The behavioral and cognitive models of phobia and anxiety have witnessed a substantial development over the last 35 years. More detailed descriptions of specific models of each anxiety disorder are provided in this chapter. However, earlier models of acquisition of phobia were initially based on the model of classical conditioning, outlined by Pavlov. The supposition was that neutral stimuli were inadvertently ‘paired’ with noxious outcomes (such as injury or unpleasant experiences), and that these associations were learned and the previously neutral stimulus was later avoided. Mowrer (1939, 1960) later viewed this simple associationist model as inadequate to explain the maintenance of fear of situations that were avoided, as the simple associationist model would imply that the strength of the fear should decline with longer avoidance. Mowrer posited a two-factor theory (explained below) that accounted for the acquisition of fear through classical associationist conditioning and the maintenance of fear through the anxiety reduction repeatedly experienced through escape of avoidance in the presence of anticipation of the feared stimulus.

Utilization of exposure paradigms—whereby the patient was urged to engage in exposure to the feared stimulus without the opportunity to escape—was expected to lessen the anxiety or fear as the patient experienced no harm during the exposure. Initially Wolpe advocated a form of reciprocal inhibition, pairing ‘responses’ such as relaxation, assertion, or the sexual response, in the presence of the feared stimulus. The rationale is that relaxation would be incompatible with fear and would replace fear as a response. Subsequent research on exposure to feared stimuli indicated that relaxation was not an important or even useful component of exposure.

The behavioral model emphasized the development of response and stimulus hierarchies that reflected increasingly more anxiety or fear for specific stimuli. Therapists were urged to begin with modeling their own exposure to the feared stimulus, while the patient later imitated this coping behavior. Use of exposure—while preventing escape or neutralization—would provide the patient with an experience of habituation to the feared stimulus and—in cognitive terms—the disconfirmation that the stimulus needed to be avoided because it conferred danger. This model was expanded to the treatment of specific phobia, SAD, and OCD.

Beck and Emery's cognitive model stressed both the biological preparedness of certain fears and the cognitive distortions associated with these fears. Thus, Beck was able to identify the role of the individual's interpretations, e.g., catastrophic interpretations of events or symptoms (‘I won't be able to stand it’, ‘I'll get so anxious I will die’), mislabeling (‘I am crazy’), and fortune-telling (‘something terrible will happen’). The initial cognitive approach advocated by Beck stressed the use of exposure in the context of identifying the patients’ predictions and testing them out through behavioral exposure.

Subsequent cognitive and behavioral models attempted to specify specific cognitive components for each of the anxiety disorders—indeed, arguing for a specific refined model for each diagnostic category. As a consequence of this greater specificity of the model, we describe how the CBT model is applied for each of the anxiety disorders in this chapter.

Having developed several psychological models of anxiety, psychoanalysts have only recently begun to focus on the treatment of specific anxiety disorders. Systematic placebo-controlled studies of specific anxiety disorders with psychodynamic approaches have not yet been accomplished. Nevertheless, psychodynamic theory has significant clinical explanatory potential for anxiety disorders through its focus on intrapsychic conflicts, unconscious fantasies, defense mechanisms, and the compromise function of symptoms, factors that are not central to other psychological or neurobiological theories. In addition, the clinical techniques of focus on the transference, examining the emotional impact of the patient's developmental history, exploring the meaning of symptoms, and the technique of using free association provide a broad array of therapeutic tools for potentially lessening symptoms and vulnerability to recurrence of disorders.

For systematic studies to be performed, psychoanalysts and psychoanalytic researchers must develop specific treatments, described in treatment manuals, for anxiety disorders focusing on dynamics specific to each of these disorders, as well as the particular treatment approaches tailored to these dynamics. As of the writing of this chapter, manuals of this sort for anxiety disorders have only been developed for panic disorder (Wiborg and Dahl, 1996; Milrod et al., 1997) and posttraumatic stress disorder (PTSD) (Lindy et al., 1983; Weiss and Marmar, 1993; Marmar et al., 1995). As will be discussed below, preliminary studies using these manuals suggest that the psychodynamic approach is a promising treatment for panic disorder (Wiborg and Dahl, 1996; Milrod et al., 2000, 2001) and PTSD (Lindy et al., 1983). In this section, we shall describe basic psychodynamic principles that can be used to develop psychodynamic models and treatment approaches to specific anxiety disorders.

In order to understand psychodynamic theories and approaches to anxiety disorders, it is useful to review certain core dynamic concepts.

Evidence suggests that neurophysiological vulnerabilities may trigger a psychological state that can increase the potential of an individual to develop an anxiety disorder. A temperamental fearfulness can affect the individual's perceptions of themselves and others, as well as the sense of safety of feelings and fantasies. Kagan and colleagues (Rosenbaum et al., 1988; Biederman et al., 1990; Kagan et al., 1990) identified a group of behaviorally inhibited children who demonstrated fear responses in the setting of environmental novelty. Children felt to be at risk for the development of panic disorder (offspring of parents with panic disorder and agoraphobia) were found to have high rates of behavioral inhibition compared with a control group, and children with behavioral inhibition were likewise found to have an increased rate of anxiety disorders. Thus, this fearfulness may have a genetic origin that in interaction with a particular set of psychological and environmental factors can trigger the development of anxiety disorders.

Psychodynamic psychotherapy operates through the identification of the unconscious and conscious fantasies and conflicts underlying anxiety disorder symptoms, bringing them into the therapeutic dialogue, where they can be understood and rendered less threatening. These fantasies can be brought to the surface by exploring the meanings of symptoms, the stressors that precede or exacerbate symptom onset, and the fantasies and feelings that emerge in the relationship with the therapist (the transference). As these fantasies and conflicts are rendered less catastrophic, the symptoms often diminish and resolve. An important component of this form of therapy is helping patients to become aware of, more tolerant of, and more effective in expressing their drives and wishes.

Exploration of underlying dynamic meanings of symptoms provides important clues about unconscious fantasies and conflicts that fuel anxiety symptoms. Although patients with anxiety disorders share general sets of symptoms, individual variations in the syndromes are an important source of information about unconscious significance. For instance, one patient's fear of choking during panic attacks when drinking liquids was linked to intense, exciting, and frightening struggles for control with her father when she was a child regarding how much food and drink she should have at the dinner table. The exploration of this symptom led to an understanding of angry and sexualized feelings in her relationship with her father that she experienced as dangerous, yet needed to reexperience over and over in the form of symptoms. Circumstances preceding symptom onset, feelings experienced other than anxiety, and defense mechanisms employed provide additional clues about the psychological origins of symptoms.

Use of the transference is a core component of psychoanalytic treatment. In the phenomenon of transference, components of central relationships are unconsciously experienced as deriving from current relationships (Freud, 1909/1953). This process takes place with the therapist as well. Understanding the patient's fantasies about the therapist and the treatment can be of value in any form of treatment, but from a psychodynamic perspective, the transference situation has far-reaching effects, and necessarily influences therapeutic outcome.

For example, a patient's fear that he will be abandoned by significant people in his life if he expresses his rage or frustration can be examined in the context of a stable, reassuring relationship with the therapist. Therapists also explore with patients how current perceptions or misperceptions of others, including the therapist, are linked with perceptions of significant others in childhood. For instance, patients who experience others and the therapist as shaming them may describe having experienced shaming behavior from their parents. Fantasies and dreams provide crucial information about intrapsychic conflicts, as well as the transference.

There is an emphasis in psychodynamic psychotherapy on monitoring one's own reactions to patients, referred to as the countertransference (Gabbard, 1995). Negative, critical, or distancing behavior, of which the therapist may or may not be aware, can have a disruptive impact on the therapeutic alliance, and can limit the impact of any treatment. Although awareness of one's own reactions to a patient is of value in any treatment, psychodynamic psychotherapists scan their own reactions as additional clues to understanding patients. For instance, the therapist may be aware of his own discomfort and avoidance when a patient with PTSD appears to be on the verge of discussing a particularly painful aspect of the trauma she experienced. Not all reactions to patients, however, are induced by particular patient behavior and attitudes, and psychodynamic psychotherapists attempt to learn about the various feelings different patients, conflicts, and disorders may elicit in them. With patients with anxiety disorders, therapists should be particularly concerned about fantasies and fears of exacerbating a patient's anxiety symptoms.

