Kozak, 1986) would predict, individuals in the increased threat
condition did not demonstrate greater fear reduction at post-treatment than participants in the
neutral or EO conditions. This finding is particularly noteworthy given that participants in the
neutral condition performed a threat-irrelevant task which likely functioned as a mild distractor.
It is important to note however, that the threat manipulation was not effective in eliciting greater
fear activation (in terms of subjective ratings and heart rate) during exposure in comparison to
the other conditions.
Most of the evidence reviewed thus far suggests that distraction has the potential to
hinder exposure treatment and to promote the return of fear. However, Page and his co-workers
have provided compelling evidence that at least some forms of distraction during exposure may
actually facilitate both short- and long-term reductions in fear and anxiety. In a first study,
Penfold and Page (1999) examined whether manipulating attentional focus during in vivo
exposure influenced anxiety reduction among individuals with strong blood and injection-related
fears. Participants completed three weekly exposure sessions (duration =10 minutes) under one
of three experimental conditions: focused exposure, distracted exposure, or exposure alone.
Participants in the distracted exposure condition were engaged in neutral (stimulus-irrelevant)
conversation with the therapist (e.g., regarding plans for the future, hobbies, etc.), while those in
the focused exposure condition were asked to discuss their thoughts, feelings, and physiological
reactions to the feared stimuli. Visual attention was directed towards fearful stimuli regardless
of condition. It was argued that stimulus-irrelevant conversation represented a more ecologically
valid form of distraction than the more artificial distractors (e.g., video games, demanding
cognitive tasks) that had been used in most prior studies. Results showed that exposure-plus-
distraction led to greater within-session anxiety reduction than focused exposure or exposure
alone. However, no group differences were found on a behavioral approach test (BAT)
immediately following exposure.
In a second study, Oliver and Page (2003) sought to replicate and extend this finding by
examining both the short- and long-term effects of manipulating focus of attention during
exposure. This study used an identical procedure and participant population (i.e., blood- and
injection-fearful individuals) as Penfold and Pages (1999) prior investigation. In line with
emotional processing theory, it was hypothesized that distracted exposure may facilitate within-
session decreases in anxiety, but hinder long-term improvement by interfering with the activation
of fear structures in memory. However, contrary to expectation, participants in the exposure-
plus-distraction condition reported the greatest amount of fear reduction both within and between
sessions. Furthermore, this advantage was maintained at post-treatment and 1-month follow-up,
suggesting that conversational distraction during exposure may facilitate both short- and long-
term fear reduction compared to focused and natural forms of exposure.
A third investigation (J ohnstone and Page, 2004) extended this line of work to include an
examination of exposure-driven fear reduction among spider phobics. Participants were
randomly assigned to either distracted or focused exposure conditions. As in the previous
studies, participants were instructed to maintain visual contact with the feared stimulus (a live
Black House Spider) throughout the exposures, and attention towards the phobic stimulus was
manipulated via conversational tactics. However, the exposure schedule was slightly altered in
this investigation, such that participants completed three successive exposures (duration =10-
minutes) during their first experimental trial, and an additional exposure session four weeks later.
Interestingly, it was found that both subjective and physiological indices of fear did not differ
between groups during the initial moments of exposure. Also, neither group demonstrated a
significant return of fear, suggesting that both distracted and focused variants of exposure
treatment are effective in reducing fear of spiders. Nonetheless, consistent with Oliver and
Pages (2003) findings, participants in the distracted exposure condition reported significantly
greater within- and between-session anxiety reduction than participants in the focused exposure
condition, and these results were maintained at 1-month follow-up. Notably, increases in
perceived control were also observed among participants in the distracted exposure group in both
of these investigations (J ohnstone & Page, 2004; Oliver and Page, 2003).
Lastly, another group of researchers (Schmid-Leuz et al., 2007) recently compared the
effects of distracted versus attention-focused exposure among a group of dental phobics.
Participants were exposed to a series of fear-eliciting dental tools (i.e., dental probe, drill, needle
and pliers) according to an idiographically-designed hierarchy. Acoustic and olfactory stimuli
that mimicked those commonly experienced during dental procedures were utilized throughout
the exposure, which lasted 60 minutes for all participants. In the attention-focused exposure
condition, participants were instructed to contemplate the function of each tool as they
manipulated it manually. In contrast, participants in the distracted condition held the instruments
in their non-dominant hand while they played puzzle games with the experimenter. Several fear
indices (e.g., self-report fear and anxiety ratings, heart rate, state and trait anxiety measures)
were collected before and immediately following exposure, as well as at 1-week follow-up.
Contrary to expectation, there were no significant group differences in fear reduction (as
measured by heart rate and self-report ratings) at post-treatment and 1-week follow-up.
Likewise, there were no group differences in avoidance of subsequent dental treatments in the
six months immediately following the study. In fact, both groups demonstrated significant and
lasting improvement in phobic symptoms (as measured by self-report anxiety ratings), and the
only significant group difference that emerged revealed a slight advantage for the focused
exposure condition in terms of state anxiety ratings. However, an examination of group means
suggests that this group difference was slight, and is unlikely to be clinically meaningful.
Safety Behavior
Salkovskis and co-workers (1996) demonstrated that Panic Disorder patients choice of
safety behaviors is logically related to their catastrophic beliefs (e.g., an individual who believes
they are susceptible to experiencing a heart attack might lay down and raise their feet during
panic episodes). The first study to examine whether safety behaviors act to maintain anxiety and
fear-related beliefs was conducted by the Oxford Group (see Wells et al., 1995). In this study,
socially anxious individuals completed two exposure sessions (duration =5-10 minutes each) in
which they encountered a situation that they had identified as being highly fearful. They were
given different instructions prior to each exposure, and the order of conditions was counter-
balanced. Prior to one exposure session, a cognitive rationale was used to instruct participants
to drop their usual safety behaviors. Prior to the other exposure session, participants were not
given any specific instructions regarding the use of safety behaviors, and instead were provided
with a habituation rationale to explain the mechanisms of fear reduction during exposure. As
predicted, decreased safety behavior under cognitive rationale was associated with greater
reductions in participants subjective anxiety ratings and anxiety-related beliefs (within-session)
than natural exposure under habituation rationale.
Salkovskis, Clark, Hackmann, Wells, and Gelder (1999) found similar results in a study
of individuals suffering with Panic Disorder with Agoraphobia (PDA). Participants in this study
performed an idiographically-designed behavioral approach test (BAT) both prior to, and within
two days of completing a single, brief exposure session (duration =15 minutes). During this
exposure, half of the participants were instructed to drop their safety behaviours and were given
a cognitive rationale, while the other half were told to behave as they normally would in the
feared situation, and were provided with a habituation and extinction rationale. Again,
