Mechanisms of recovery
The findings from our studies reviewed so far may not be surprising to behavior therapists, who have long recognized the critical role of avoidance in anxiety disorders. Previous research has consistently shown that avoidance is often linked to overall illness severity in anxiety disorders (Başoğlu et al., 1988; Başoğlu et al., 1994b) and that reducing avoidance through exposure treatment leads to improvement in most patients with anxiety disorders, including PTSD (Marks & Dar, 2000). However, what warrants further explanation is the remarkable potency of an intervention that focuses exclusively on fear and behavioral avoidance. This becomes a particularly compelling question given its efficacy, which is comparable to that of other, more intensive and lengthy exposure-based treatments, even when delivered in just a single session.
Could this effectiveness be attributed to CFBT’s focus on enhancing a sense of control over fear, rather than relying on habituation? In other words, does prioritizing a sense of control yield better outcomes than focusing on habituation? The most direct way to address this question would be through a head-to-head comparison between CFBT and traditional exposure treatment. While such a study has not yet been conducted, several alternative approaches could provide valuable insights.
For instance, one could investigate whether an increase in a sense of control over fear can occur without a corresponding reduction in fear during exposure. If increases in sense of control and habituation are not inherently linked, demonstrating their relative independence would be highly informative. Additionally, researchers could examine the relative contributions of increased sense of control and habituation to overall improvement or a subjective sense of well-being as reported by patients. Such investigations would help clarify the mechanisms underlying CFBT’s effectiveness and shed light on whether its unique focus on control represents a more potent therapeutic approach than traditional habituation-based methods.
Anxiety reduction or increased sense of control?
Habituation, defined as the waning of a defensive response to repeated stimulation (Marks, 1987), is a universal phenomenon observed across species, from humans to invertebrates. For example, a snail retreats into its shell when initially touched but eventually ceases to respond after repeated stimulation. This concept has long been central to behavioral treatments for anxiety, particularly exposure therapy. Marks (1987) argued, based on the extinction model of anxiety reduction, that exposure should continue until significant anxiety reduction occurs. Similarly, emotional processing models of fear reduction (Foa & Kozak, 1986; Foa & McNally, 1996) propose that the initial fear response and the degree of fear reduction within and across exposure sessions are key indicators of successful emotional processing, which in turn determines treatment outcomes.
However, a review of the processes of change in exposure therapy (Craske et al., 2008) found no conclusive evidence linking treatment outcomes to the intensity of initial fear or the extent of fear reduction during and between sessions. The authors concluded that within-session habituation may be mediated by mechanisms distinct from those responsible for long-term therapeutic outcomes. This raises important questions about the role of habituation in exposure therapy and whether it is as central to treatment success as previously thought.
These findings challenge the traditional emphasis on habituation and suggest that other mechanisms, such as an enhanced sense of control over fear, may play a more critical role in achieving long-term improvements. This aligns with the principles underlying CFBT, which prioritizes fostering a sense of control over fear rather than focusing solely on fear reduction. By shifting the therapeutic focus from habituation to empowerment, CFBT may tap into more potent and enduring mechanisms of change, offering a promising alternative to traditional exposure-based approaches. This perspective not only reframes our understanding of how exposure therapy works but also opens new avenues for developing more effective and efficient interventions for trauma survivors.
The concept that habituation alone may not necessarily lead to an increased sense of control over fear is a critical insight. For habituation to translate into a lasting sense of control, exposure must be conducted as a motivation-driven effort aimed at achieving control over fear, with the individual attributing the success to their own efforts (e.g., “I did it all by myself without any help”) and generalizing this sense of control to future encounters with similar stressors (e.g., “I will be able to handle similar situations in the future”). Evidence from a study by Marks et al. (1993) on exposure and alprazolam treatment for panic disorder and agoraphobia strongly supports this idea.
