Meta-analyses have demonstrated the efficacy and effectiveness of psychodynamic psychotherapy [1-4]. While continued controlled studies demonstrating the effectiveness of psychodynamic psychotherapy are both needed and ongoing, there have also been calls by scholars for more research examining the process by which psychodynamic psychotherapy works [4]. Therefore, importance is increasingly being placed on the specific techniques, interventions and processes within dynamic theoretical frameworks and how these relate to therapeutic outcome.

Along with transference interpretations, defense interpretations are amongst the core technical techniques found in psychodynamic psychotherapy. This is true in both long-term [5,6] as well as shortterm [7] models of psychodynamic psychotherapy. The overall aim of these two technical interventions is to make the unconscious or latent material conscious, and point out how the distortive processes that transference and defense mechanisms involve can cause and maintain psychological duress [5,6,8-10]. Although there is extensive literature on the general concept of interpretation in psychodynamic psychotherapy [11-14] including on transference interpretations [15], much less is known about defense interpretations. This is surprising given the amount of research that has been conducted on defence mechanisms in general. Indeed, a flurry of research activity over the past few decades has been dedicated to studying defensive functioning. This research has led to several reviews that synthesize the theoretical, methodological, and empirical aspects of defenses [16-18]. There is evidence showing that patients who undergo successful psychodynamic treatment demonstrate more adaptive defense use [19-27]. Furthermore, the use of mature defenses has been shown to be associated with healthy psychological and physical functioning [18,28,29].

Though defense mechanisms are considered a pillar of psychodynamic theory, there remain significant gaps that need to be addressed. For example, there is no definitive text that outlines specific technical guidelines for how therapists should deal with defenses employed by patients. Without such a text, work that integrates clinical acumen derived from theory and empirical studies is lacking. As a result, very few resources are available to clinicians that show how recent empirical findings can inform the contemporary practice of psychodynamic therapy. In order to synthesize current thinking on how therapists should intervene with patients’ defenses in psychotherapy, this review set out to identify therapeutic principles by examining theoretical and empirical sources that pertain to therapeutic interventions aimed at patients’ use of defenses. The goal of creating this list of principles was twofold. First, it could help in establishing consensus amongst different scholars on how to address defense mechanisms in therapy, and highlight points of contention. Second, it would allow for the examination of the existing empirical evidence for these principles in order to spur future research into these principles.

Literature searches were performed using PsychInfo, Psychoanalytic Electronic Publishing (PEP) web and Medline with the following keywords: “defense” or “defence” or “defense mechanism” or “defence mechanism” and “therapeutic technique” or “interpretation”. No specific time period was used for literature searches. In addition, a number of other sources were obtained through informal channels by consulting experts in the field, including through listservs of psychodynamic researchers (i.e., Association for Psychodynamic Research, Society for Psychotherapy Research).

Figure 1 contains a graphical representation of the process by which the literature was retrieved and selected. A team of three researchers, including one senior psychodynamic researcher and clinician and two advanced doctoral students, identified and examined 136 sources. These sources included 45 books, 51 empirical articles and 40 theoretical articles that were reviewed to determine if they provide information on how to work with patient defenses in psychotherapy. The team discarded obvious false positives because they were unrelated to therapeutic technique and defenses. Subsequently, 29 books, 49 empirical articles and 19 theoretical articles were retained. These sources were then thoroughly examined for core ideas, suggestions, or guidelines by authors having to do with therapeutic technique and patient defenses. One researcher independently completed the review of all texts; a second researcher completed a similar analysis for approximately 70% of all materials. The two raters worked independently. Each rater recorded detailed notes and bibliographical information including author, text (actual citation) and page number. At this point, the material identified ranged from several phrases in a paragraph to a few pages, all of which had some bearing on therapeutic technique and defense mechanisms. The resulting information was independently compiled in a table by each rater; references that stated similar concepts were grouped together. Information that was unspecific or too vague was removed. For example, a number of historical case studies that were examined [30-34] were removed because their specific focus on an individual case made it difficult to extract broad themes. The two final tables, one for each of the two raters, were then compared in a series of consensus meetings.

Figure 1: Flow chart: Data retrieval and analysis.