individuals who decreased their use of safety behavior during exposure reported significantly
less subjective anxiety and catastrophic beliefs during the follow-up BAT than individuals in the
control group.
Similar findings have emerged from several other investigations. For example, Morgan
and Raffle (1999) found that specific instructions to diminish the use of safety behavior during
exposure significantly increased the effectiveness of standard group CBT for Social Phobia. In
addition, two studies recently conducted at the University of Texas suggest that safety-seeking
behavior may be an important factor in the maintenance of claustrophobic fear. In the first study,
Sloan and Telch (2002) compared the outcomes of exposure with safety behavior utilization
(SBU) vs. exposure with guided threat reappraisal (GTR) and exposure only (EO) in a sample of
claustrophobic undergraduate students. Participants were randomly assigned to complete six
brief (5-minute) exposures in a claustrophobic chamber under one of the three conditions
described above. Although participants in the SBU condition were made aware of the
availability of safety aids (e.g., small window in chamber, intercom, etc.), they were not
specifically instructed to use them. Meanwhile, participants in the GTR condition were
instructed to test their catastrophic beliefs while in the chamber, and participants in the EO
condition were given no instruction. Results revealed a general advantage for the GTR
condition, as participants in the SBU group reported the highest ratings of peak fear during BATs
at post-treatment and 2-week follow-up, while participants fear ratings in the EO group fell in
between those of the other two groups. Furthermore, participants in the SBU condition exhibited
significantly less clinical improvement and between-trial habituation than those in the GTR
condition, whose advantage also generalized (although to a lesser extent) to a second BAT.
Two additional findings from this study deserve mention. First, contrary to expectation,
participants heart rate reactivity during the first 5 minutes of treatment did not differ between
groups, suggesting that levels of initial fear activation did not affect subsequent fear reduction.
Furthermore, all participant groups exhibited mean heart rates in the normal (albeit high normal)
range for adults at rest during BAT tests at pre-treatment (i.e., all group means <97 bpm).
Although Foa and Kozak (1986) do not offer specific guidelines regarding the degree of
physiological arousal (e.g., heart rate bpm) required to facilitate fear extinction following
exposure, the amount of arousal exhibited by these participants does not appear to have been
sufficient to fully activate fear structures in memory. Nevertheless, all groups demonstrated
significant reductions in heart rate reactivity and peak fear from pre- to post-treatment. This
finding appears to contradict emotional processing theory. Second, a significant number of
participants in the SBU condition reported that they did not actually use any safety behavior
during exposure. Accordingly, Sloan and Telch (2002) suggested that perhaps it is not safety
behavior utilization per se that inhibits fear reduction during exposure, but rather, the availability
of safety aids that causes this detriment. A second study conducted by this research team
(Powers, Smits & Telch, 2004) attempted to address this theoretically important question. A
large sample of undergraduates reporting severe claustrophobic fears were randomly assigned to
one of five conditions: (i) exposure with safety behavior utilization (SBU), (ii) exposure with
safety behavior availability (SBA), (iii) exposure only (EO), (iv) credible placebo treatment
(PL), or (v) wait list control (WL). Results indicated that the PL and WL conditions were the
least effective in reducing fear, and that approximately twice as many participants in the EO
condition achieved high end-state functioning at post-treatment and 2-week follow-up as those in
the SBU and SBA conditions. Importantly, individuals in the SBU and SBA conditions showed
equally poor rates of improvement, suggesting that mere availability of safety aids during
exposure is sufficient to hinder fear reduction.
Finally, Kim (2005) recently explored the specific mechanisms involved in reducing fear
via exposure with decreased safety behavior. Kim noted that the Oxford Group studies cited
earlier (Salkovskis et al., 1999, Wells et al., 1995) were confounded by the fact that participants
in each group were given a different rationale prior to exposure. In both studies, participants
who were instructed to stop using safety behavior were also provided with a cognitive rationale
for completing exposure treatment. In contrast, a habituation rationale was provided to
participants in the control conditions. As such, the precise mechanisms by which fear reduction
had been enhanced in the decreased safety behavior groups could not be determined from these
studies. It was unclear whether the benefits of reducing safety behavior in these studies were due
solely to a decrease in the behavior, or whether a cognitive rationale (with an emphasis on the
disconfirmation of negative beliefs) was also required to achieve these benefits. To address this
issue, Kim randomly assigned socially phobic individuals to one of three groups: (i) exposure
with decreased safety behavior under habituation rationale, (ii) exposure with decreased safety
behavior under cognitive rationale, or (iii) exposure only. It was predicted that decreased safety
behavior under cognitive rationale would produce the greatest reduction in anxiety and
catastrophic beliefs, and this is indeed what was found. Thus, Kim concluded that an emphasis
on disconfirming negative beliefs is crucial in reducing fear via exposure with decreased safety
behavior.
Contrary to the above-reviewed findings, a number of often-neglected studies suggest
that the use of safety-seeking strategies during exposure may not be universally detrimental. In
fact, it has recently been suggested that the judicious use of safety behavior may be entirely
appropriate under certain circumstances, particularly during the early phases of graded exposure
treatment (Rachman, Radomsky & Shafran, 2008).
Bandura, J effrey and Wright (1974) provided some early evidence for this view in a
study which examined the effects of providing safety aids to snake phobics during exposure. In
their study, participants were offered minimal, moderate or high use of response induction aids
(e.g., gloves) when they were unable to engage in an exposure exercise even after it was modeled
to them by the therapist-experimenter. This study found that participants who relied on moderate
or high levels of what would today be labeled safety behaviors experienced marked and
significantly better improvement (fear reduction) than those who were offered only mild levels
of these aids. Given the date of this important experiment, it is not surprising that outcome was
reported in terms of self-report and approach behavior. As such, current hypotheses about safety
behavior preventing the disconfirmation of maladaptive negative beliefs were not assessed. A
more recent study (Milosevic, 2006; Milosevic & Radomsky, in press) sought to address this
issue by randomly assigning snake fearful participants to either a treatment as usual (exposure)
condition or an exposure-plus-safety-gear condition in which participants could select one or
more safety aids (e.g., gloves, goggles, protective clothing) for use during 45 minutes of
exposure-based treatment. Results indicated that both groups experienced significant and nearly
identical treatment gains (measured through post-treatment approach behavior, self-reported
anxiety and negative cognitions/beliefs in the absence of safety gear), indicating that safety gear
neither interfered with outcome nor with disconfirmatory experiences (Milosevic, 2006;
Milosevic & Radomsky, in press). It is important to note that during the first part of the
exposure session, participants who used safety gear were able to get significantly closer to the
snake than those in the control group. Likewise, a larger number of participants in the safety
gear group were able to touch and/or hold the snake compared to those in the control group,
suggesting that the judicious use of safety behaviour (Rachman, Radomsky & Shafran, 2008)
may present advantages over traditional exposure-based treatments.
Additionally, Rachman and colleagues demonstrated that escape behavior (a form of
safety-seeking) does not always hinder treatment of agoraphobic avoidance in a set of innovative