In this study, patients received either alprazolam or placebo, combined with either exposure treatment or relaxation as a psychological placebo. Psychological treatments were terminated at week 8, and drug tapering occurred between weeks 8 and 16. Blindness to the drug condition was maintained until the end of the taper period. Patients who improved by week 8 completed a Tablet Attributions Questionnaire, which measured their treatment attributions, sense of control over their problem, and expectations of coping without treatment. This allowed the researchers to examine the effect of sense of control on long-term outcomes using a prospective design.
The findings revealed that patients who attributed their improvement to the medication were more likely to relapse during the withdrawal phase than those who attributed their improvement to their own efforts during exposure treatment (Başoğlu et al., 1994a). This occurred despite the fact that both groups had habituated to anxiety during treatment. This suggests that the sense of control, rather than habituation or the extinction of anxiety, is the critical factor determining long-term improvement. Interestingly, many patients expressed a strong desire to know whether they had been taking alprazolam or placebo, as this would clarify what had truly helped them. One patient, who had shown significant improvement, admitted during the tapering phase that she had been praying her tablets were not the “real thing.” Toward the end of the taper, some patients reported increasing their exposure to agoraphobic situations on their own, without therapist encouragement, to take greater responsibility for their improvement in case the tablets turned out to be the active drug. These findings and observations were pivotal in inspiring the paradigm shift in exposure treatment that led to the development of CFBT.
These findings are further corroborated by a later study (Powers et al., 2008) that examined the effects of attributional processes related to medication-taking on the return of fear following exposure treatment. In this study, 95 participants with claustrophobic fears were randomly assigned to one of several conditions: a single-session exposure-based treatment, exposure combined with a placebo pill, exposure combined with a psychological placebo, or a waitlist control. To manipulate attributions about the pill, participants in the exposure-plus-pill condition were further divided into three groups, each receiving a different instructional set immediately after treatment and post-treatment assessment. These instructions framed the pill as:
- A sedating herb that likely made exposure treatment easier,
- A stimulating herb that likely made exposure treatment more difficult, or
- A placebo with no effect on exposure treatment.
The results were striking: 39% of participants who were told the pill was a sedating herb (and thus likely made treatment easier) experienced a return of fear, compared to 0% in the other two conditions. This return of fear was mediated by participants attributing their anxiety reduction to the pill rather than to their own efforts. Additionally, the study found that low self-efficacy also played a mediating role in the return of fear.
These studies underscore the pivotal role of a sense of control in the recovery from fear and anxiety. They also reveal that fear reduction can occur independently of an increased sense of control, particularly when individuals attribute their improvement to external factors (e.g., medication) rather than to their own efforts. This suggests that the ultimate goal in the path to recovery from anxiety is not merely habituation or fear reduction but the cultivation of a sense of control over one’s fear.
This insight is further supported by a study involving patients with PTSD who were treated with exposure therapy, cognitive restructuring, or a combination of both (Livanou et al., 2002). The findings demonstrated that a sense of control played a critical role in maintaining long-term improvement, regardless of the specific treatment modality. Patients who developed a stronger sense of control over their symptoms were more likely to sustain their gains over time, highlighting the importance of this factor in the therapeutic process.
The relationship between sense of control and fear reduction is complex, and while Marks and Dar (2000) suggested that a sense of control might be a result rather than a cause of fear reduction, several lines of evidence challenge this view. First, extensive research in both animal and human studies demonstrates that manipulations of control (as an independent variable) are causally linked to fear. Reviews by Başoğlu and Mineka (1992) and Mineka and Zinbarg (2006) highlight that fear is often the outcome of a loss of control, rather than the reverse (see also Başoğlu et al., 2011 for further evidence). This suggests that control plays a foundational role in modulating fear responses.