The defining characteristics of specific phobia are intense fear of anxiety in the presence of a specific stimulus or situation, where this fear results in impairment or discomfort, and the individual realizes that the fear is excessive. Typical specific phobias include fears of animals, blood or injection, heights, water, insects, rats, and other stimuli or experiences. About 11% of the general population has a lifetime prevalence of specific phobia (Wittchen et al., 1994).

Specific phobia is intense fear and arousal in the presence of a specific stimulus or feared object (such as heights, animals, water). This is distinguished from panic disorder (where the fear is that the individual's arousal will go out of control and cause a medical emergency or insanity) and from SAD where the individual fears that the symptoms of anxiety will be observed by others resulting in humiliation or embarrassment. Specific phobia is also distinguished from PTSD in that patients with PTSD fear intrusive memories or images. Specific phobia can be evaluated by use of a variety of instruments, including the Fear Questionnaire (Marks and Mathews, 1979) and the Fear Survey Schedule (Wolpe and Lang, 1964).

The most widely used theoretical model of specific phobia is based on learning theory. Since Watson's (1919) observations of a conditioned fear of furry objects in a young child (by pairing shock with a rabbit), behavior therapy has viewed specific phobia as resulting from a learned association of a negative consequence paired with a neutral stimulus. This classical, or Pavlovian, model was later modified in the two-factor theory of ‘conservation of fear’ proposed by Mowrer (1960). According to Mowrer, the initial fear was established through classical conditioning (e.g., the neutral stimulus of the stove was paired with the negative experience of being burned). However, avoidance of the stove in the future was based on operant conditioning—that is, when the individual approached the stove there was an increase of fear. Avoiding or escaping was associated with reduction of fear (thereby negatively reinforcing the operant of escape or avoidance through the consequence of fear reduction). The two-factor model thus accounted for the acquisition of fear through classical conditioning and the avoidance of feared stimuli through the negative reinforcement of reducing fear through the operants of escape of avoidance. Fear was thereby ‘conserved’.

The implication of the classical and operant models was that fear could be overcome by direct exposure without escape. In addition, Wolpe (1958) introduced the idea of responses incompatible with fear or anxiety with the concept of ‘reciprocal inhibition’. This refers to the fact that certain responses (or experiences) (e.g., relaxation, sexual behavior, and assertiveness) are incompatible with the response of fear. By pairing these incompatible responses (e.g., inducing relaxation in the presence of the feared stimulus) the individual can decondition the learned fear. Related to this model is the use of habituation techniques and extinction—that is, repeated exposure of the stimulus will reduce its potentiating effect (habituation) or repeated exposure without reinforcement (e.g., escape is negatively reinforcing) reduces the acquired associative link of the conditioned stimulus (CS) (e.g., the stove) with the learned (conditioned) response (e.g., fear).

While recognizing the value of conditioning and negative reinforcement for escape and avoidance, there has been a growing recognition of the importance of ‘prepared’ behaviors (Seligman, 1971), innate fears, or innate predispositions. According to these Darwinian influenced ethological models there are certain stimuli that the human infant is predisposed to fear. These stimuli reflect dangers in the evolutionary expected environment—that is, the primitive environment of danger from predators, natural catastrophes, and abandonment. For example, research on the distribution of fears in various cultures reveals that the same stimuli are largely equally feared and that these stimuli reflect primitive dangers. This nonrandom distribution of fears, with heights, water, animals, thunder/lightening topping the list, suggests that human infants and children are preadapted to fear events that confer danger. The Dunedin study in New Zealand offers further support to the ethological model of fear. In this study a large number of children were followed from early infancy to early adulthood and records of their fears and their experiences with feared events was obtained. Contrary to the ‘learned fear’ model proposed by associationist and operant theories, children who previously have suffered injuries from falling were less afraid of falling in the future. The learning models would have predicted the opposite—but the ethological model suggests that fears may be protective and innately predisposed. Moreover, an overwhelming high percentage of parents of children who feared water were afraid of the water on the very first presentation of a pool of water. Now, despite the argument that fears may be predisposed through evolution, the ethological model argues for some plasticity—that is, fears can be unlearned through exposure.

The cognitive model of specific phobia suggests that, in addition to the two-factor theory and the ethological model, there are specific cognitions and behaviors that may add to fear and avoidance. These include beliefs that the threat/danger of a stimulus is related to the fear that it elicits (see Ost, 1997; Ost and Hugdahl, 1981) and that safety behaviors may protect the individual from the threat. Examples of these cognitive distortions in fear include the following: ‘If I am anxious, then it must be dangerous’ and ‘I must get rid of the anxiety immediately’. Safety behaviors include superstitious behaviors or thoughts that attempt to neutralize the fear or provide some protection from the fear. Examples of safety behaviors that fearful individuals may utilize include repeated self-assurance (praying, self-talk), magical rituals (wearing specific clothing on an airplane), hypervigilant scanning of the environment (e.g., checking for sounds and movements on an airplane), collecting information about danger (e.g., checking the weather forecasts or safety records of airlines), and requiring someone to accompany them when in the presence of a feared stimulus. The cognitive model of specific phobia suggests that these safety behaviors act as a disattribution error—that is, ‘The only reason that I am safe is that I engaged in my safety behaviors’. Thus, safety behaviors might reduce the efficacy of the exposure used in behavioral treatment—a supposition now supported by empirical data.

There is overwhelming support for the efficacy of behavioral exposure treatment for specific phobia—in some cases, over 90% of patients being effectively treated with exposure treatment with some use of anxiety management techniques (Ost, 1997). Most fears can be successfully treated in fewer than five sessions, with massed practice or prolonged exposure yielding more rapid results.

Given the importance of the role of avoidance and escape in the maintenance of fear, behavioral treatments rely on repeated exposure to feared stimuli. The rationale for treatment is to identify the feared situations or stimuli, introduce the use of relaxation techniques (if needed), and engage the patient in gradual but prolonged exposure to the stimulus. We have found it helpful to educate the patient about the evolutionary significance of phobias—that is, that most of the stimuli that are feared (e.g., heights, water, insects, animals) would confer danger in a primitive environment where these feared stimuli were present and dangerous. This preparedness of phobia leads to the emergence of a fear later, but that the use of behavioral exposure can reverse this process. The two-factor theory of anxiety ‘conservation’ outlined by Mowrer (1939, 1960) can be helpful in understanding that fears may be acquired through being ‘paired’ with a noxious experience, but that they are maintained or conserved through the anxiety reduction of escape or avoidance.

Behavioral treatment of specific phobia follows a set pattern of interventions. During the assessment phase the therapist evaluates which stimuli or situations are avoided or experienced with discomfort. The Fear Survey is a useful assessment measure as is the Initial Fear Evaluation for Patients (Leahy and Holland, 2000). The patient's Fear Hierarchy (see Leahy and Holland) provides information for the assessment of a ranking or hierarchy of feared situations as well as the rating of degree of fear and whether the situation is actually avoided. Although anxiety management (such as breathing exercises and relaxation) are helpful, they are not necessary for exposure to the feared stimulus.

Socialization to treatment begins with providing the patient with the Information for Patients about Specific Phobia (Leahy and Holland, 2000) or by informing the patient of the nature of acquired and predisposed fear. Patients often find the Darwinian model provides them with a demystifying and nonstigmatizing explanation of their fear. Initial interventions involve training the patient in relaxation techniques (deep muscle relaxation, breathing, meditative techniques). Patients are trained in identifying Subjective Units of Distress (SUDs), rating their fear or anxiety from 0 to 100% (or 0–10), with higher numbers corresponding to greater fear. Imaginal exposure is used whereby the patient begins with imagining, in session, the least feared situation in the hierarchy and holding this image in mind until SUDs are reduced by 50% or more and then moving up the hierarchy to gradually more feared stimuli. In vivo exposure involves actual exposure to the feared stimulus. It is useful to obtain initial SUDs right before, during, and after the exposure and to elicit predictions from the patient about what he or she fears will happen (e.g., ‘the elevator will crash’ or ‘I will drive off the bridge’).

Safety behaviors are important impediments to exposure efficacy and these can be identified by asking patients if they do anything to make themselves feel safer. For example, asking the patient, ‘When you drive across the bridge, when you are afraid, do you do any of the following to make yourself feel safer—talk to yourself, avoid looking to the side, clench the steering wheel, slow down, or anything else?’