studies designed to test the validity of the golden rule of exposure (i.e., in order to prevent fear
sensitization try never to leave a situation until the fear is going down; Mathews, Gelder &
J ohnston, 1981, p.182). In the first of these two studies (de Silva & Rachman, 1984), individuals
with Agoraphobia performed eight weekly exposure sessions with instructions to either: (i) stay
in the feared situation until their peak fear had declined by at least 50% (anxiety endurance
condition), or (ii) withdraw from the situation if their anxiety reached 75% of the highest level
they could imagine (escape condition). Measures of self- and clinician-rated anxiety as well as
agoraphobic avoidance (as indicated by a BAT test) were taken at pre- and post-treatment.
Notably, both groups showed significant pre- to post-treatment reductions in fear and
agoraphobic avoidance compared to a wait-list control group, and contrary to the golden rule
of exposure, individuals in the anxiety endurance condition did not exhibit greater improvement
than those in the escape condition. A replication of this study (Rachman, Craske, Tallman &
Solyom, 1986) found similar results, and most importantly, demonstrated that treatment gains
were maintained among both endurers and escapers at a 3-month follow-up assessment.
Consistent with Banduras theory, fear reductions were accompanied by increased control
ratings, especially among participants who were allowed to escape the situation. Together, these
studies provide compelling evidence that the judicious use of anxiety-control strategies may not
be detrimental under all circumstances, and prompt a reconsideration of whether it is necessary
for clients to endure prolonged high levels of anxiety/distress to benefit from exposure therapy.
Neutralization
Neutralization behavior, which is generally associated with OCD, has been defined as an
attempt to put matters right or undo the potential negative consequences of ones thoughts
and/or actions that the individual perceives as dangerous (e.g., repeating positive phrases or
prayers following blasphemous thoughts to cancel out the thoughts and prevent divine
retribution) (Rachman, 1976b; Rachman, Shafran, Mitchell, Trant & Teachman, 1996). A
decade ago, Rachman and colleagues (Rachman et al., 1996) set out to validate an experimental
method for studying neutralization, in a study that was designed to test the hypothesis that
neutralization resembles overt compulsions. Based on this hypothesis, it was predicted that
neutralization of intrusive thoughts would lead to an immediate decrease in anxiety, and that
preventing neutralization would result in a gradual decay of anxiety and urges to neutralize. A
group of non-clinical volunteers who demonstrated thought-action fusion (TAF) were selected to
participate in this study. TAF refers to the belief that having unwanted, immoral thoughts: (i)
might increase the likelihood that negative events will occur (i.e., Likelihood TAF) and/or (ii)
is morally equivalent to performing the inappropriate imagined actions (i.e., Moral TAF)
(Shafran, Thordarson & Rachman, 1996). Participants were asked to imagine a close friend or
relative and then to insert their name into the following sentence: I hope _____ is in a car
accident. Next, half of the participants were told that they could do whatever they wished (for 2
minutes) to undo (i.e., neutralize) the effects of the sentence, while the other participants were
instructed to do nothing to neutralize their thoughts during the next twenty minutes (although
they were allowed to read a magazine). The variables of interest were then measured and
instructions were reversed for each group. Thus, a final assessment was conducted twenty-two
minutes after the initial anxiety provocation. The authors predictions were largely supported,
and it was concluded that neutralization behavior is likely to be counter-productive in the long
run.
A group of researchers from the Netherlands have since conducted two investigations
using a slight variation of this protocol. The first of these studies (Hout, Pol, & Peters, 2001)
aimed to replicate Rachman et al.s (1996) findings. However, measures of anxiety and urges to
neutralize were taken at equal points in time for both experimental groups, unlike Rachman and
colleagues study. Hout and colleagues found that participants in both the neutralization and the
neutralization prevention groups reported a significant decrease in anxiety and urges to check
after only two minutes. Given that spontaneous decay of compulsive urges takes much longer to
achieve (Rachman, de Silva, & Roper, 1976), the authors questioned the assumption that
neutralization is functionally equivalent to overt compulsions. The second study conducted by
this group (Hout, Kindt, Weiland, & Peters, 2003) used a similar protocol, however, participants
were instructed to either (i) neutralize, (ii) perform a cognitive distractor task (mental
arithmetic), or (iii) they were given no particular instructions following the anxiety provocation.
Interestingly, participants in the no instruction group reported neutralizing their thoughts as
much as participants in the instructed neutralization group, suggesting that even non-clinical
individuals may spontaneously neutralize. Consistent with their previous results, it was found
that all three groups demonstrated equal reductions in anxiety from initial provocation to the 2-
minute assessment point. Furthermore, there were no group differences in subjective anxiety
when the anxiety-provoking thought was later re-introduced into consciousness. Thus, it was
concluded that reductions in anxiety following unpleasant thoughts are not likely due to the
effects of neutralization, but instead might result from other processes such as rationalization.
In a study aimed at uncovering factors involved in the etiology of OCD, Salkovskis,
Westbrook, Davis, J eavons and Gledhill (1997) examined the effects of neutralization on
individuals subsequent discomfort and urges to neutralize. In this experiment, non-clinical
participants recorded a 20-second narrative describing their most frequently experienced and
highly repugnant obsession onto a looped tape. Next, they listened to this intrusive thought
repeatedly during two exposure sessions. During the first session, half of the participants were
instructed to neutralize their intrusion, while the other half of participants were instructed to
count backwards, in order to control for any maladaptive effects that may be associated with
distraction. During the second session, all participants refrained from neutralizing or counting.
As expected, participants who neutralized their obsession reported greater decreases in
discomfort during the first phase of the experiment. However, they also reported greater
discomfort and urges to neutralize during the second exposure than individuals who had
previously used distraction. A replication of this study was recently carried out in a clinical
OCD sample, and similar results were found (Salkovskis, Thorpe, Wahl, Wroe, & Forrester,
2003). Taken together, these results suggest that neutralization of intrusive thoughts is likely to
be counter-productive in the long run, and may contribute to the development of clinical
obsessions and compulsions.
Summary of Findings
Fear Reduction
Contrary to predictions set forth by habituation/extinction models of fear reduction, a
large number of studies examining the effects of attentional focus during exposure (Craske et al.,
1991; Grayson et al., 1982, 1986; J ohnstone & Page, 2004; Oliver & Page, 2003; Penfold &
Page, 1999; Schmid-Leuz et al., 2007) have failed to demonstrate negative short-term effects in
association with distraction. In fact, several studies have shown that in comparison to attention
focusing (Craske et al., 1991; Grayson et al., 1986; J ohnstone & Page, 2004; Oliver & Page,
2003; Penfold & Page, 1999) and exposure alone conditions (Oliver & Page, 2003; Penfold &
Page, 1999), distraction may actually facilitate within-session fear reduction. Likewise, several
studies have shown that the judicious use of safety aids does not interfere with exposure
treatment (de Silva & Rachman, 1984; Rachman et al., 1986; Milosevic, 2006; Milosevic &
Radomsky, in press), and may actually improve treatment outcomes under certain circumstances
(Bandura et al., 1974).
On the other hand, recent studies have also suggested that cognitive load during exposure
is associated with less within-session fear reduction than threat reappraisal (Kamphius and Telch,