Second, there is no conclusive evidence establishing a direct association between fear reduction and the outcome of exposure treatment (Craske et al., 2008). This implies that fear reduction alone may not be the primary driver of therapeutic success. Third, while sense of control and fear are correlated, they are also relatively independent phenomena. For example, individuals can experience intense fear without losing control. In Şalcıoğlu’s (2004) study of earthquake survivors, 76% reported high levels of fear during the earthquake, but only 40% experienced a total loss of control. Similarly, in studies of war and torture survivors, while greater loss of control was generally associated with more intense anxiety, the ratings of loss of control were typically lower than those of anxiety intensity. This indicates that high anxiety can occur without significant loss of control. For instance, resilient political activists, despite experiencing minimal loss of control, reported levels of anxiety comparable to less resilient survivors in response to certain forms of torture involving physical pain (Başoğlu et al., 1997).
The relative independence of sense of control and fear also means that fear reduction can occur without an increase in sense of control—such as when fear reduction is attributed to external factors or mediated through safety signals. Conversely, sense of control can increase without any reduction in fear. This was observed in some torture survivors who, despite showing no fear reduction during their first exposure session, reported a dramatic increase in their sense of control simply from confronting a long-standing, debilitating fear.
Further evidence supporting the independence of these constructs comes from studies showing that phobic patients can improve even when allowed to terminate exposure sessions before complete fear reduction occurs (Emmelkamp & Mersch, 1982; Rachman et al., 1986). Additionally, extending exposure trials beyond complete fear reduction does not yield better outcomes than partial fear reduction in a single trial (Rachman & Lopatka, 1988), and complete fear reduction does not necessarily prevent a return of fear (Rachman et al., 1987).
These findings collectively suggest that while fear reduction and sense of control are interrelated, they are not synonymous. A sense of control appears to be a critical, independent factor in recovery, often driving long-term improvement even in the absence of immediate fear reduction.
One of the most informative studies regarding the relationship between fear reduction and sense of control is the alprazolam/exposure treatment study of panic disorder and agoraphobia conducted by Marks and colleagues (1993), as reviewed earlier. In this study, we identified four distinct subgroups of patients with different treatment outcomes based on their improvement in panics and behavioral avoidance (Başoğlu et al., 1994b) :
- No improvement in panics or avoidance,
- Improvement in avoidance but not in panics,
- Improvement in panics but not in avoidance, and
- Improvement in both panics and avoidance .
The existence of the first and fourth subgroups demonstrates that fear and avoidance (an indicator of a lack of sense of control over fear) are closely interrelated in some cases. However, the second subgroup reveals that a sense of control can increase without a corresponding reduction in fear. [Prospective monitoring of daily anxiety levels and panics in the same study sample showed that most panics were triggered by prior elevated levels of general anxiety and anticipatory fear (Başoğlu et al., 1992). Thus, the term “fear” in this context refers not only to recurring panics but also to high anxiety levels and anticipatory fear between episodes.]
The third subgroup, on the other hand, demonstrates that fear reduction can occur without an increase in sense of control. Importantly, patients who improved in avoidance but not in panics were more likely to rate themselves as improved compared to those who improved in panics but not in avoidance. This suggests that improvement in avoidance—or an increased sense of control—holds greater significance for an individual’s subjective sense of well-being than fear reduction alone.
These findings imply that the critical factor driving recovery (when defined as a subjective sense of well-being) is an increased sense of control over the problem, rather than fear reduction. Fear reduction appears to be neither necessary nor sufficient for recovery. Instead, fostering a sense of control over fear and avoidance behaviors emerges as the central mechanism underlying lasting improvement.
Marks and Dar (2000), in their review of the limitations of the exposure principle, highlighted several shortcomings of the habituation model. They noted that habituation fails to explain why some phobic patients experience rapid improvement without reporting fear or avoidance during their first exposure session, or why fear can decline with interventions that do not involve exposure, such as hypnosis, reassuring information, and placebo tablets. In exploring the common therapeutic ingredients across various treatments—such as exposure, cognitive restructuring, coping skills training, problem-solving, stress immunization, and mindfulness meditation—they posed the question, “How many roads lead to the Rome of fear reduction?” Perhaps the questions they rightly raised about habituation might be more effectively addressed if we search for the roads to the Rome of increased sense of control, rather than habituation.