As the patient is able to tolerate situations higher in the hierarchy the therapist can indicate that continued exposure—far beyond normal experiences with the stimulus—should be continued after treatment has been completed. For example, a patient with a fear of elevators should be told to continue taking elevators up and down for weeks—even when it is not necessary—in order to overpractice exposure. Any ‘setbacks’ or ‘relapses’ should be followed by re-initiating the program of exposure. Relaxation should be continued on a daily basis in order to reduce physiological arousal.

The patient was an executive in his fifties who had suffered from fear of heights for 9 years—with this fear increasing in the past 3 years. The patient indicated that he feared crossing bridges, climbing mountains, driving in the mountains, and standing close to the edge of precipices. He indicated very little fear of flying and pointed out that his fear of heights was due to his fear that he might lose control of the vehicle or himself and fall over the side. He utilized a number of safety behaviors that he believed lessened his fear, including having his wife drive or accompany him as a passenger (‘She could take over the driving’), planning far ahead so as to anticipate trouble, avoiding looking to the side of the bridge, clenching the steering wheel, driving very slowly, alternating with the break and accelerator, talking to himself, avoiding the rear-view mirror, and avoiding bridges or heights totally.

The therapist explained to the patient both the Darwinian model and the learning theory model and provided him with the information sheet from Leahy and Holland (2000). He was quite skeptical of both models and said he would take a ‘wait and see attitude’. The therapist encouraged this and suggested, ‘Let's collect some data about what happens with your fear as we proceed’. A fear hierarchy for heights was obtained and the first intervention was imaginal exposure for thinking about specific bridges. The in-session imaginal exposure suggested little initial fear, so the imagined stimulus was changed to thinking about himself standing at the edge of a cliff. This immediately increased fear, which abated with prolonged exposure.

Specific safety behaviors were targeted. The therapist explained how these safety behaviors made him believe that he could not face the situation without these magical behaviors and thoughts and then relinquishing them would be important. The therapist utilized a role-play where the therapist played the role of the safety behavior thoughts (e.g., ‘You need to clench the steering wheel or you will go over the side’) while the patient argued against these thoughts. Furthermore, the patient was asked to imagine and later actually produce the opposite behaviors of his safety behaviors. For example, rather than clenching the wheel, he was asked to loosen his grip, rather than driving slower, he was to drive normally, rather than avoid the rear view mirror, he was to look at it on and off, and rather than avoiding looking over the side, he was to gaze on and off over the side. These were first practiced with imaginal training and later with in vivo training. Finally, he was to write out his predictions of what would happen and the actual outcome for various exposures.

Closer questioning revealed that the patient was inadvertently hyperventilating by taking very deep breaths during these experiences. Apparently he had ‘heard’ that you should take deep breaths to calm yourself. It was explained that this might add to his sense of light-headedness and that he should breathe normally.

After seven sessions (spaced over a 3-month period) after the initial intake, the patient had engaged in all of the feared behaviors in his hierarchy, including driving across numerous long bridges, driving for hours in the mountains, and standing at the edge of cliffs. These exposures became boring in themselves, but he was encouraged to continue to look for further opportunities after his treatment was completed.

The DSM-IV [American Psychiatric Association (APA), 1994] defines obsessions as

Compulsions are defined as

Insight into illness is no longer necessary for the diagnosis so long as the excessiveness or senselessness of obsessions and compulsions is recognized at some point during the course of the disorder.

OCD may be diagnosed using semistructured clinical interviews such as the Structured Interview for the DSM (SCID-P; Spitzer et al., 1987) or the Anxiety Disorders Interview Schedule (ADIS-IV; DiNardo and Barlow, 1988; DiNardo et al., 1993). Dimensional measures may also be used to assess for the severity and content of symptoms. The Yale-Brown Obsessive-Compulsive Scale is the most widely used rating scale in assessing severity of OCD symptoms (Y-BOCS; Goodman et al., 1989a,b). Other rating scales include the Mandsley Obsessive–Compulsive Inventory (Hodgson and Rachman, 1977), the Padua Inventory (Sanavio, 1988), the Obsessive-Compulsive Inventory (Foa et al., 1998b), and the Compulsive Activity Checklist (Freund et al., 1987). Finally, two recent questionnaires, the Obsessional Beliefs Questionnaire and the Interpretation of Intrusions Inventory, have been developed by an international consortium of researchers to identify and rate cognitive aspects of intrusive thoughts and obsessions (Obsessive Compulsive Cognitions Working Group, 1997, 2001).

Other measures to assess for general severity of illness include the Beck Anxiety Inventory (BAI; Beck et al., 1988a) and the Beck Depression Inventory (BDI; Beck et al., 1988b). Patients may also be given general measures of disability such as the Sheehan Disability Scale (Leon et al., 1992) to assess the degree to which the symptoms are interfering with the patient's functioning.

Treatment forms utilized over the course of treatment included the automatic and revised thought log, the obsession-compulsion monitoring form, the imaginal and in vivo exposure form, and the exposure monitoring form (McGinn and Sanderson, 1999).

Traditionally considered to be refractory to treatment, many treatments now effectively treat OCD. Treatments that have demonstrated efficacy include cognitive and behavioral therapies and serotonergic medications. Psychodynamic psychotherapy and many psychotropic medications have not proven effective in treating OCD (Knight, 1941; Black, 1974; Malan, 1979; Perse, 1988) and hence should not be considered first-line treatments.

Symptoms treated within a cognitive-behavioral framework include the obsessive thoughts, images, impulses, or urges, and the compulsions that may take the form of ritualized thoughts or behaviors. Also targeted in treatment are the secondary automatic thoughts that develop among patients with OCD (e.g., I am a bad person for having such thoughts). Essentially, two primary goals of cognitive-behavioral strategies are to (1) alleviate the anxiety associated with obsessions, thereby reducing the frequency and persistence of these thoughts, images, impulses, or urges, and (2) reduce compulsions and alleviate feelings of relief associated with compulsions.

Before treatment is initiated, detailed information is obtained on the nature and exact count of the patient's external (e.g., knives) and internal (e.g., images) triggers of obsessive anxiety, catastrophic fears (e.g., my baby will die), compulsive rituals (e.g., checks 25 times a day), and passive avoidance or escape behaviors (e.g., does not cook).

Michele is a 28-year-old woman who presented with longstanding obsessive fears of becoming contaminated by germs. She washed her hands multiple times a day and used gloves to attend to the simplest of household chores. More recently, she reported developing obsessive fears about her baby coming to harm. Michele began to ritualistically repeat a series of numbers (e.g., ‘6, 6, 6, 6, 6, 6’), phrases (‘I repent’), and images (e.g., imagined her baby playing with his toys) fairly continuously throughout the day. She dropped out of graduate school during her first semester, could not leave the house without her baby, stopped cooking (‘I can't touch knives’), and cleaning (‘I feel the germs will seep into my pores’).

During the initial treatment session, Michele was given a simple but detailed description of OCD, including facts and figures on demographics, prevalence, etiology, and so on. The cognitive-behavioral model was explained to Michele along with a description of the strategies she would learn in treatment. The importance of completing in between sessions was emphasized and her husband was identified as a co-therapist to facilitate completion of weekly assignments. Michele was prescribed Foa and Wilson's book (1991) titled Stop obsessing! How to overcome your obsessions and compulsions and was encouraged to join the Anxiety Disorders Association of America (http://www.adaa.org) and the Obsessive Compulsive Foundation (http://www.ocfoundation.org).

Using a thought log, Michele learned to identify and monitor secondary automatic thoughts during periods of obsessive anxiety. Michele learned that these habitually occurring thoughts and images typically followed obsessional thoughts and stimuli and typically led to anxiety and the urge to ritualize. Illustrative automatic thoughts were identified such as ‘if I eat this meat, I (or my baby) will get germs and die,’ ‘the fact that I imagined my baby getting stabbed means that he will die unless I think of him safely playing with his toys,’ or ‘I am immoral for thinking that he is dead.’ It soon became evident that these thoughts reflected her underlying dysfunctional beliefs that she was fundamentally evil, vulnerable to creating and experiencing harm, was personally responsible for any misfortune that befell her or her family and incapable of coping well during adversity. As her maladaptive automatic thoughts were replaced to reflect an appropriate degree of responsibility and vulnerability to harm, and as her beliefs about her morality and ability to cope were modified through rational self-examination, her anxiety associated with obsessions and compulsions began to decline. Within a few weeks, Michele's obsessions and compulsions became less frequent. As a result of daily practice in her own environment, Michele grew adept at restructuring her cognitions and soon began to feel confident that she could reduce her anxiety on her own.