The next step involved a comparative analysis where similar themes were grouped together using guidelines established by Hill and associates [35,36] for consensual qualitative research (CQR). CQR involves an inductive process by which conclusions are built up from the data; the consensual process helps to reduce the risk of researcher bias. While Hill and colleagues [35,36] have made suggestions for preset themes or domains based on Strauss and Corbin’s work [37], in this review the domains were abstracted from the tables to reflect core ideas about working with patient defences. Grouped themes had to have at least two different sources stating similar or equivalent information in order to be considered a principle. Each rater first worked independently to identify principles, then met for discussion. Once the raters reached consensus on the principles, an auditor reviewed the principles identified by each rater and those agreed upon in their consensus meeting. Another consensus session was then held. Agreement was found on eight principles. In comparing other principles that had been highlighted by each of the raters, the group concluded that two additional principles could be agreed upon, for a total of 10 principles. Table 1 shows the ten principles with a breakdown of the sources that were included in the final list of principles.

Defense Principle	Summary of Notes as Derived from Sources Associated with Principle	References
1. Considering the “Depth” of an Interpretation. Specific to Projection Specific to Immature defenses seen in PD patients	1) Recognize the defense used by the patient 2) Undo what has been done by the defense (e.g. “to find out and restore what has been repressed, rectify the displacement, to bring back what has been isolated into its true context”) 3) Analyst should then returns his analysis from ego back to the id, meaning examining impulses, drives, wishes, etc.\ 4) Interpret at the level of the Ego.     5) Interpret defenses before impulses and conflicts. 6) Interpret from surface to depth.  7) Analyze resistance before content Ego before Id. 8) Begin with the “surface”.  9) Confront: show patient that he is resisting show patient why he is resisting show patient what he is resisting show patient how he is resisting 10) Interpret motive for resistance mode of resistance (i.e., the defense mechanism) 11) Address inconsequential areas first and avoid the underlying wish and the anxiety it produces. 12) Examine why the defense was used, how it was used, the context in which it appears, and the conflict associated with it. 13) Interpret affects first followed by wishes since wishes are “deeper”. 14) Do not fail to identify the latent or unconscious meaning of the defense. 15) Confront defenses in order to call the patient’s attention to what is happening earlier in therapy rather than later. Leave out motive or impulse until later when interpretation is used. 16) Do not include information that is too deep or contains Id content too early in the therapy.  17) Focus of attention; surface and context: stick to observable “data” within the analytic hour (the verbal expressions of the patient), focus on the immediacy of the material. E.g. even though a patient talks about the past or future pay attention to the way it is described in the present (through undoing for example). No need to elaborate on what the patient says (i.e. the conflict associated with the defense). 18) Demonstrating the defensive manifestation to the patient: Task of the analyst in this stage is to show the “observed data” to the patient with the task of making the patient aware of how his or her ego handles intrapsychic conflict. 19) Invite the patient to suspend the defense and examine the defense “rationally” for a moment and observe along with the therapist. “The analyst’s task is to clearly include the essential package of defense and what it interfered with, without overloading the patient’s ego”  20) Address defenses as oppose to what is being defended against.   21) Do not interpret from “surface to depth” because that could reinforce the idea that there is something “behind” people’s behavior and keep the projection in place.   22) Use supportive interventions: acknowledgement, questions, associations, reflections, clarifications, Support Strategies (these are to be used in early and middle phase of therapy and should lead to more emotional elaboration). 23) Use exploratory interventions: Interpret and help patient develop insight: Progressively more deepening Defense interpretations used that combine affect, the defensive operation, motive, objects affected, and how the defense was learned in formative relationships.      24) Do not use interpretation: may works for neurotic patients but not personality disorders (alienates the patient and usually comes across as hard, judgemental, etc.) “Rather the therapist should inquire about, and help patients to think through, the consequences of their actual or intended actions.” 25) “Too rich a mixture of interpretative techniques early on may correlate with poorer outcome especially with more disturbed patients”   26) Mix between support and interpretation must be adapted to patient defenses. Also support alone is not enough for the alliance. “At each level of a patients’ defensive functioning there appears to be some specific range of more exploratory (interpretative) interventions that will be optimal to facilitate growth of the alliance” 27) “Adjust” supportive and interpretative techniques to patients’ defensive functioning.	[68]   [69]  [41]  [83] [6]      [49][6]  [6] [6]       [46]   [46]      [44]  [44]  [45]    [18]     [25]  [71]    [74]
2. Intervene with Patients' most Prominent Defenses.	1) Confront more prominent defenses whenever they are obvious especially when they are defending important repressed material.   2) Address both characterlogical defenses as well as those that are “out of character” because they may be related to a symptom.	[6]  [5]
3. Interpretations should begin with Defenses used as Resistance.	1) Address defense used as resistances first during the session.	[6,10,13,48,59]