The findings from our treatment studies with earthquake survivors provide strong evidence supporting the central role of a sense of control in recovery from trauma. In Study 4, which combined self-exposure instructions with Earthquake Simulation Treatment, significant improvements in sense of control were observed, as measured by the Sense of Control Scale. This increase in control was strongly correlated with reductions in PTSD symptoms. In Study 2, where self-exposure instructions were provided alone, behavioral avoidance was the first symptom to improve early in treatment, reflecting an enhanced sense of control, followed by improvements in other PTSD symptoms, suggesting a causal link between the two. Study 3 demonstrated that exposure to simulated earthquake tremors, even without self-exposure instructions from the therapist, led to reduced behavioral avoidance, indicating that confronting fear-inducing stimuli can naturally boost a sense of control over anxiety. Across all four studies, the treatment not only reduced PTSD symptoms but also alleviated depression, consistent with the learning theory model of traumatic stress, which posits that depression is the outcome of helplessness and hopelessness. Additionally, the treatment appeared to enhance resilience against the traumatic effects of earthquakes, as evidenced by remarkably low relapse rates. Only three cases relapsed after significant improvement, even though some participants experienced further earthquakes after treatment. In Study 4, 11 out of 13 survivors who encountered another earthquake after treatment reported much less fear and loss of control during the tremors, reflecting a learned sense of control over stressors.
Another factor that may have contributed to an enhanced sense of control is the minimal involvement of the therapist in the treatment process. As previously noted, a single-session intervention with no further therapist contact means that the treatment is largely self-administered. This approach likely fosters a sense of control by encouraging individuals to attribute their improvement to their own efforts rather than to the therapist or external factors, such as changes in life circumstances. This aligns with evidence demonstrating that internal attributions of control help consolidate treatment effects and reduce the risk of relapse in trauma survivors (Livanou et al., 2002).
Noncompliance with exposure treatment is a well-documented challenge in behavioral therapy. Marks (2002) noted that noncompliance rates, including refusal of treatment and dropout, reached 30% among patients with anxiety disorders receiving therapist-delivered exposure treatment in outpatient settings. In contrast, our studies observed noncompliance rates of less than 10%, even when treatment was delivered in a single session without subsequent monitoring or reinforcement. This lower rate of noncompliance may reflect the motivational impact of the treatment rationale (i.e. fight your fear to take control over your life or surrender and live the rest of your life in misery and despair). This type of discourse, often used by political leaders, military commanders, or leaders of activist groups, has the power to instill courage, self-confidence, and hope, motivating individuals to take action against their fears. The effectiveness of such discourse is evident in its ability to inspire people to engage in high-risk activities, such as combat, political activism in repressive regimes, or even acts involving certain death, as seen with Kamikaze pilots in World War II or modern-day suicide bombers. These examples underscore that cognitive and behavioral control over fear is achievable even in the face of genuine threats to safety, highlighting the profound influence of motivational framing on human behavior.
The effectiveness of Earthquake Simulation Treatment in enhancing the impact of self-exposure instructions can be attributed to its strong impact on sense of control. This intervention involves exposure to unconditioned stimuli, such as simulated earthquake tremors, as well as distressing trauma memories. By reconstructing the original uncontrollable stressors experienced during a real earthquake—such as a moving physical environment, sounds of shifting objects, and the disorienting effects of tremors on the vestibular system, postural control, and proprioception—it creates opportunities for individuals to develop a sense of control over a wide range of stressor cues and the emotional responses they evoke. This type of exposure is likely to have a far more potent effect on sense of control than exposure to mere reminders of the trauma or conditioned fear cues, which may generalize to various objects, situations, or activities.