Her list of anxiety triggers was now organized hierarchically from least to most anxiety provoking on a scale of 1–100 (e.g., garbage, meat, knives) and a working hierarchy was created to reflect increasing contact (and anxiety) with each item (e.g., imagining touching garbage, touching garbage). Prolonged, systematic exposure was initiated with the least anxiety-provoking item until she habituated to it, after which she was exposed to the next item and so on. For example, Michele first imagined touching garbage, then touched the lid with gloves, then without gloves. When she was successfully able to handle garbage using her bare hands with minimal anxiety, the next item on her overall hierarchy was selected (e.g., knives) and was again organized hierarchically to reflect increasing contact (and anxiety) with that item (e.g., looking at a picture of a knife, imagining holding a knife, looking at a knife. As Michele also presented with obsessive fears that could not be implemented through in vivo exposure (e.g., obsessions of baby being stabbed), she was exposed to increasingly anxiety-provoking scenes in her imagination until her anxiety declined (imaginal exposure).

Because Michele could not tolerate the anxiety associated with completely abstaining from rituals at the outset, a graded response prevention was formulated and administered in conjunction with exposure. Michele was prohibited from performing rituals to neutralize the anxiety associated with items currently or previously the subject of exposure but was permitted to ritualize to items to which she had yet to be exposed.

Although Michele was able successfully to tolerate exposure and was able to abstain from performing rituals during exposure sessions, she found it difficult to conduct exposure and refrain from performing rituals at home, even with her husband's assistance. To facilitate response prevention, a weekly contingency plan was instituted wherein Michele rewarded herself for conducting exposure sessions at home (e.g., bought herself a CD) and a response cost was instituted when she performed rituals (e.g., was not able to watch her favorite show, had to send money to a despised politician). Self critical thoughts were also modified to reduce feelings of excessive guilt on the occasions she inadvertently performed rituals.

Michele's overall mood improved as her anxiety began to decline. As Michele's obsessions and compulsions declined to manageable levels and she was able to go about her daily life with minimal impairment, sessions now focused on helping her maintain gains and prevent relapse. For example, Michele was encouraged to take charge of her continued treatment with less and less guidance from the therapist, understand the difference between symptom recurrence and relapse, learn how to cope with symptom recurrence, and encouraged to pursue new activities to fill in the long gaps of time that she had previously spent performing her rituals. Sessions were tapered down to biweekly, and then monthly sessions and so on as Michele learned to manage her symptoms on her own. Michele was encouraged to identify stressors that led to increased symptoms and contact the therapist if she experienced a resurgence in between sessions.

Although with our present state of knowledge treatment of severe OCD should be primarily psychopharmacological or cognitive-behavioral (Stein, 2002), psychodynamic approaches can provide additional understanding and insights into the illness, particularly in milder or more moderate forms (Gabbard, 2000, 2001). Patients may benefit from exploring the meanings and defensive functions of obsessions and compulsions. Shame or embarrassment about symptoms and the fantasies associated with them can interfere with treatment. The atmosphere of safety with the therapist and the therapist's nonjudgmental exploratory stance can aid the patient in easing his intense self-criticisms and more openly discussing his symptoms. Identifying and reducing these sources of resistance to treatment can also increase compliance with medication and CBT. OCD symptoms are highly disruptive of relationships; problematic interactions with others secondary to the symptoms can be productively examined in the transference–countertranserence work of the therapy (Gabbard, 2000, 2001).

Linda was a 40-year-old single woman who presented with multiple rituals and obsessional thoughts for many years that had become disruptive of her daily routine in the preceding 3 months. These included knocking on wood, checking the stove and locks, and being preoccupied with Zodiac signs to try to gain information as to whether something terrible was about to happen to her, spending about an hour a day on these rituals. In addition, Linda felt threatened by a very close relationship with her boyfriend, fearful of his betraying and rejecting her despite his expressing interest in marrying her. She had lost her job 4 months prior to presentation, apparently unrelated to her OCD symptoms, and was concerned about finding a new one. Although sertraline provided some relief, her symptoms continued at a reduced level and her fears about her boyfriend persisted.

Linda reported a difficult childhood with a father whom she experienced as neglectful or only interested her academic achievement. Although she made many efforts to gain his affection through her schoolwork she felt that he still rejected her. Her mother was an anxious and preoccupied woman, and Linda believed that she had to take care of her mother rather than receive maternal comfort. Furthering her problems, social unrest in her country of origin forced the family to move to the US when she was age 11. Thus she had to cope with the loss of friends and her home, and adapt to a strange new environment, a task that she found at times to be overwhelming.

The therapy explored the many functions of Linda's symptoms. She and her therapist noted that her feelings of helplessness and uncertainty that were triggered by the loss of her job reminded her of the upheaval she experienced when she had to leave her home as a child. The checking behavior was a coping mechanism to control these feelings of helplessness, by displacing them to potential fantasied disasters (fire, burglary) that she could avert by her rituals. In addition, Linda felt deeply threatened by her growing ties to her boyfriend and possible marriage. In particular, she felt certain at times that he would reject her once she committed to him, just as her father rejected her despite her efforts. She attempted to ward off this expected disaster with her rituals as well with horoscope checking, which focused on whether others with her sign were having problems with relationships. Helping Linda to understand the origins of her fears and the function of her obsessions and compulsions aided in the further reduction of her symptoms. In particular, helping her to tolerate her feelings of helplessness, and linking them to the anxiety and frustration of her childhood traumas, led her to feel less threatened by her current life challenges.

The hallmark feature of SAD (formerly social phobia) is excessive and persistent anxiety (or panic attacks) in situations in which the person is exposed to unfamiliar people or subjected to scrutiny by others while performing specific tasks (e.g., public speaking, eating in a restaurant). Such individuals fear that they will act in a way (or display visible anxiety symptoms) that will be humiliating or embarrassing. DSM-IV (APA, 1994) require that individuals recognize that their fears are excessive or unreasonable. According to the DSM, exposure to the feared social situation almost invariably provokes anxiety and hence these situations are avoided or endured with dread. As a result, these symptoms create significant distress and impairment in functioning. Individuals with SAD suffer from extreme loneliness and isolation and report impairment in social, occupational, marital, and other spheres of their life.

Commonly feared situations include formal speaking or interactions (70%), informal speaking or interactions (46%), problems with assertion (31%), and being observed by others (22%) (Holt et al., 1992). Individuals with SAD may fear one or two specific social situations such as public speaking, but the vast majority present with evaluative fears in multiple social situations. Finally, a small proportion of individuals fear almost any social contact with others and if such broad-based fears are present, the individual is classified as having Generalized Social Anxiety Disorder (APA, 1994).

SAD may be diagnosed using semistructured clinical interviews such as the Structured Interview for the DSM (SCID-P; Spitzer et al., 1987) or the Anxiety Disorders Interview Schedule (ADIS-IV) (DiNardo and Barlow, 1988; DiNardo et al., 1993). These interviews also help clinicians rule out other disorders that may explain the presenting symptoms and rule in other disorders that may co-occur with SAD. The Fear of Negative Evaluation Scale (FNE) and the Social Avoidance and Distress Scale (SADS) may be used in conjunction with diagnostic tools to measure concerns with social-evaluative threat and distress and avoidance in situations (D. Watson and Friend, 1969). The Leibowitz Social Anxiety Scale (LSAS) is a newer scale and is widely used to assess the range of performance and social difficulties experienced by individuals with social anxiety (Liebowitz, 1987). Behavioral assessment tests are also frequently used. Such tests typically ask individuals to role-play a social situation (e.g., give a speech or converse with a stranger) while the therapist monitors their discomfort level on several indices, including their subjective rating of distress, as well as behavioral (e.g., speed of performance), and psychophysiological (e.g., heart rate is monitored) measures.

The BAI (A. T. Beck et al., 1988a) may also be used to measure general anxiety levels and given the high rate of depression among individuals with social anxiety, the BDI (A. T. Beck et al., 1988b) is often administered. The Sheehan Disability Scale (Leon et al., 1992) may also be used to assess the degree to which the symptoms are interfering with the patient's functioning.

Treatment forms utilized over the course of treatment included the automatic and revised thought log, the imaginal and in vivo exposure form, and the exposure monitoring form (McGinn and Sanderson, 1999; Leahy and Holland, 2000).