4. Attend to Defenses used both Inside and Outside of the Therapeutic Hour. For immature defenses only	1) Confront defenses (used both in and out of therapy) before moving on towards an “in-depth analysis” of the unconscious meanings of these defenses so that structural change is achieved.2) Point out reality based problems first.3) Point out the cost of the defensive action to the patient and others.  4) Use defense interpretations only within the context of the transference relationship with the therapist and not in relationships outside of therapy. 5) Interpret defenses as they are used in psychotherapy and not outside of the analytic hour.  6) Interpret current sources of stress (outside the therapy). 7) Interpret the interaction of personality needs and defenses. 8) Acknowledge when reality based problems interfere with the mode of resistance (defense)  9) “Pursue this and similar mode of activities in and outside of the analysis.”  10) Assist patient is addressing their internal world once external stressors are manageable then we 11) Confront the defenses with the permission of the patient.	[6] [6]   [83,84] [46,85,86]  [10]  [5]  Lemma, 2003, p. 193  [18]
5. Consider the Timing of Interventions.	1) Interpret defenses in the "middle phase" of therapy.    2) Defense interpretation observed in mid-phase of treatment.   3) Defenses tend to change in the last half of therapy   4) Focus on defenses in the mid phase of treatment but should also be addressed throughout therapy.   5) Do not interpret prematurely since the intervention will have little use but will not be harmful. 6) Interpretation should be done at the beginning of the session rather than later because there will be more time to work through.7) Interpret when the patient is about to gain insight on his own   8) Do not interpret prematurely or it will damage the alliance.  “Therapist does not wait for defenses to be expressed as resistances in transference reactions or toward the therapy to interpret them”9) Do not underuse interpretations  10)“Sustained intervention by the therapist” addressing defenses is necessary throughout therapy	[48] [58] [24] [8]  [48] [48]   [6] [6] [6] [27]
6. Consider the Affect Associated with the Defense when Appropriate. Acting Out Isolation Intellectualization, Compartmentalization  Reaction Formation Idealization & Devaluation of Other/Self Passive-Aggression	1) Assist patient in focusing on anxiety to counteract the defense. That is, get the patient to focus on the raising level of anxiety before the act so as to limit the probability that anxiety will lead to uncontrollable impulse. 2) Address the idea that emotions are associated with being childish or weak. 3) Interpretations that address the cognitive level of understanding, before affective responses have been disinhibited, will be counterproductive. Be careful for the difference between intellectual and emotional insight.  4) Address the idea that emotions such as anger are sinful and should not be expressed. The obsessive tries to obscure magical thinking with this defense. 5) Confront these defenses and interpret affects of envy and greed that are assoicated with them. 6) Interpret underlying belief that anger drives people apart or causes “bad” things to happen.   7) Assist patient in venting angry feelings 8) Help patients acknowledge that they are in fact angry.   9) “We attempt to find out what the painful affect is making the patient resistant…”	[44]  [44]  [44]  [44]  [44,61,62] [44]  [18][59]
7. Consider the Degree of Emotional "Activation" Associated with the Defense. Acting Out	1) Interpret when the defense is neither too emotional nor emotional cold or detached. 2) Use interpretation that produce enough activiation. Interpretations that do not produce enough activation it will be effectual, if on the other hand the interpretations that produce too much activation will result in “an explosion of anxiety”. 3) Goal is to control the discharge of anxiety; too much anxiety is overwhelming and unproductive, too little does not lead to meaningful insight.“It will be necessary to interpret the ego defence” 4) Interpret when the defense is cold or no longer emotionally active. This way the defense will not be too emotionally charged and will thus be more amenable to interpretation. The patient will be less likely to use denial or other disavowal defenses to keep the acting out in place.   5) Assist patient in focusing on anxiety to counteract the defense. That is, get the patient to focus on the raising level of anxiety before the act so as to limit the probability that anxiety will lead to uncontrollable impulse. 6) Confront acting out as it is an important part of intervening with this defense.	[42] [11]  [68]  [44]  [6]
8 Avoid using Technical Language in Interpretations.	1) Do not use technical language in interpretations or it will promote isolation and intellectualization.   2) Scan the associations of the patient to build the interpretations.3) Making tentative interpretations; official objection to the use of technical terms was nearly unanimous.	[6]   [10] [8]
9. Balance between Supportive and Interpretive Interventions Splitting Acting Out Pass-Aggression. Turning against the self Projection	1) Therapists’ use of supportive interventions did not impact the development of either maladaptive or adaptive defences.  2) Have patient envision the positive and negative aspects of an object at the same time by using unconditional positive regard, firmness and safety.3) Acknowledge that the complaint is “as severe” as you have ever seen, turn up the volume of the complaint with statements like “I don’t know how you can stand it”.4) Do not forbid the acting out but help the patient to use displacement instead (e.g. hit a bunching bag).5) Help patient to vent angry feelings and also get them to assert themselves outward and not inward. Help patients acknowledge that they are in fact angry. If patient describes cutting himself or herself then therapist should understand it “matter-of-factly”. Say “I wonder if there is some other way you could make yourself feel better, Can you put your feelings into words?” Point out the probable result of the passive aggressive behavior as it manifests itself. “What do you want for yourself?” 6) Validate the distress but not the projective content and be empathetic.7) Use Counter projective techniques.	[76]  [18]     [18]    [18] [87]
10. Accurately Identify Defense Mechanisms used by Patients.	1) Therapists should master the “technique” of addressing what defense or defensive level the patient is using.   2) Accurately address the defense used by the patient or the alliance and outcome may suffer.   3) Point out how (process) the patient is defending.      4) Be aware that any cognitive process can be used by the mind for defense, idea of individual mechanisms not important.	[79] [6,8] [5]   [78]