The impact of this intervention is evident in survivors’ responses to real earthquakes. While traumatized individuals often exhibit panic reactions during earthquakes—such as senselessly jumping out of windows even during mild tremors, resulting in injuries—treated survivors frequently report no loss of control, despite experiencing some reduced fear. This aligns with evidence (Mineka et al., 1999; Mystkowski et al., 2002; Rodriguez et al., 1999) showing that fear extinction is more stable when exposure occurs in a context closely resembling the one in which the fear was originally acquired. Additionally, experiencing the original uncontrollable earthquake stressors in a controlled environment, where the stressors can be initiated or stopped at will, may reduce the perceived uncontrollability of earthquake tremors, potentially creating an illusion of control over them. This could partly explain the dramatic reduction in anticipatory fear of future earthquakes observed immediately after the session.
Concluding remarks
Learning theory formulation of traumatic stress (presented in Başoğlu et al 2011) posits that helplessness responses that block natural recovery in trauma survivors can be overcome by an intervention designed to enhance sense of control over anxiety. Generalized improvement in all traumatic stress reactions with such an intervention is consistent with causal associations among helplessness, PTSD, depression, and social disability hypothesized by this formulation. Thus, the evidence supports not only the efficacy of the intervention but also its underlying theoretical framework. The fact that the intervention prompted high rates of treatment compliance and recovery when delivered in a single session is consistent with evidence pointing to an evolutionarily determined readiness in humans to utilize risk-taking behaviors in overcoming fear.
Compared to habituation-based exposure treatments, CFBT’s emphasis on enhancing a sense of control appears to have greater therapeutic potential, though this remains to be confirmed by future comparative studies. That said, even without such comparisons, the effectiveness of an intervention focused solely on fostering a sense of control—particularly one delivered in a single session involving only an explanation of the treatment rationale and self-exposure instructions, without any therapist-guided in-session exposure or between-sessions monitoring—demands further explanation. This is a critical question that anxiety researchers, as well as cognitive and behavioral therapists, must address.
References
Bandura, A. (1997). Self-efficacy: The Exercise of Control. New York: Freeman.
Başoğlu, M., Lax, T., Kasvikis, Y. and Marks, I. M. (1988). Predictors of improvement in obsessive-compulsive disorder. Journal of Anxiety Disorders, 2, 299–317.
Başoğlu, M., Marks, I. M., Kılıç, C., Brewin, C. R. and Swinson, R. P. (1994a). Alprazolam and exposure for panic disorder with agoraphobia: Attribution of improvement to medication predicts subsequent relapse. British Journal of Psychiatry, 164, 652–659.
Başoğlu, M., Marks, I. M., Kılıç, C., Swinson, R. P., Noshirvani, H., Kuch, K. and O’Sullivan, G. (1994b). Relationship of panic, anticipatory anxiety, agoraphobia and global improvement in panic disorder with agoraphobia treated with alprazolam and exposure. British Journal of Psychiatry, 164, 647–652.
Başoğlu,M.,Marks, I. M. and Şengün, S. (1992). A prospective study of panic and anxiety in agoraphobia with panic disorder. British Journal of Psychiatry, 160, 57–64.
Başoğlu, M. and Mineka, S. (1992). The role of uncontrollable and unpredictable stress in posttraumatic stress responses in torture survivors. In Torture and its Consequences: Current Treatment Approaches, ed. M. Başoğlu. Cambridge: Cambridge University Press, 182–225.
Başoğlu, M., Mineka, S., Paker, M., Aker, T., Livanou, M. & Gök, S. (1997). Psychological preparedness for trauma as a protective factor in survivors of torture. Psychological Medicine, 27, 1421-1433.
Başoğlu, M. & Şalcıoğlu, E. 2011) A Mental Healthcare Model for Mass Trauma Survivors: Control-Focused Behavioral Treatment of Earthquake, War, and Torture Trauma. Cambridge University Press.