Prior to the introduction of social phobia into the diagnostic nomenclature, few studies examined the efficacy of psychotherapy treatments for this condition. Since its introduction into the DSM-III (APA, 1980), numerous studies have been undertaken to determine the efficacy of psychotherapy treatments for SAD. A growing body of evidence now suggests that cognitive-behavioral treatments are efficacious in treating SAD and have been found to be superior to waiting-list conditions (see Hope et al., 1993; for a complete review, Taylor, 1996; Turk et al., 2002). Studies have also found that the effects of CBT are maintained in the long run, even for periods up to 5 years following therapy discontinuation.

There are clinical and psychodynamic similarities between panic disorder and social phobia. Clinically, social phobia shares the symptoms of anticipatory anxiety, panic-like symptoms, or panic attacks in feared situations, and phobic avoidance of feared situations. In addition, the two disorders may share a similar neurophysiological vulnerability, as behavioral inhibition described by Kagan et al. (1990) is associated with social phobia as well as panic disorder. Parents of children with behavioral inhibition have been found to be at greater risk for the development of anxiety disorders, particularly social phobia (Rosenbaum et al., 1991a,b).

Whether through physiological predisposition, developmental stressors, traumatic experiences, or a combination of these factors, these patients typically have internalized representations of parents, caretakers, or siblings who shame, criticize, ridicule, humiliate, abandon, and embarrass them. These perceptions are established early in life and then are repeatedly projected on to persons in the environment who are avoided, for fear of their being critical and rejecting. Avoidance adds to difficulties in developing coping strategies.

As with panic patients, in patients with social phobia, anger is threatening due to fears of rejection by important attachment figures. For social phobics, anger and disdain for others are typically denied and projected on to others in order to avoid acknowledging these feelings (Gabbard, 1992; Zerbe, 1994). However, with this projection, the patient views others as critical and rejecting of him, triggering social anxiety. Additionally, patients experience guilt about their anger at others for being critical or rejecting, and for their own aggressive yet denied wishes for attention. Social anxiety can serve as a punishment for this guilt.

In addition to conflicts with the experience of anger, social phobic patients struggle with intense feelings of inadequacy. Alongside their low self-esteem, they can develop a compensatory grandiosity, with fantasies of others being very responsive or adoring of their specialness (Kaplan, 1972). This is typically associated with a desire to exhibit oneself sexually (Fenichel, 1945), which must be denied. This grandiosity adds to the recurrent disappointments that these patients experience in social situations, and may intensify the pain and anger they experience in response to rejection.

As with other psychological symptoms, from a psychodynamic view, social phobia also represents a compromise formation. Social phobics are conflicted about the wish to exhibit themselves sexually, and social anxiety is both an expression of the conflict, and a punishment for the wish. Avoidance of social situations aids in avoidance of the conscious experience of these wishes. Similarly, anxiety and avoidance punishes the individual for angry feelings and fantasies. Efforts at idealization of self or others attempt to ward off painful feelings of low self-esteem but then add to the potential for disappointment.

The therapist must be particularly alert to the patient's shameful feelings in treatment of social phobia. The patient may anticipate that the therapist will be as critical and rejecting of him as he expects others will be. This can be used as an opportunity to explore an early transference reaction to the therapist and to examine the patient's fantasies that he experiences as conflicted. In particular, angry fantasies and exhibitionistic wishes may emerge. The therapist explores the patient's fears upon entering a social setting, and why the patient may have difficulty confronting these fears. This inquiry will often aid or encourage the patient to confront his social anxiety directly.

The goal of cognitive-behavioral strategies is to alleviate the anxiety and avoidance behaviors associated with the social or performance situations. When possible, group treatment is the format of choice for patients with social anxiety because it is cost-effective, gives participants the opportunity to learn vicariously, see others with similar problems, and make a public commitment to change (Sank and Shaffer, 1984; Heimberg, 1991). Group treatment also provides the opportunity for multiple role-play partners and a range of people to provide evidence to challenge distorted thoughts (Sank and Shaffer, 1984; Heimberg, 1991).

Treatment is initiated once the therapist has established the diagnosis of social anxiety and assessed the extent to which symptoms create distress and impair functioning. The therapist identifies key cognitive, behavioral, and physical symptoms of anxiety, lists all the social situations that patients endure with dread or avoid altogether along with the safety-seeking behaviors they employ to cope with their anxiety in social situations.

James is a 32-year-old computer analyst who described his social anxiety as a curse passed down from generations. He recalled that he was shy as a child and never spoke up in class. He remembers rejecting a variety of career options including his dream to become a musician. He feels that he was trapped behind what he called an ‘invisible barrier’ and feels that he never allowed people to see his ‘true’ personality. Although he is attractive, James was afraid of dating and had never had a meaningful relationship until he was actively pursued by a woman whom he ultimately married. He decided to begin treatment after he was promoted to the position of a manager. He initially turned down the position but after he read an article on SAD in Time magazine, James decided to accept the new position and pursue treatment.

Although James had begun the process of reading on SAD, the psychoeducation phase reinforced his growing understanding of his condition. Realizing that he had a disorder that could be treated effectively quickly reduced the symptoms of depression he had been experiencing for the past 2 years. He began to feel optimistic that he could be helped and expressed an eagerness to continue with treatment. James was assigned self-help books such as Ronald Rapee's Overcoming shyness and social phobia (1998) and was encouraged to join the Anxiety Disorders Association of American (http://www.adaa.org).

James reported that he had been experiencing increased physical tension as he had accepted the new position. To combat these symptoms, he was taught deep muscle relaxation and breathing retraining and was instructed to practice exercises daily. As he mastered the exercises over the next few weeks, James was encouraged to use them as and when he needed before he faced anxiety-provoking situations.

Although James had many social or performance-based situations that triggered anxiety, his decision to accept the promotion at work necessitated a focus on interpersonal situations at work related to his new position. Using a thought log, James learned to identify and monitor automatic thoughts during periods of social anxiety at work or in anticipation of social encounters in his new position. Representative automatic thoughts were identified such as ‘he will think I am stupid,’ ‘I am going to mess up,’ ‘they will be able to see that I am nervous,’ ‘they will be waiting for me to fall on my face,’ ‘they won't listen to me,’ and ‘I will not be able to cope with the stress of this new job.’ Once James was able to identify his own automatic thoughts, he was encouraged through guided discovery and Socratic questioning to consider the fact that he did not know for sure what others were thinking, and to help broaden his perceptions away from the most catastrophic predictions (e.g., he may not notice that I am nervous, she may be thinking that I am better than the previous manager). His perfectionistic standards of performance (e.g., I cannot expect that I will be able to be an effective manager immediately) and his belief that others were critical and evaluative (e.g., she will think I am a loser) were also modified. Within a few weeks, James grew skilled at identifying and challenging his automatic thoughts using Socratic questioning. As a result of daily practice, he began to notice a reduction in anxiety, particularly during moments when he anticipated social encounters at work.

Using a fear and avoidance hierarchy, the therapist and James identified key interpersonal situations that he would face in his new position. Key situations such as interfacing with clients and his team at work in his new position were transformed into specific, behavioral tasks such as meeting clients face to face, calling clients on the telephone, holding a meeting with his staff, asking his staff to conduct tasks, and so on. Once these tasks were rated it became clear that even the smallest task (e.g., calling clients on the telephone) was creating more than a moderate level of anxiety (e.g., over 50 on a scale of 1–100). Consequently, imaginal exposure and anxiety provoking tasks unrelated to his place of employment were first used in initial exposure sessions. For example, James practiced exposure with tasks such as asking strangers for the time (30) for directions (35), asking acquaintances for simple favors (40), imagining talking to clients on the telephone (45) before he confronted more anxiety-provoking tasks at work. In addition, other exposure tasks such as mispronouncing a word in front of others (60) and slipping and falling in front of strangers (75) were used later on in the hierarchy to help James learn that he was capable of coping even if he did place himself in a position where negative evaluation might occur. By integrating cognitive restructuring into exposure, James was able to acknowledge that he could not expect to become a skilled manager rightaway and consequently, was able to set nonperfectionistic goals during exposure. Simulated exposure exercises were also used with the therapist and James role-playing key situations. For example, other individuals were bought in to the session to simulate work meetings during an exposure session.