Table 1: Defense principles table.

The analysis led to the identification of 10 principles of therapeutic technique related to patient defences. Each one is reported below, along with its clinical implications as reported in the literature that was retrieved; this includes the empirical evidence that supports it.

Consider the “Depth” of an interpretation

One common element that is consistent among various authors is Freud’s original proposition from the “The Interpretation of Dreams” [38] identifying the major goal of psychoanalytic work as making the unconscious conscious. Through the employment of interpretation on the part of the therapist, it is believed that patients can understand their typical ways of defending thus rendering the defensive processes more controllable, less automatic, and bringing these processes into their conscious awareness [9].

Fenichel [39,40] was the first to propose what Greenson [5], Langs [6] and Wolberg [10] would later expand upon, namely leaving out “deeper” material until later in therapy. The psychodynamic concept of depth refers here to those aspects of the patient’s psychic structures that lay more deeply imbedded in the unconscious and are consequently more difficult to bring to light. Fenichel [39] restructured the concept of interpreting from surface to depth on the basis of two critiques of Reich’s [41] advice to repeatedly work through characterological defenses in order to arrive at more deeply unconscious material [11]. First, attacking character armour, as Reich puts it, can lead to narcissistic wounds because these character defenses are presumably interwoven into the personality structure and is therefore ego-syntonic. Reich [41] referred to the term “latent negative transference” to describe what he believed were destabilizing interpretations that patients were not ready to hear which subsequently damaged the working relationship. Thus, Fenichel’s [39,40] suggestions can be viewed more as a technical refinement of Reich’s [41] earlier approach rather than a complete rejection of it.

Langs [6] and Lowenstein [42] arranged interpretive work with patients into two different yet related technical tasks. The first technique in this process is to use “clarification” as a means of allowing patients to verbalize and elucidate their own defensive process without addressing any underlying meaning or unconscious process at work. At this point, therapists confront or draw attention to the process of what patients are doing during the course of therapy; however, no deeper unconscious material (e.g. wishes, fantasies, or impulsive urges) would be included in the therapist interpretations until a more thorough understanding of the unconscious conflict that underlie the defense is evident. The second technical task in this process involves deeper exploratory work that is achieved through the use of interpretation which is considered essential to changing problematic defensive patterns in patients.

Other authors have made statements similar to Lowenstein [42,43] and Langs [6] about confrontations and clarifications making it probably the most recognizable therapeutic axiom regarding technical considerations in psychoanalytic psychotherapy with respect to defense mechanisms. In other words, therapists should proceed in their work from “surface to depth” [39] and address those aspects of patient material that are readily discernible first, and then move on to deeper more unconscious material. Aspects of this axiom can be found in numerous texts (Table 1) that discuss defensive functioning and usually serve as an overall guiding principle of how to work with defenses over the course of psychotherapy.