2000) and exposure only conditions (Telch et al., 2004). Moreover, it has been found that the
availability and/or use of safety behavior during exposure inhibits short-term fear reduction in
comparison to threat reappraisal (Sloan and Telch, 2002) and exposure only (Powers et al., 2004)
conditions.
Return of Fear
The group of studies reviewed above that assessed return of fear following exposure
reported mixed results. Four of these studies (Grayson et al., 1982; Kamphius & Telch, 2000;
Powers et al., 2004; Sloan and Telch, 2002) reported findings consistent with habituation-based
and emotional processing accounts. Grayson et al. (1982) found that distraction during exposure
significantly hampered between-session habituation, while Kamphius and Telch (2000) reported
a trend towards greater return of fear in distracted vs. natural and focused (i.e., threat reappraisal)
exposure conditions. Likewise, Sloan and Telch (2002) found a significantly greater return of
fear (at 2-week follow-up) among participants who were allowed to use safety behavior during
exposure compared to individuals who were instructed to test their catastrophic cognitions.
Finally, Powers et al. (2004) demonstrated that a mere availability of safety behaviors during
exposure was sufficient to compromise between-session habituation in comparison to natural
exposure.
In contrast, Page and co-workers (J ohnstone & Page, 2004; Oliver & Page, 2003)
findings appear to be more consistent with Banduras (1977) self-efficacy model. In conjunction
with significantly greater increases in perceived control, participants who were mildly distracted
during exposure exhibited less return of fear than participants in focused (J ohnstone & Page,
2004; Oliver & Page, 2003) and exposure only (Oliver & Page, 2003) conditions. Similarly,
Rachman and colleagues (1986) found that providing clients with the option to escape during
exposure to agoraphobic situations led to higher ratings of control relative to clients who were
instructed to endure their anxiety, and did not lead to a subsequent return of fear.
Subjective Anxiety
Consistent with cognitive theory, a number of investigations have demonstrated that
neutralization of intrusive thoughts (Hout et al., 2001, 2003; Rachman et al., 1996; Salkovskis et
al., 1997, 2003) and distraction (Schmid-Leuz et al., 2007) are associated with short-term
reductions in subjective anxiety. It has also been shown that comparable reductions in self-
reported anxiety can be achieved within two minutes of anxiety provocation when neutralization
is prevented (Hout et al., 2001, 2003). This finding suggests that other processes such as
rationalization may be involved in alleviating anxiety provoked by distressing thoughts (Hout et
al., 2003). Furthermore, Hout et al. (2003) found that anxiety experienced upon re-exposure to
intrusive thoughts did not differ between individuals who had previously neutralized their
thoughts and those who did not.
In contrast, the Oxford Group found that neutralization was associated with the
maintenance of discomfort and urges to neutralize in studies of both non-clinical (Salkovskis et
al., 1997) and clinical (Salkovskis et al., 2003) individuals. In addition, safety behavior use
during exposure to anxiety-provoking situations has been shown to inhibit anxiety reduction
immediately following the exposure(Kim, 2005; Salkovskis et al, 1999; Wells et al., 1995).
Catastrophic Beliefs
Similar to subjective anxiety, probability estimates regarding the likelihood of feared
consequences were found to immediately decrease following neutralization (Rachman et al.,
1996). Although follow-up data were not collected in this study, two related investigations
(Salkovskis et al., 1997, 2003) found that individuals who neutralized distressing thoughts
exhibited subsequent increases in discomfort and urges to neutralize. Based on Salkovskis et
al.s (1996) observation that anxiety-neutralizing behavior is motivated by catastrophic beliefs, it
is possible that the return of discomfort and urges to neutralize among these individuals resulted
from a re-emergence (or re-activation) of catastrophic beliefs. Lastly, three additional studies
that included explicit measures of catastrophic belief (Kim, 2005; Salkovskis et al, 1999; Wells
et al., 1995) found that a reliance on safety behavior during exposure to anxiety-provoking
situations was associated with greater belief in the likelihood of feared consequences. However,
it has also been shown that the use of safety behavior during exposure does not necessarily
prevent the disconfirmation of catastrophic beliefs (Milosevic & Radomsky, in press).
Discussion
The purpose of this paper was to assess whether there are circumstances in which
anxiety-neutralizing behavior may facilitate exposure and treatment of anxiety disorders.
Toward this aim, a number of empirical studies investigating potential moderators of exposure-
based treatment (i.e., distraction, safety behavior, neutralization) have been reviewed. While
several factors limit our ability to make solid inferences based on the available evidence, some
tentative conclusions seem justified.
Overall, the studies reviewed above support the increasingly accepted view that anxiety-
neutralizing strategies have the potential to become counter-productive by promoting
misattributions of safety, undermining self-efficacy and/or interfering with other possible
mechanisms of fear reduction during exposure (e.g., emotional processing). This was especially
apparent among studies which examined the disruptive effects of neutralization (Salkovskis et
al., 1997, 2003), certain types of safety-seeking behavior (Kim, 2005; Morgan & Raffle, 1999;
Powers et al., 2004; Salkovskis et al., 1999; Sloan & Telch, 2002; Wells et al., 1995) and heavy
distraction during exposure (Grayson et al., 1982; Kamphius & Telch, 2000; Telch et al., 2004).
Although Hout et al. (2003) found that participants in their neutralization condition did not
report greater increases in anxiety upon re-exposure to the thought than those who were
instructed to do nothing, minimal efforts to neutralize the thought were reported by participants
in both conditions (i.e., mean ratings of 22 [SD =28] vs. 10 [SD =22] out of 100, respectively).
As such, the clinical relevance of this finding must be called into question. Thus, the majority of
evidence provided by these studies indicates that anxiety-control strategies commonly used by
phobic clients may be detrimental in the long run.
On the other hand, there is also evidence to suggest that clients anxiety-control strategies
are not inevitably detrimental to exposure treatment (Bandura et al., 1974; deSilva & Rachman,
1984; Rachman et al., 1986; Schmid-Leuz et al., 2007; Milosevic & Radomsky, in press). In
fact, it appears that certain types of subtle avoidance might actually help anxious individuals to
achieve greater success in exposure-based treatments. More specifically, stimulus-irrelevant
conversation during exposure has been shown to produce anxiolytic effects which are both
durable and robust (J ohnstone & Page, 2004; Oliver & Page, 2003), unlike some other forms of
distraction (e.g., demanding cognitive tasks) which have been shown to hinder treatment
(Kamphius & Telch, 2000; Telch et al., 2004). Therefore, it is important to clarify what precise
mechanisms distinguish conversation from other types of distraction which serve to maintain fear
and anxiety. A detailed inspection of the studies reviewed above offers some preliminary clues
to this distinction.
First, both Oliver and Page (2003) and J ohnstone and Page (2004) found that participants
who engaged in stimulus-irrelevant conversation during exposure reported subsequent increases
in perceived control. As predicted by social learning theory (Bandura, 1977), this increase in
perceived control was associated with reduced fear responding in the presence of anxiety-
provoking stimuli. Although direction of causality cannot be inferred from these investigations,
this finding lends support to the theory that perceived mastery over feared situations is an
important moderator of fear reduction (Bandura, 1977; see also Powers, Smits, Whitley,
Bystritsky & Telch, in press). However, a comparison of changes in perceived control following
helpful vs. harmful forms of distraction is not currently possible, as this construct was not
measured in those studies that found distraction to be counter-productive. Nevertheless, it is
conceivable that stimulus-irrelevant conversation during exposure contributed to subsequent