As James did not possess leadership skills, exposure sessions were often preceded by sessions where requisite skills were practiced during sessions through instruction, modeling, behavior rehearsal, corrective feedback, and positive reinforcement. James was also assigned to read books on leadership and effective communication strategies in the workplace. Finally, to refocus his attention on conversations with clients and staff instead of on how he appeared to them, James was taught attention refocusing exercises. He was required to read increasing long and complex articles on computer programming, first under optimal conditions such as in his home after his wife went to bed, and then under increasingly distracting situations such as with music on, in the subway and so on. Finally, James learned to become aware of situations in which his attention wandered away from the task at hand (e.g., a conversation with a client) and learned to apply the new skills to refocus his attention away from the mental representation of himself and towards the task at hand.

James was encouraged to use cognitive restructuring before and after exposure situations to ensure that his fear of negative evaluation changed as a result of successful exposure. He was also encouraged to use exposure situations as behavioral experiments in which to test out irrational predictions. James was also encouraged to continue using daily relaxation exercises but was not permitted to use them during exposure sessions, in order to prevent his anxiety from reducing artificially. He practiced social and attention skills prior to exposure and soon began to feel less anxious, more confident about his ability to handle his new job and his ability to cope with his symptoms.

As his symptoms reduced and he was able to perform effectively at work, other social and performance situations were targeted in treatment. Acute treatment was discontinued once his overall symptoms reduced to manageable levels, his social functioning was no longer impaired, and he was able to guide his own treatment. Strategies to maintain gain and prevent relapse now became the focus of treatment and session were tapered to monthly sessions until James was able to manage on his own.

Panic disorder is defined by the occurrence of panic attacks, which are marked by intense physical sensations (heart palpitations, shakiness, sweating, shortness of breath, sensation of choking, chest pain, nausea, dizziness, feelings of detachment or unreality (depersonalization or derealization), fear of losing control or going insane, fear of a medical crisis (e.g., heart attack), numbness or tingling, and hot or cold flashes (APA, DSM IV)). Agoraphobia is characterized by fear of open spaces, places where exit is blocked or other stimuli (such as heights, bright sunlight), where the fear is that the situation may elicit a panic attack. The lifetime prevalence of panic disorder is 1.5–3.8%, with females twice as likely to manifest this disorder. Age of onset for panic disorder with agoraphobia is in the early twenties.

Panic disorder is distinguished from SAD in that in SAD the main fear is that others will see the individual's anxiety and that this will be a humiliating experience. Panic disorder is distinguished from OCD in that in OCD the main fear is of making mistakes or being contaminated or leaving something undone—rather than the fear of the consequences of one's own anxiety, as is characteristic of panic disorder. Although in the general population there are many individuals who manifest agoraphobia without prior history of panic disorder, it is individuals with both panic disorder and agoraphobia who are more likely seek treatment.

People with panic disorder and agoraphobia are 18 times more likely to try to commit suicide than people without any psychiatric disorder (Weissman et al., 1989) and are more likely to have an increased risk of cardiovascular disease, including aneurysm, congestive heart failure, and pulmonary embolism (Coryell et al., 1982, 1986). These people eventually have a risk of stroke that is twice the rate for other psychiatric disorders (Weissman et al., 1990; McNally, 1994).

Many of the situations that are feared by the agoraphobic are situations that might confer greater danger in an evolutionary adaptive environment (Leahy and Holland, 2000). For example, situations that might elicit panic attacks are open spaces (greater vulnerability to predators), closed spaces (vulnerability to suffocation or being trapped), bright sunlight (more visible to predators), and heights (danger of falling). Although the fear in panic disorder is of the consequences of one's own anxiety symptoms (that is, the fear of going insane, losing control, or a medical crisis) it may be that this ‘fear of fear’—elicited in these specific situations was adaptive to primitive ancestors. There is a reasonably high heritability component for panic disorder, suggesting a genetic link of some importance.

The cognitive-behavioral theoretical model is derived from the work of A. T. Beck et al. (1985), Clark (1986), and Barlow (1988). The initial physiological arousal—rapid breathing, dizziness, or sweating—may, in some cases, be due to greater exertion, fatigue, undiagnosed illness, life stressors—that are often underestimated by the panicker. This initial ‘panic attack’ is accompanied by a catastrophic interpretation—‘I am going crazy’—leading to hypervigilance for other signs of anxious arousal. This increased self-focus on one's own arousal increases the likelihood of arousal being detected or escalated—leading to false confirmations that another panic attack is imminent. Many panickers rely on ‘safety behaviors’—such as being accompanied by another person, stiffening one's posture, ‘taking deep breaths’ (that augment the hyperventilation syndrome). Situations that ‘trigger’ increased arousal—such as open spaces, heights, closed spaces, or behaviors that trigger arousal (exercise) are anticipated with dread or tolerated with increased discomfort.

Gould et al. (1995) have provided a meta-analysis of 48 controlled studies of cognitive-behavioral treatment of panic disorder with agoraphobia. The authors concluded from this analysis that CBT was highly effective in yielding panic-free outcomes, with an effect size of 0.88 (compared with an effect size of 0.47 for pharmacological treatment). The range of percent of patients who received CBT who were panic free after treatment was between 32% and 100%. In most of the studies reviewed, the percentage of panic free exceeded 80%. When CBT was compared with an emotion-focused approach, the former was significantly more effective than the latter (Shear et al., 2001).

The model for panic disorder described by Busch et al. (1991) and Shear et al. (1993) weaves neurophysiological factors with psychodynamic concepts and data to develop a psychodynamic formulation for panic disorder. This model was employed for the development of treatment interventions and manualization (Milrod et al., 1997). The authors describe that an inherent tendency toward fearfulness in unfamiliar situations results in a state of fearful dependency on significant others in the child's environment to provide a sense of safety. This anxious attachment causes a narcissistic humiliation for the child, as he cannot feel safe without the help of others, and a propensity toward anger at others for being unable to provide sufficient comfort to relieve his anxious state. Children may also develop a state of fearful dependency in environments in which parents behave in a critical, threatening, or rejecting manner.

Thus these children develop representations of others as abandoning, rejecting, and controlling. Anger at others is fueled by these perceptions, but the child is fearful of experiencing or expressing anger for fear of driving away or damaging the needed parent. Fearful dependency can be triggered again in adulthood by life events that represent danger or separation from a significant other. Angry feelings, which are often unconscious, are experienced as a danger to centrally important relationships, and signal anxiety is triggered. Defenses such as reaction formation, in which anger is converted into positive or helping feelings, or undoing, in which any negative feelings that do emerge into consciousness are taken back, attempt to quell the danger experienced from frightening angry feelings. However, these defenses fail, and patients experience the onset of traumatic anxiety in the form of a panic attack. The panic attack represents a compromise formation, in which the patient can express anger via demands for help from others, can desperately seek help in the setting of feared loss or separation, and can shut out angry feelings considered to be dangerous with a focus on intense, overwhelming anxiety. From the standpoint of the pleasure principle, patients experience a panic attack as less painful than the potential risk of loss of an important attachment figure, or of a conscious awareness of other symbolic meanings that the panic attack carries.

Case reports and psychological assessments of patients with panic disorder formed the basis for the development of a systematic approach to the psychodynamic treatment for panic disorder (Busch et al., 1991; Milrod et al., 1997). Milrod and Shear (1991) found 35 case reports of successful treatment of panic with psychodynamic psychotherapy or psychoanalysis in the psychoanalytic literature. A 15-session manualized psychodynamic psychotherapy for panic disorder, when combined with clomipramine treatment, was found to reduce the risk of relapse over an 18-month period following treatment termination compared with a group treated with clomipramine alone (Wiborg and Dahl, 1996). This study did not match treatment groups for frequency of therapist contact.

Milrod et al. (2000, 2001) conducted an open trial of panic-focused psychodynamic psychotherapy (PFPP) (Milrod et al., 1997), a manualized psychodynamic treatment that focuses on exploring the underlying unconscious meanings of panic symptoms and associated psychodynamic conflicts. This therapeutic approach was employed as a 24-session, twice weekly treatment intervention for 21 patients with DSM-IV panic disorder, using standardized panic disorder assessment measures recommended by the National Institute of Mental Health Collaborative Report (Shear and Maser, 1994). At study entry, patients had significant panic disorder and agoraphobia, along with functional impairment. Of 17 treatment completers (four patients were dropouts), 16 experienced remission of panic disorder and agoraphobia, and also experienced statistically significant, clinically meaningful improvements in phobic symptoms and psychosocial function, both at treatment termination and at 6-month follow-up following a 6-month no-treatment interval. The results of the open trial suggested that PFPP is a promising treatment for panic disorder. A randomized controlled trial of PFPP in comparison with applied relaxation training (ART) is in progress.