The axiom is based on the assumption that working with the unconscious is akin to unpeeling an onion. As each layer is removed we find another layer waiting to be examined, understood, and analyzed. In some cases, it is referred to as addressing “the defense as opposed to what is being defended against” [44], which implies that therapists should avoid naming material to their patients that they have reason to believe will be too anxiety provoking or difficult to accept until a point in therapy when the patient is ready to accept such material.

The surface to depth idea can also be used to organize the order in which the therapist uses therapeutic techniques. For instance, Perry and Bond [45] suggest that when working with patients diagnosed with personality disorders, the therapist should structure his or her techniques accordingly from surface to depth with “lighter” interventions such as questions and clarifications at first, before moving on to interpretive work in order to give the patient enough time to assimilate understanding in a step-by-step approach.

The responsibility of the therapist to address defensive behaviour does not end at confrontation according to Langs [6]. He suggests that the next step to dealing with defenses is the action by which deeper understanding and insight helps patients give up the more problematic defenses for more adaptive ones. He proposes that interpretations be used once the more unconscious material is better understood by the therapist and take precedence over confrontations after that point.

Not all authors agree regarding the usefulness of interpreting defenses in this manner. For example, Vaillant [18] questions whether individuals with character pathology can actually benefit from “deeper” interpretations of the variety that Greenson [5] and Langs [6] discuss. In fact, Vaillant states that especially in the early phases of treatment, the use of any form of interpretation in general can be “disastrous” [18] because these patients experience interpretations as critical attacks by the therapist. Even advocates in favour of the use of interpretations warn against “too rich a mixture of interpretative techniques early on” [25] with this population as it may negatively influence the process of therapy. Specifically, for patients who tend to rely mostly on “immature” or “lower-level” defenses, management may be more useful than uncovering. Management assumes a more limit-setting and active approach to helping patients, one that does not confront the behaviour head on but rather encourages the individual to change problematic defense patterns. On the other hand, interpretations, especially with reference to depth, are more indicative of confrontation and thus may be too anxiety provoking for at least those patients with more serious psychopathology. However, it should be noted that Vaillant’s [18] suggestions are not directly related to any empirical investigation and he focuses solely on defenses used by patients diagnosed with personality disorders.

Implications for practice and empirical evidence: This principle suggests that therapy should unfold in such a way that deeper material is presented to patients as treatment progresses. This is especially true when patients are relatively well functioning and do not rely too heavily on immature defenses to manage unconscious conflict. Therapists would refrain from using deep interpretations in psychotherapy for those patients with immature defense patterns, such as individuals diagnosed with personality disorders. This would be particularly true early in treatment.

Despite the promising theoretical material presented thus far for informing clinical practice it should be noted that there is little, if any, empirical evaluation of depth of interpretation and change in defensive functioning.

Intervene with patients’ most prominent defenses

The second principle suggests that therapists should confront more prominent defenses whenever they are obvious, especially when these defenses are obscuring important repressed material [6]. Essentially, because all individuals use a multitude of defenses in any given psychotherapy session, therapists should focus on those defenses that seem to be most closely related to conflicts associated with symptoms, anxiety, presenting problems, or other difficulties associated with functioning. Furthermore, therapists should address both characterological defenses as well as those that are “out of character” because they are also most likely related to a symptom [5].

Implications for practice and empirical evidence: The clinical implication of this principle is for therapists to pay attention to their decision-making process when deciding to intervene with one particular defense of a patient as opposed to another, as this decision may potentially affect the overall outcome or progress of psychotherapy.

Despite the claims made by both these clinician-theorists [5,6], very little research has examined how therapists choose to intervene with patients’ defenses during psychotherapy sessions and what exactly constitutes the “most prominent defenses” of a patient. For example, based on the literature reviewed, it is unclear if prominence refers to the most frequently used defense, least frequently used, or atypical defense or rather if the intervention should depend on the degree to which a defense is considered developmentally adaptive (i.e., defensive maturity). The most frequently used defense in this case refers to the defense the patient employs most often in-session and thus is believed to indicate some aspect of the patient’s habitual way of responding. Least frequent would on the other hand indicate defenses that arise rarely. Finally, an atypical defense would be when a patient uses a defense in a moment or situation that appears out of context given their character and what the therapist understands of their psychological organization. These nuances remain largely unaddressed in the literature at present.