increases in perceived control and self-efficacy through a variety of possible means. For
example, the pleasant topics of conversation used to distract participants in the aforementioned
studies (e.g., discussion of hobbies, travel plans, etc.) may have helped to promote a state of
relaxation during exposure. According to Bandura, relaxation in the presence of feared stimuli
can greatly affect perceptions of self-efficacy, as emotional arousal is hypothesized to be an
important source of information regarding ones ability to cope. Similarly, Wolpes (1954,
1968) theory of reciprocal inhibition states that inducing relaxation during exposure to phobic
stimuli should facilitate fear extinction via counter-conditioning processes. Although relaxation
(e.g., progressive muscle relaxation [PMR]) has not been found to be an essential component of
exposure treatment (McNally, 2007), it is plausible to hypothesize that anxious individuals may
have felt empowered by their ability to relax during the exposure, resulting in adaptive belief
change (e.g., If I am able to relax in the presence of my worst fear and nothing terrible happens,
I must not be in danger after all) (J ohnstone & Page, 2004), and a re-interpretation of the
meaning of previous anxious symptoms (e.g., Perhaps my anxiety does not always signal real
danger). In contrast, other forms of distraction may be less conducive to relaxation and mastery
experiences. In fact, distractors that were used in some of the other studies reviewed above (e.g.,
video games, cognitive load tasks) may have exerted the opposite effect, giving rise to
hyperarousal, frustration, excitement or anticipation (Rodriguez & Craske, 1993). Thus, unlike
more intense distractors, calming conversation and other mild distractors might promote
increases in self-efficacy during exposure to feared stimuli through increased relaxation,
increased perceptions of control and/or belief change, which in turn, may facilitate fear
reduction. Furthermore, presuming that affect and physiological arousal experienced during
exposure influence fear-related cognitions and beliefs, moderate distractors such as the puzzle
game utilized by Schmid-Leuz and co-workers (2007) may fail to exert an effect on exposure-
driven fear reduction since they are less likely to promote a heightened state of anxiety or
relaxation.
Another factor that may have contributed to differences between helpful and disruptive
forms of distraction is the amount of cognitive demand placed on the individual by each type of
task. Presumably, stimulus-irrelevant conversation required less attentional resources than the
distractors employed in the other studies reviewed above (e.g., cognitive load and dual-process
tasks, puzzle and video games). In comparison to other forms of distraction, conversation may
have allowed greater integration of corrective information during exposure, thereby promoting
the disconfirmation of negative beliefs and enabling emotional processing to occur. Consistent
with this interpretation, Telch et al. (2004) have suggested that it is not distraction per se that
interferes with fear reduction, but the extent to which the distractor task makes attentional
resources less available for cognitive processing during exposure (p.230). However, both
Penfold and Page (1999) and Oliver and Page (2003) found that distracted exposure led to
greater fear reduction than exposure alone. Given that exposure-only conditions do not require
any additional cognitive resources, this finding appears somewhat counter-intuitive, and remains
to be fully explained.
One possible explanation for this finding is that treatment is facilitated by an optimal
level of attentional focus toward feared stimuli during exposure. According to this hypothesis,
one would expect that excessive focus on feared stimuli might increase perceptions of threat and
heighten anxiety to levels that undermine emotional processing or increase subsequent fear
predictions, while too much distraction might inhibit fear reduction by diminishing cognitive
resources available for emotional processing (see Antony, McCabe, Leeuw, Sano and Swinson,
2001; see also J ohnstone & Page, 2004, McNally, 2007; Rodriguez & Craske, 1993). Consistent
with this theory, J ohnstone & Page (2004) found that only individuals who were low on initial
anxiety benefited from focused exposure in their study. Therefore, it is possible that mild
distractors such as stimulus-irrelevant conversation may provide an optimal amount of
distraction from threatening stimuli, while helping to maintain a sufficient level of anxiety to
activate fear structures, thereby facilitating the integration of corrective information during
exposure. Similarly, mild distraction during exposure might create an optimal environment for
the consolidation of extinction learning (e.g., by stimulating the medial prefrontal cortex,
increasing NMDA receptor activity, etc.), unlike heavy forms of distraction, which may inhibit
these cortical processes. Of course, these hypotheses are purely speculative, and await empirical
validation.
In summary, clients anxiety-control strategies appear less likely to become counter-
productive when the following four conditions are met: (i) they promote increases in self-
efficacy (via relaxation, positive affect, belief change, and/or other means), (ii) they do not
demand excessive attentional resources, (iii) they enable greater approach behavior and
integration/consolidation of corrective information (via disconfirmatory experiences), and (iv)
they do not promote misattributions of safety. However, further research is required to
determine the validity of these postulates. Also, it remains to be established which of these
conditions are necessary and/or sufficient to prevent return of fear following exposure.
Importantly, these observations also highlight the fact that clients array of anxiety-control
strategies cannot be classified as helpful or disruptive solely on the basis of presentation (i.e.,
form). Rather, such classifications require a consideration of the individuals intention in
performing the behavior, the perceived function and consequences of the behavior, and the
context in which the act is carried out (Thwaites & Freeston, 2005).
Clinical Implications
A number of practical implications follow from the evidence reviewed in this article.
First of all, the extant literature does not support the notion that anxiety-control strategies are
always detrimental to exposure therapy, or that it is necessary to completely eliminate their use
in order to achieve positive treatment outcomes. In fact, non-clinical individuals routinely
employ anxiety-control strategies (e.g., distraction, superstitious acts, etc.) in a variety of
circumstances and without significant negative consequences, suggesting that normative use of
these strategies may serve an adaptive function under certain circumstances. Similarly, the
judicious use of anxiety-control strategies during the early stages of exposure may facilitate
treatment by better enabling clients to approach and/or attend to feared stimuli and to process
corrective information (Rachman et al., 2008). Such techniques are commonly employed in
clinical practice (e.g., when creating exposure hierarchies), and may be particularly effective if
they are framed within the context of behavioural experiments (i.e., hypothesis testing), or if they
are conceptualized as stepping stones en route to patients/clients mastery of feared situations.
That said, the potentially adverse effects of anxiety-control strategies on treatment
effectiveness should not be ignored. Rather, the evidence clearly indicates that clients should be
warned about the potential negative consequences of neutralization, and should be encouraged to
promptly discontinue their use of any anxiety-control strategies that are intended to prevent
feared events, lest they promote misattributions of safety (Salkovskis et al., 1996). Likewise,
clients should be instructed to stop utilizing anxiety-control strategies that are likely to foster
complete cognitive avoidance of feared stimuli, as such avoidance strategies may prevent
emotional processing and/or extinction learning. Accordingly, it is essential for clinicians to
conduct a detailed functional assessment of anxiety-control strategies when treating clinically
anxious individuals. This crucial step helps to ensure that therapists understand their
clients/patients intentions in performing these behaviours, as well as their explanation for why
they have successfully averted feared catastrophes in the past.
1