In treatment of panic disorder, therapists focus on the conflicts surrounding separation and anger as they emerge in precipitating events, interpersonal relationships, and in the transference. Examining the use of defenses is of value in bringing frightening feelings and fantasies to consciousness (Busch et al., 1995; Milrod et al., 1997). For example, the therapist treating a panic patient can identify the use of reaction formation when a patient is avoiding the experience of anger by being overly helpful to those with whom they are actually angry. For instance, a patient may refer to ‘loving to death’ a boyfriend whom she actually experiences as depriving and hurtful. Undoing, in which angry feelings are expressed and then taken back, provides an important opportunity to identify and explore the threat the patient experiences from angry feelings. By examining these defenses the therapist can help the patient with the core conflicts in panic, and with the fear of disrupting attachment to others who are considered essential to safety.

Generalized anxiety disorder (GAD) is characterized by physiological arousal (restlessness, fatigue, difficulty concentrating, irritability, muscle tension, insomnia) and apprehensive worry. Unlike other anxiety disorders where the fear or anxiety is about a specific event or stimulus, GAD is characterized by worry about several events (e.g., relationships, illness, finances, work). Lifetime prevalence of GAD is about 5% and 1-year prevalence is 4% reflecting the fact that GAD is widespread and chronic (Blazer et al., 1991; Wittchen et al., 1994; Kessler et al., 1999; Newman et al., 2003).

GAD is characterized by worry about a number of different things, the sense that worry is dangerous or out of control and physical arousal and tension. Measures assess the degree of worry in GAD (Penn State Worry Questionnaire), examination of beliefs about worry (Metacognitions Questionnaire), areas or topics of worry (Worry Domains Questionnaire), and the Intolerance of Uncertainty Scale (IUS). GAD status may also be evaluated using the Anxiety Disorders Interview Schedule (Brown et al., 1994) and the Generalized Anxiety Disorder Questionnaire (GAD-Q; Newman et al., 2002).

The behavioral model of GAD stresses both individual differences in arousal and experiences that are coupled with negative consequences. According to this model, specific events or stimuli become associated through conditioning with anxiety or fear. Treatment implications of the strict behavioral model include emphasis on decreasing anxious arousal through relaxation, coupling this relaxation with the feared stimuli (reciprocal inhibition; Wolpe, 1958), increasing exposure without escape, and enhancing assertion.

The cognitive-behavioral model, developed over the past 15 years, emphasizes the central role of worry in GAD (Borkovec, 1994; Wells, 1997). Worry primarily involves thoughts (rather than images) that are experienced as ego-syntonic, but which are associated with predictions of negative outcomes. In particular, worriers with GAD are more likely to perceive threats that are either not there or are ambiguous (MacLeod et al., 1986; Borkovec, 1994; Matthews and Wells, 1999, 2000), they underestimate their ability to cope with negative outcomes, and their negative predictions are often extreme. Borkovec noted that worriers with GAD often believe that the worry itself will cause negative consequences for them (such as sickness or insanity) and that their worry is ‘out of control’. However, worriers also believe that worry protects and prepares them and, therefore, cannot be easily abandoned.

A recent model of worry as intolerance of uncertainty has gained significant empirical support. Dugas, Ladouceur, Freeston and colleagues have indicated that worriers are often so intolerant of uncertainty that they continue to worry (or seek solutions to hypothetical problems) until the uncertainty can be reduced (Freeston, 1994; Dugas and Ladacoeur, 1998; Dugas et al., 2004). Ironically, though, given the intolerance of uncertainty, this search for a perfect solution above the threshold of certainty will lead to failure, thereby leading to further worry and further search for perfect solutions.

Borkovec and others have proposed that worry is an attempt to avoid negative emotions by relying on abstract, linguistic processing rather than direct emotional processing (Borkovec and Hu, 1990; Borkovec, 1994; Heimberg et al., 2003). When GAD patients engage in worry, they are actually less anxious or aroused, resulting in the inhibition of emotion during the worry phase. This inhibition of emotion regarding unpleasant content prevents ‘exposure’ or ‘emotional processing’, resulting in a later rebound of anxiety after the worry abates. Wells and his colleagues have described this as the ‘incubation’ of anxiety that results from relying on worry.

Developmental histories of GAD patients reveal an interesting pattern of experiences that may give rise to later vulnerabilities related to uncertainty, negative outcome, and concern over the feelings of others. For example, GAD adults report that during childhood that they had more disruptions in attachment relationships, experienced ‘reversed parenting’ (such that they attended to the emotional needs of a parent who often neglected the patient's needs), unpredictability of outcomes (or noncontingency), and had parents who combined both overcontrol and coldness. Presumably, these socialization experiences would sensitize worriers to the needs of others—for example, GAD patients rank high on empathy and their most common worries relate to interpersonal issues. Moreover, the child growing up in this kind of family would learn to inhibit emotional experience and expression and rely on anticipatory problem solving—either to soothe the emotional needs of the parent or to solve problems that others could not solve or that the child could not rely on to solve.

Most intriguing, in support of the emotional avoidance model, is that worriers as children were the most likely of all anxiety disorder patients to have experienced a physical trauma or threat of physical trauma. Yet, they are the least likely of adult anxiety patients to worry or fear such trauma. This may reflect that worriers engage in focusing on relatively irrelevant concerns as a way of avoiding the more troublesome physical vulnerability.

There is considerable support that cognitive-behavioral treatments are effective in the treatment of GAD (Gould et al., 2003), with some evidence that treatment gains are maintained 6 months after CBT is completed. Moreover, combining cognitive and behavioral treatment is more effective than behavior therapy alone (Butler et al., 1991).

The patient was a 53-year-old manager who complained of worrying all his life. Always someone concerned with being conscientious, he noted that his worry had become more pronounced in the past 15 years, as he had taken on more responsibilities at work that involved deadlines and uncertainty of outcome. He relied on sedatives for sleep, had found antidepressant medication to be unhelpful and had several short experiences in traditional psychotherapy that were not productive.

The general GAD model was explained to him, distinguishing between productive and unproductive worry, and emphasizing the importance of uncertainty tolerance. A distinction was made between worries that can rapidly (almost immediately) be turned into a ‘to do’ list of specific action versus worries about ‘possible’ events over which he had almost no control. Specific ‘to do’ lists were utilized daily, along with tracking actual accomplishments and behaviors on a daily basis. Worry time was set aside for flooding himself with worries and listing these worries. This resulted in a recognition that his worries primarily focused on a few areas—work to be done, meetings he would have, and the concern about being on time.

The worry time was immensely helpful to him and ran against his initial prediction that he would not be able to set aside worries until later. This gave him more of a sense of control. Specific predictions were elicited that characterized these worries—‘I won't get the report in’ and ‘People will be hostile toward me’—and these were tested weekly. Uncertainty training was implemented—with the therapist telling him that much of worry is the intolerance of uncertainty. He was urged to practice both in session and as self-help homework repeating, ‘It's possible that I can make mistakes and people will be angry with me’. In addition, he practiced visualizing (as exposure) images of negative outcomes until these outcomes became boring.

On a daily basis the patient was instructed to practice PMR, forming visual images of relaxing settings. In addition, he was encouraged to increase the frequency of aerobic exercise, which he did to a moderate degree.

In regard to his insomnia, he was instructed to avoid naps and to use the bed only for sleep and sex. Thus, he refrained from reading in bed, given this guideline. Like many insomniacs, his sleeplessness was due to mental activity. He was instructed to write out his worries and his action to do list at least 3 hours before bedtime. If he had difficulty falling asleep, he was instructed to practice repeating ‘I will never fall asleep’. The rationale for this instruction is that his insomnia was based on a worry—‘I might never get to sleep’—that he tried to neutralize by ‘trying to sleep’. This generally failed.

Over the course of nine biweekly sessions his worry diminished substantially and his sleep improved. He was urged to continue with the worry time, to do lists, uncertainty training, and practicing feared thoughts at the termination of treatment.