Interpretations should begin with defenses used as resistance

Another principle found in this review was that those defenses seen specifically as resistance in-session should be addressed first by therapists in psychodynamic psychotherapy [5,6,10,13,46-49]. Resistance is defined as any defensive process aimed at interfering with the natural unfolding of therapy and thus prevents the further exploration and elaboration of unconscious material [50]. Although resistance is defensive in nature because it keeps certain affects, thoughts, ideas or impulses from consciousness, this construct is generally used when discussing the therapeutic setting. As Blum [51] has stated, “the concept of defense is broader than that of resistance since resistance is a treatment function that takes meaning from the analytic process”. Thus, while patients can use various defensive processes in their everyday life, they are only classified as resistance when these processes take place within the context of therapy. In fact, Freud [52] himself made this distinction quite clearly by stating, that“defensive mechanisms directed against former danger recur in the treatment as resistance against recovery”.

Some short-term approaches to psychodynamic psychotherapy, such as the ones proposed by Davanloo [7], Malan [53] and Sifneos [54] underscore the fundamental role of the clinician as that of addressing resistance. As Weiner and Bornstein [49] suggest, often pursuing or interpreting patients’ resistance is “more fruitful than a patient’s recalled memory” [46].

Implications for practice and empirical evidence: The implication for practice is that when a defensive process is interfering with therapy it should be addressed first so as not to hamper or interrupt the treatment; no real therapeutic progress would be possible until resistance is overcome since by its very nature, resistance blocks the progression of psychotherapy. Examples of this type of phenomenon include arriving late for sessions, cancelling appointments, inappropriate silences, or therapeutic ruptures.

Although there is widespread acceptance of this principle and several case studies [30-34] are presented to support the hypothesis that resistance should be addressed first, there is no empirical evidence for this claim in the scientific literature. Given our analysis of the sources identified in this review, it seems apparent that the concepts of resistance and defense are often confused in the literature and there is no current methodology available for differentiating the two.

However, the short-term approaches mentioned above [7] have an emerging evidence base supporting their overall effectiveness for patients diagnosed with a multitude of disorders including depression, anxiety, somatic disorders, and personality disorders [1-4,55]. Although these studies are not directly related to the principle discussed, they indirectly suggest that the investigation of resistance in psychotherapy is a potentially fruitful avenue of study.

Attend to defenses used both inside and outside of the therapeutic hour

The fourth principle refers to the difference between those defenses used within the therapeutic hour, which includes defenses used insession not pertaining to resistance, as well as those defenses that patients recount from their everyday lives. For example, both Langs [6] and Greenson [5] propose that therapists should in fact acknowledge when “reality-based” problems are influencing the defensive behaviour of patients. Wolberg [10] indicates that current sources of stress (outside the therapy), and their interaction with personality needs and defenses be addressed before therapy can unfold in a productive fashion. These authors imply that what unfolds outside of therapy is of value and understanding the defensive processes that patients recount from their “outside” lives could be an equally valuable pursuit in-session [55-68].

This is in line with Vaillant’s work [18] that suggests that events from outside the therapeutic hour should not only be acknowledged but also dealt with before systematic intervention focusing on defenses in-session is undertaken. The author explains that stressful life events could actually make a patient appear more “defensive” in-session than what their typical personality would suggest. An example of this situation would be if a patient describes using the defense of splitting in their everyday life but no evidence of splitting is observed during the session; the therapist must hence choose whether or not to make this part of the therapeutic work and to address it. Once externally based problems are under control, then patients may be in a position to address their internal world. Vaillant [18] maintains that this is the only way for patients who suffer from substance abuse problems or are diagnosed with personality disorders to benefit from therapy.

However, Gray [46,69-72] rejects this approach and pinpoints the therapeutic relationship as the only true context for interpreting defenses to patients. Where Gray suggests that therapists interpret almost exclusively inside the therapeutic setting, Vaillant argues that this would ignore a number of important events that are outside the therapeutic context. Although Gray is discussing a more traditional psychoanalytic approach to therapy than Vaillant, they both agree that therapists need to make a distinction between defenses used in-session and those used out-of-session when making interpretations.