Future Directions
A major obstacle to further progress in this area is the current lack of operational
definitions for the primary constructs of interest. For example, descriptions of neutralization
have varied widely from attempts to put matters right (Rachman, 1976) to any voluntary,
effortful cognitive or behavioral act that is directed at removing, preventing, or attenuating a
thought or the associated discomfort (Freeston & Ladouceur, 1997). Similar ambiguity exists
for the concepts of safety-seeking behavior and distraction (see Rodriguez & Craske, 1993,
for a detailed discussion of this issue). Consequently, different studies have used different
methods to examine the effects of various coping strategies, making it extremely difficult to
compare findings across these investigations (Antony et al., 2001). Therefore, the first priority
for researchers in this area should be to establish widely-accepted operational definitions for
these constructs, in order to facilitate cross-study comparisons and to promote the generation of
new and testable theories. Furthermore, it is important that definitions of these constructs do not
include reference to their effects on exposure, lest they rely on circular reasoning. Instead, the
operational definitions for these anxiety-control strategies should focus primarily on the form
and function of these acts (Thwaites & Freeston, 2005).
In addition, future studies should place a greater emphasis on measuring the long-term
effects of anxiety-neutralizing behavior. There is a grave lack of longitudinal research in this
area, which is surprising, given the theorized importance of long-term maladaptive consequences
associated with anxiety-control strategies. Similarly, future investigations would benefit from
attempts to maximize the ecological validity of their experimental manipulations. In order to
translate research findings into practical advice for clients, it will be essential to study the effects
of anxiety-control strategies that are actually used by clinically anxious individuals. Lastly, the
development of standardized protocols for studying these behaviors would greatly facilitate the
comparison of results across studies (Rose & McGlynn, 1997).
Once these goals have been achieved, several other questions can be tackled. For
example, what are the primary mechanisms through which anxiety-neutralizing behaviors exert
their disruptive effects (e.g., interference with anxious arousal and/or emotional processing,
maintenance of low self-efficacy beliefs, misattributions of safety, etc.)? Likewise, what are the
cognitive and neurological mechanisms involved where anxiety-control strategies improve the
outcome of exposure treatment (e.g., increased self-efficacy, decreased harm expectancy,
activation of NMDA receptors)? Is it necessary to fully activate fear structures during
exposure (e.g., to induce anxious hyperarousal in clients) in order to modify catastrophic beliefs
and achieve lasting fear reduction? (How) does the use of anxiety-neutralizing behavior affect
subsequent fear predictions? Do clients moods influence their choice of anxiety-control strategy
and/or their interpretation of exposure treatment outcomes? Are anxiety-control strategies more
disruptive when used to counter hypothetical fears (such as those frequently encountered in OCD
and health anxiety)? Can universally adaptive coping strategies (e.g., rationalization?) be
identified and taught to clients? All of these questions are ripe for investigation, and progress in
each of these areas should help to bridge the gaps between theory, research and clinical practice.
References
Agras, W.S. (1965). An investigation of the decrement of anxiety responses during
systematic desensitization therapy. Behaviour Research and Therapy, 2, 267-270.
Antony, M.M., McCabe, R.E., Leeuw, I., Sano, N., & Swinson, R.P. (2001). Effect of
distraction and coping style on in vivo exposure for specific phobia of spiders.
Behaviour Research and Therapy, 39, 1137-1150.
Bandura, A. (1977). Self-efficacy: Toward a unifying theory of behavioural change.
Psychological Review, 84, 191-215.
Bandura, A., J effery, R. and Wright, C. (1974). Efficacy of participant modeling as
a function of response aids. Journal of Abnormal Psychology, 83, 56-64.
Battersby, M. (2000). Response to Does reducing safety behaviours improve treatment
response in patients with social phobia? Australian and New Zealand Journal of
Psychiatry, 34, 871-872.
Beck, A.T., Emery, G, & Greenberg, R.L. (1985). Anxiety disorders and phobias. New
York: Basic Books.
Borkovec, T.D., & Grayson, J .B. (1980). Consequences of increasing the functional
impact of internal emotional stimuli. In K. Blandstein, P. Pliner, & L.H. Polivcy (Eds.),
Advances in the study of communication and affects (pp. 117-137). New York: Plenum
Press.
Bouton, M.E. (2004). Context and behavioral processes in extinction. Learning and Memory,
11, 485-494.
Clark, D.M. (1999). Anxiety disorders: Why they persist and how to treat them.
Behaviour Research and Therapy, 37, S5-S27.
Craske, M.G., Street, L.L., & Barlow, D.H. (1989). Instructions to focus upon or distract
from internal cues during exposure treatment of agoraphobic avoidance. Behaviour
Research and Therapy, 27, 663-672.
Craske, M.G., Street, L.L., J ayaraman, J ., & Barlow, D.H. (1991). Attention versus
distraction during in vivo exposure: Snake and spider phobias. Journal of Anxiety
Disorders, 5, 199-211.
Davis, M. (2002). Role of NMDA receptors and MAP kinase in the amygdale in extinction of
fear: Clinical implications for exposure therapy. European Journal of Neuroscience, 16,
395-398.
de Silva, P., & Rachman, S. (1984). Does escape behaviour strengthen agoraphobic
avoidance? A preliminary study. Behaviour Research and Therapy, 22, 87-91.