From a dynamic perspective, anxiety is linked to the potential emergence of threatening unconscious wishes into consciousness, and to early interpersonal relationships that form an internal psychological template in which attachments are experienced as easily disrupted. In GAD, defenses have been ineffective at neutralizing or disguising unconscious wishes, leading to persistent anxiety, or somatization may be operating as a primary defense. Crits-Christoph et al. (1995, 1996) suggest that early relationships in GAD patients trigger feelings of rejection, potential loss, anger, and a sense of needing to protect the caregiver to maintain the relationship. Ongoing anxiety derives from these conflicted feelings and the sense of unstable relationships. In addition, they hypothesize that past traumas can set off a pattern of generalized worry.

PTSD is defined by exposure to a life-threatening or injury-threatening experience in which the individual experienced intense fear, helplessness or horror and after which the individual experienced one of the following: intrusive recollections of images of the event, recurrent distressing dreams, experiencing the event as if it is recurring, psychological distress with exposure to the event, or physiological reactivity to stimuli similar to the event. In addition, there are attempts to avoid the stimulus and increased and recurrent arousal (insomnia, irritability, hypervigilance, etc.) (APA, DSM IV). The lifetime prevalence of PTSD in the National Comorbidity Study was 7.8% (males 10.4% and females 5.0%, with 60% of males and 51% of females exposed lifetime to traumatic events). Younger individuals are at greater risk for PTSD than older individuals.

PTSD differs from panic disorder in that the individual with PTSD has had these symptoms for longer than 1 month following the trauma (versus acute stress disorder) and re-experiences the traumatic event through intrusions, dreams, and a sense of the recurrence of the event (versus panic disorder). Evaluation instruments for PTSD include the Clinician-Administered PTSD Scale (CAPS), the PTSD Symptom Scale, and the Impact of Events Scale-Revised (Weiss and Marmar, 1997).

The behavioral model of PTSD entails both classical and operant conditioning, following Mowrer's two-factor theory. Specifically, it has been proposed that the original traumatic event results in a learned association of the emotional trauma that has occurred with the stimuli (visual images, sensations, sounds, etc.) of this event. Future encounters or memories of the event activate the traumatic experience, resulting in increased anxiety. Avoidance or even numbing following the event results in decreased anxiety, thereby reinforcing avoidance or escape and consequently maintaining the traumatic association. Foa and her colleagues have expanded on the behavioral model by proposing that PTSD is characterized by a combination of the associations described above and by the meanings given to the experience. This model stresses the importance of the ‘fear structure’, which includes the problematic interpretations given to the event, such as ‘I am never safe’, ‘I can be killed at any time’, ‘The world is not fair’, or ‘I am all alone’. Foa's model stresses the importance of both information and emotional processing and places the cognitive-affective ‘fear structure’ at the heart of PTSD. According to this model, attempts to assimilate the feared experience—in order to process it and give it meaning—occur during the intrusive ‘re-experiencing’, but are so overwhelming that complete processing is not obtained. This results in further attempts to avoid and, consequently, emerging interpretations that one is helpless and always vulnerable. Similar to ‘shattered assumptions’, the traumatic event may have more generalized implications for the individual about the nature of physical and interpersonal security and meaning.

Specific cognitive processes involved in PTSD include dissociative experiences (derealization and depersonalization), increased recall of vivid imagery associated with the trauma, but also a tendency in some cases to have vague or overgeneral recall (Loftus and Burns, 1982; Brewin and Holmes, 2003). McNally provides an extensive review of the literature related to memory processes, ‘repressed’ memory, and so-called ‘traumatic amnesia’. His review casts considerable doubt on sensational claims of ‘recovered memory’ related to abuse and trauma. Rather, it appears that traumatic events generally are more memorable and account for the intrusive nature of subsequent PTSD. There is mixed evidence for attentional biases—but some evidence suggests that individuals with PTSD manifest the Stroop effect of interference with subliminal stimuli (Harvey et al., 1996). Shame and anger are also often associated with the traumatic experience, mental defeat (a combination of helplessness and dissolution of personal identity, Elhers et al., 2000), negative beliefs associated with depression and PTSD (Foa et al., 1999). Brewin and colleagues have proposed a dual representation model of trauma, suggesting that information is encoded and experienced as verbally accessible memory (VAM) or situationally accessible memory (SAM), with sights, sounds, and sensations experienced at the more ‘primitive’ level of SAM (Brewin, 1996; Brewin and Holmes, 2003). Thus, effective treatment of PTSD would entail both the verbal or narrative meanings associated with trauma (VAM) and the more concrete stimuli and sensations entailed in SAM. Interpersonal factors are also associated with PTSD, with lack of social support predicting continuation of symptoms (see Brewin et al., 2000). Finally, eye movement desensitization and reprocessing (EMDR) was developed by Shapiro and has been utilized for treatment of PTSD by associating the elicited images of trauma with rapid eye movements produced by the patient following the therapist's hand. Although some studies have found this to be as effective as exposure and anxiety management interventions, the findings are mixed.

There is considerable support for cognitive-behavioral treatments of PTSD, with some protocols utilizing a combination of various interventions and other utilizing other CBT interventions. It is not unusual for CBT outcome studies to utilize extended or double sessions (60–120 minutes) so as to allow for sufficient exposure and habituation to the feared stimulus. Empirical support for the efficacy of these treatments can be found in numerous reports (Foa et al., 1991, 1995; Tarrier et al., 1999).

In Freud's conceptualization (1920), trauma pierces the ego's ‘stimulus barrier’, overwhelming the ego. In an attempt to cope with resulting traumatic anxiety, the ego employs dissociation, minimizing painful feelings through denial, or separating the feelings from thoughts and memories surrounding the trauma. Any intense affect state can create fears of a recurrence of the trauma (Krystal, 1988). In addition, the individual is driven to repeat the trauma in an attempt to assuage feelings of overwhelming helplessness and lack of control.

As with other anxiety disorders, the vicissitudes of anger play an important role in the psychodynamic underpinning of PTSD symptoms. Patients with PTSD experience intense rage at those they view to be responsible for their trauma (Brom et al., 1989). This rage is projected on to others not connected to the event, who are consequently viewed as dangerous, intensifying anxiety. Patients may employ the defense of identification with the aggressor, in which they ally themselves with the individual or group responsible for the trauma (Lindy et al., 1983). This mental operation can help to allay feelings of helplessness and may provide a sense of empowerment. However, identification with the aggressor often triggers guilt, and fears of becoming like the abuser.

Survivor guilt, a core feature of one type of posttraumatic reaction, can occur when an individual survives a traumatic experience in which others have died or have been severely injured. The individual who survives unconsciously identifies with the victims of trauma, but may also develop an unconscious identification with the perpetrators of the trauma, as described above, triggering guilt.

Developmental experiences can affect the risk of developing PTSD in response to a trauma. Traumatic developmental experiences can disrupt the early sense of autonomy and cause a regression from the developmental level that has been attained. Traumatic experiences in adulthood also cause regression, and often reawaken past experiences of disillusionment and anger at parents for failures to protect children from earlier traumas.

Lindy et al. (1983) studied 30 survivors of a fire using a brief (six to 12 sessions) manualized psychodynamic therapy. Nineteen of the survivors met DSM III criteria for PTSD. The patients demonstrated significant improvement and were less symptomatic than a control group of untreated survivors at follow-up. Brom et al. (1989) found improvement in patients with PTSD in three treatment groups (psychodynamic psychotherapy, hypnotherapy, and systematic desensitization) compared with a control group. The psychodynamic treatment was more effective with avoidance symptoms compared with the other treatments, which were more effective with intrusive symptoms. A manualized psychodynamically oriented group psychotherapy for Vietnam veterans with PTSD has also been developed (Weiss and Marmar, 1993), but has not been systematically tested.

In the psychodynamic treatment of PTSD, efforts are made to explore the precipitating traumatic event to give the patient an opportunity to discharge feelings of rage and terror and to investigate the unconscious significance of the event. Unlike other exposure-based treatments of PTSD (Resick and Schnike, 1993; Foa et al., 1999; Jaycox et al., 2002), therapeutic focus is not on reexperiencing the trauma. Exploration and reexperiencing of a traumatic experience may be harmful or disruptive to some patients, and therapists need to be careful to modulate their exploration to what the patient can tolerate (Krystal, 1988; Gabbard, 2000). It is important to investigate what made this event traumatic to this particular patient, and what factors in the patient's background, including prior traumas, rendered them susceptible to PTSD. The therapist should identify unconscious fantasies of identification with the aggressor related to the trauma. In therapy of all anxiety disorders, but particularly in PTSD, exploration of the patient's need to be punished by the symptoms themselves as a result of intense guilt is essential.