Craske, M. G., Kircanski, K., Zelikowsky, M., Mystkowski, J., Chowdhury, N. and Baker, A. (2008). Optimizing inhibitory learning during exposure therapy. Behaviour Research and Therapy, 46, 5–27.
Emmelkamp, P. M. and Mersch, P. P. (1982). Cognition and exposure in vivo in the treatment of agoraphobia: Short-term and delayed effects. Cognitive Therapy and Research, 6, 77–90.
Foa, E. B., and Kozak, M. J. (1986). Emotional processing of fear: Exposure to corrective information. Psychological Bulletin, 99, 20–35.
Foa, E. B., and McNally, R. J. (1996). Mechanisms of change in exposure therapy. In Current Controversies in the Anxiety Disorders, ed. R. M. Rapee. New York: Guilford Press, 329–343.
Livanou, M., Başoğlu,M.,Marks, I.M., De Silva, P., Noshirvani, H., Lovell, K. and Thrasher, S. (2002). Beliefs, sense of control and treatment outcome in post-traumatic stress disorder. Psychological Medicine, 32, 157–165.
Marks, I. and Dar, R. (2000). Fear reduction by psychotherapies: Recent findings, future directions. The British Journal of Psychiatry, 176, 507–511.
Marks, I. M. (1987). Fears, Phobias, and Rituals. Oxford: Oxford University Press.
Marks, I. M. (2002). The maturing of therapy: Some brief psychotherapies help anxiety/depressive disorders but mechanisms of action are unclear. The British Journal of Psychiatry, 180, 200–204.
Marks, I. M., Swinson, R. P., Başoğlu, M., Kuch, K., Noshirvani, H., O’Sullivan, G., Lelliott, P. T., Kirby, M., Mcnamee, G. and Şengün, S. (1993). Alprazolam and exposure alone and combined in panic disorder with agoraphobia: A controlled study in London and Toronto. British Journal of Psychiatry, 162, 776–787.
Mineka, S., Mystkowski, J. L., Hladek, D. and Rodriguez, B. I. (1999). The effects of changing contexts on return of fear following exposure therapy for spider fear. Journal of Consulting and Clinical Psychology, 67, 599–604.
Mineka, S. and Zinbarg, R. (2006). A contemporary learning theory perspective on the etiology of anxiety disorders: It’s not what you thought it was. American Psychologist, 61, 10–26.
Mystkowski, J. L., Craske, M. G. and Echiverri, A. M. (2002). Treatment context and return of fear in spider phobia. Behavior Therapy, 33, 399–416.
Powers, M. B., Whitley, D., Telch, M. J., Smits, J. A. J. and Bystritsky, A. (2008). The effect of attributional processes concerning medication taking on return of fear. Journal of Consulting and Clinical Psychology, 76, 478–490.
Rachman, S., Craske,M., Tallman, K. and Solyom, C. (1986). Does escape behavior strengthen agoraphobic avoidance? A replication. Behavior Therapy, 17, 366–384.
Rachman, S. and Lopatka, C. (1988). Return of fear: Underlearning and overlearning. Behaviour Research and Therapy, 26, 99–104.
Rachman, S., Robinson, S. and Lopatka, C. (1987). Is incomplete fear-reduction followed by a return of fear? Behaviour Research and Therapy, 25, 67–69.
Rodriguez, B. I., Craske, M. G., Mineka, S. and Hladek, D. (1999). Context-specificity of relapse: Effects of therapist and environmental context on return of fear. Behaviour Research and Therapy, 37, 845–862.
Şalcıoğlu, E. (2004). The effect of beliefs, attribution of responsibility, redress and compensation on posttraumatic stress disorder in earthquake survivors in Turkey. PhD Dissertation. Institute of Psychiatry, King’s College London, London.
Williams, J. and Maier, S. (1977). Transsituational immunisation and therapy of learned helplessness in the rat. Journal of Experimental Psychology: Animal Behavior Processes, 3, 240–252.