Foa, E.B., Huppert, J .D., & Cahill, S.P. (2006). Emotional processing theory: An update. In
B.O. Rothbaum (Ed.), Pathological anxiety: Emotional processing in etiology and
treatment (pp. 3-24). New York: Guilford.
Foa, E.B., & Kozak, M.J . (1986). Emotional processing of fear: Exposure to
corrective information. Psychological Bulletin, 99, 20-35.
Freeston, M.H., & Ladouceur, R. (1997). What do patients do with their obsessive
thoughts? Behaviour Research and Therapy, 35, 335-348.
Grayson, J .B., Foa, E.B., & Steketee, G. (1982). Habituation during exposure treatment:
Distraction vs. attention-focusing. Behaviour Research and Therapy, 20, 323-
328.
Grayson, J .B., Foa, E.B., & Steketee, G. (1986). Exposure in vivo of obsessive-
compulsives under distracting and focusing conditions: Replication and extension.
Groves, P.M. & Thompson, R.F. (1970). Habituation: A dual-process theory.
Psychological Review, 77, 419-450.
Halstead, W.C. (1947). Brain and intelligence: A quantitative study of the frontal lobes.
Chicago: University of Chicago Press.
Hofmann, S.G. (2008). Cognitive processes during fear acquisition and extinction in animals
and humans: Implications for exposure therapy of anxiety disorders. Clinical Psychology
Review, 28, 199-210.
Hofmann, S.G., Meuret, A.E., Smits, J .A.J ., Simon, N.M., Pollack, M.H., Eisenmenger, K., et al.
(2006). Augmentation of exposure therapy for social anxiety disorder with d-cycloserine.
Archives of General Psychiatry, 63, 298-304.
Hout, M. van den, Kindt, M., Weilend, T., & Peters, M. (2002). Instructed neutralization,
spontaneous neutralization and prevented neutralization after an obsession-like thought.
Journal of Behavior Therapy and Experimental Psychiatry, 33, 177-189.
Hout, M. van den, Pol, M. van, & Peters, M. (2001). On becoming neutral: Effects of
experimental neutralizing reconsidered. Behaviour Research and Therapy, 1439-1448.
J ohnstone, K.A., & Page, A.C. (2004). Attention to phobic stimuli during exposure: The
effect of distraction on anxiety reduction, self-efficacy and perceived control. Behaviour
Kim, E. J . (2005). The effect of the decreased safety behaviours on anxiety and negative
thoughts in social phobics. Journal of Anxiety Disorders, 19, 69-86.
Lang, P.J . (1977). Imagery in therapy: An information processing analysis of fear. Behavior
Therapy, 8, 862-886.
Lang, P.J . (1984). Cognition in emotion: Concept and action. In C. Izard, J . Kagan, & R. Zajonc
(Eds.), Emotion, cognition and behavior (pp. 193-206). New York: Cambridge
University Press.
Oliver, N.S., & Page, A.C. (2003). Fear reduction during in vivo exposure to blood-
injection stimuli: Distraction vs. attentional focus. British Journal of Clinical
Psychology, 42, 13-25.
Mathews, A., Gelder, M., & J ohnston, D. (1981). Agoraphobia. London: Tavistock
Press.
McNally, R.J . (2007). Mechanisms of exposure therapy: How neuroscience can improve
psychological treatments for anxiety disorders. Clinical Psychology Review, 27, 750-
759.
Milad, M.R., & Quirk, G.J . (2002). Neurons in medial prefrontal cortex signal memory for fear
extinction. Nature, 420, 70-74.
Milosevic, I. (2006). Safety behaviour does not interfere with exposure therapy: The case
of specific phobia. Masters thesis, Concordia University.
Milosevic, I., & Radomsky, A.S. (in press). Safety behaviour does not necessarily interfere
with exposure therapy. Behaviour Research and Therapy.
Mowrer, O.H. (1939). Stimulus response theory of anxiety. Psychological Review, 46, 553-565.
Myers, K.M., & Davis, M. (2002). Behavioral and neural analysis of extinction. Neuron, 36,
567-584.
Powers, M.B., Smits, J .A.J ., & Telch, M.J . (2004). Disentangling the effects of safety-
behavior utilization and safety-behavior availability during exposure-based treatment: A
placebo-controlled trial. Journal of Consulting and Clinical Psychology, 72, 448-454.
Powers, M.B., Smits, J ., Whitley, D., Bystritsky, A., & Telch, M. (in press). The effect of
attributional processes concerning medication taking on return of fear. Journal of
Consulting and Clinical Psychology.
Purdon, C. (1999). Thought suppression and psychopathology. Behaviour Research
and Therapy, 37, 1029-1054.
Purdon, C. (2004). Empirical investigations of thought suppression in OCD. Journal of
Behavior Therapy and Experimental Psychiatry, 35, 121-136.
Purdon, C., Rowa, K., & Antony, M.M. (2005). Thought suppression and its effects on
thought frequency, appraisal and mood state in individuals with obsessive-compulsive
disorder. Behaviour Research and Therapy, 43, 93-108.
Rachman, S. (1976a). The passing of the two-stage theory of fear and avoidance: Fresh
possibilities. Behaviour Research and Therapy, 14, 125-131.
Rachman, S. (1976b). The modification of obsessions: A new formulation. Behaviour
Rachman, S. (1980). Emotional processing. Behaviour Research and Therapy, 18,
51-60.
Rachman, S. (2002). A cognitive theory of compulsive checking. Behaviour Research and
Therapy, 40, 625-639.
Rachman, S., & Bichard, S. (1988). The overprediction of fear. Clinical Psychology
Review, 8, 303-312.
Rachman, S., Craske, M., Tallman, K., & Solyom, C. (1986). Does safety behavior
strengthen agoraphobic avoidance? A replication. Behavior Therapy, 17, 366-384.
Rachman, S., & Lopatka, C. (1986a). Match and mismatch in the prediction of fear I.
Rachman, S., & Lopatka, C. (1986b). Match and mismatch of fear in Grays theory II.
Rachman, S., de Silva, P., & Roper, G. (1976). The spontaneous decay of compulsive urges.
Rachman, S., Radomsky, A.S., & Shafran, R. (2008). Safety behaviour: A reconsideration.
Rachman, S., Shafran, R., Mitchell, D., Trant, J ., & Teachman, B. (1996). How to
remain neutral: An experimental analysis of neutralization. Behaviour Research and
Therapy, 34, 889-898.
Rassin, E., Merckelbach, H., & Muris, P. (2000). Paradoxical and less paradoxical
effects of thought suppression: A critical review. Clinical Psychology Review, 20, 973-
995.
Rescorla, R.A. (1996). Preservation of Pavlovian associations through extinction. Quarterly
Journal of Experimental Psychology, 49, 245-258.
Ressler, K.J ., Rothbaum, B.O., Tannenbaum, L., Anderson, P., Graap, K., Zimand, E., et al.
(2004). Cognitive enhancers as adjuncts to psychotherapy: Use of d-cycloserine in
phobics to facilitate extinction of fear. Archives of General Psychiatry, 61, 1136-1144.
Rodriguez, B.I., & Craske, M.G. (1993). The effects of distraction during exposure to
phobic stimuli. Behaviour Research and Therapy, 549-558.
Rose, M.P., & McGlynn, F.D. (1997). Toward a standard experiment for studying post-
treatment return of fear. Journal of Anxiety Disorders, 11, 263-277.
Salkovskis, P.M. (1988). Phenomenology, assessment and the cognitive model of panic.
In S.J . Rachman, & J . Maser, Panic: psychological perspectives. New J ersey: Erlbaum.
Salkovskis, P.M. (1991). The importance of behaviour in the maintenance of anxiety and
panic: a cognitive account. Behavioural Psychotherapy, 19, 6-19.
Salkovskis, P.M. (1996). The cognitive approach to anxiety: Threat beliefs, safety-
seeking behaviour, and the special case of health anxiety and obsessions. In Salkovskis,
P.M. (Ed.), Frontiers of cognitive therapy (pp. 48-74). New York: Guilford Press.
Salkovskis, P.M. (1999). Understanding and treating obsessive-compulsive disorder. Behaviour
Research and Therapy, 37, S29-S52.
Salkovskis, P.M., Clark, D.M., & Gelder, M.G. (1996). Cognition-behaviour links in the
persistence of panic. Behaviour Research and Therapy, 34, 453-458.
Salkovskis, P.M., Clark, D.M., Hackmann, A., Wells, A., & Gelder, M.G. (1999). An
experimental investigation of the role of safety-seeking behaviours in the maintenance of
panic disorder with agoraphobia. Behaviour Research and Therapy, 37, 559-574.
Salkovskis, P.M., Thorpe, S.J ., Wahl, K., Wroe, A.L., & Forrester, E. (2003).
Neutralizing increases discomfort associated with obsessional thoughts: An experimental
study with obsessional patients. Journal of Abnormal Psychology, 112, 709-715.
Salkovskis, P.M., Westbrook, D., Davis, J ., J eavons, A., & Gledhill, A. (1997). Effects
of neutralizing on intrusive thoughts: An experiment investigating the etiology of
obsessive-compulsive disorder. Behaviour Research and Therapy, 35, 211-219.
Schmid-Leuz, B., Elsesser, K., Lohrmann, T., J hren, P., & Sartory, G. (2007). Attention
focusing versus distraction during exposure in dental phobia. Behaviour Research and
Therapy, 45, 2691-2703.
Shafran, R., Thordarson, D., & Rachman, S. (1996). Thought-action fusion in obsessive-
compulsive disorder. Journal of Anxiety Disorders, 10, 379-392.
Sloan, T., & Telch, M.J . (2002). The effects of safety-seeking behaviour and guided
threat reappraisal on fear reduction during exposure: An experimental investigation.
Telch, M.J ., Valentiner, D.P., Ilai, D., Young, P.R., Powers, M.B., Smits, J .A.J . (2004).
Fear activation and distraction during the emotional processing of claustrophobic fear.
Journal of Behavior Therapy and Experimental Psychiatry, 35, 219-232.
Thwaites, R., & Freeston, M.H. (2005). Safety-seeking behaviours: Fact or function?
How can we clinically differentiate between safety behaviours and adaptive coping
strategies across anxiety disorders? Behavioural and Cognitive Psychotherapy, 33, 1-12.
Walker, D.L., Ressler, K.J ., Lu, K.T., & Davis, M. (2002). Facilitation of conditioned fear
extinction by systematic administration or intra-amygdala infusions of d-cycloserine as
assessed with fear-potentiated startle in rats. Journal of Neuroscience, 22, 2343-2351.
Watts, F.N. (1971). Desensitization as an habituation phenomenon: I. Stimulus intensity
as a determinant of the effects of stimulus lengths. Behaviour Research and
Therapy, 9, 209-217.
Watts, F.N. (1974). The control of spontaneous recovery of anxiety in imaginal
desensitization. Behaviour Research and Therapy, 12, 57-59.
Watts, F.N., Trezise, L., & Sharrock, R. (1986). Processing of phobic stimuli.
British Journal of Clinical Psychology, 25, 253-259.
Wolpe, J . (1954). Reciprocal inhibition as the main basis of psychotherapeutic effects.
Archives of Neurology & Psychiatry, 72, 205-226.
Wolpe, J . (1968). Psychotherapy by reciprocal inhibition. Conditional Reflex, 3, 234-
240.
Author note
Chris L. Parrish, Department of Psychology, Concordia University; Adam S. Radomsky,
Department of Psychology, Concordia University; Michel J . Dugas, Department of Psychology,
Concordia University.
The authors are grateful to Norman Segalowitz and J ennifer Knight for their helpful
comments and suggestions on earlier drafts of this manuscript. We would also like to
acknowledge the valuable feedback and suggestions provided by the anonymous reviewers.
Correspondence concerning this article should be addressed to Adam S. Radomsky at:
Concordia University, Department of Psychology, 7141 Sherbrooke St. West, Montral, QC,
H4B 1R6, CANADA, phone: +1(514) 848-2424, ext.2202, fax: +1 (514) 848-4523, email:
[email protected].
Footnotes

1
Also, see Powers et al. (in press) for a detailed discussion of the importance of assessing
clients attributions of treatment success.

Anxiety-Control Strategies Is There Room For Neutralization in Successful Exposure Treatment

Uploaded by

Document Informationclick to expand document information

Copyright:

Available Formats

Anxiety-Control Strategies Is There Room For Neutralization in Successful Exposure Treatment

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Anxiety-Control Strategies Is There Room For Neutralization in Successful Exposure Treatment

Uploaded by

Copyright:

Available Formats

Anxiety-control strategies 1

Running Head: ANXIETY-CONTROL STRATEGIES

You might also like