Malan [53] further illustrated the importance of this distinction. He proposed a schema to demonstrate psychodynamic conflict that he considered the overarching principle of psychodynamic psychotherapy. This schema involves two triangles: the triangle of conflict and the triangle of persons. The triangle of conflict is comprised of three poles: defenses, anxiety, and feelings whereby defenses and anxiety block the expression of feelings. The triangle of persons is comprised of three poles: therapist, current persons (e.g. spouse/partner, boss, friends, children) and past persons (e.g. family of origin: parents, siblings, relatives). The triangle of persons is where psychodynamic conflicts are experienced. Therefore, within the triangle of persons, defenses could manifest themselves with the therapist (inside therapy) and with current or past persons (outside therapy) and both would be of importance to treatment. The ability of the therapist to effectively explore how intrapsychic conflicts play out amongst the triangle of persons would therefore be important in helping patients become aware of the pervasiveness of their patterns and how these patterns play out in multiple relationships in their lives, both past and current.

Implications for practice and empirical evidence: According to this principle, clinicians should focus more closely on the interpersonal process of defenses within the therapeutic hour with patients who are relatively high functioning and try to get patients to recognize how they use defensive strategies exhibited in everyday life within the therapeutic context. This follows from Malan [53] and emphasizes the importance of integrating the different poles in the triangle of persons. Conversely, for those patients with immature defense profiles, clinicians should first examine potential environmental stressors (e.g. unemployment, divorce, financial difficulties, etc.) that could be contributing to less mature management of conflict before moving on to the process outlined above.

This principle has not been investigated in the literature and is based predominately on theoretical assumptions and expanded upon in case examples by clinicians [28].

Consider the timing of interventions

The question of timing is an important aspect of psychodynamic technique with respect to the use of interpretation. Many, if not all texts that aim to educate practitioners regarding technique use invariably discuss the intricacies involved in choosing the correct timing when formulating one’s hypothesis about the patient and then vocalizing it during the therapeutic hour. This issue of timing can be divided up into two subcategories. First, the global idea of timing examines when to focus on the defensive functioning of patients over the course of the entire therapy. This would include both shorter and longer therapy durations. The second aspect of timing has to do with choosing the correct moment within the session to interpret.

With respect to the more global idea of timing, Reid [48] states that therapists should address defenses in the middle phase of long-term therapy so that the alliance has had sufficient time to develop before the more uncovering and slightly more anxiety- provoking aspects of defensive behaviour are pointed out. Reid [48] also states that early interpretation is neither helpful nor harmful. Langs [6], however, has suggested that early interpretation can damage the alliance and should be avoided when possible. Similarly, therapists should not address defenses too late in therapy as there may not be sufficient time to work through the material and thus may be more harmful to patients. Glover [8] concludes that although defenses are “focused on” during the middle phase of therapy, they should be interpreted throughout therapy.

There is much less work aimed at understanding the issue of when to interpret during a psychotherapy session. Reid [48] suggests the beginning as the most appropriate so as to allow enough time for patients to process the information. However, if patients are on the verge of gaining insight regarding their defensive behaviour on their own, the therapist may aid the process with an interpretation regardless of when this occurs during the session.

Implications for practice and empirical evidence: Many psychodynamic psychotherapy training manuals teach this principle by getting trainees to judge when the patient is “ready to hear” certain conclusions the therapist wishes to share [56,57]. Therapists should be aware of patient readiness, place in the treatment (i.e., early, middle, late), as well as timing during the session.

Additionally, the application of this principle would depend on the type of psychodynamic therapy a clinician is practicing as it will determine how and when a therapist should address defenses. For example, working within an intensive short-term dynamic therapy model (e.g. Davanloo’s approach), therapists would interpret defenses much earlier in the process of therapy and with more frequency and intensity (e.g. the ‘pressure and challenge’ technique; [7]).

Hersoug et al. [58] conducted a study to determine if patient characteristics (e.g. defenses) were associated with the use of specific psychotherapeutic techniques. They found that defenses tended to be interpreted during the middle phase of brief dynamic therapy (approximately 1 year), although individual differences between therapists were significant. Another study of short-term dynamic therapy [27] found that “sustained intervention by the therapist” throughout therapy was more predictive of change in maladaptive defenses. Winston and colleagues [27] suggest that timing is based less on the idea of “readiness to hear” and more on the idea that constant dosage of interpretations is required. Since dosage is essentially related to timing in that “how much” is a function of “when” the authors are arguing in favor of the liberal use of interpretations in short-term dynamic therapy throughout the course of treatment and not simply in the middle phases as suggested by Hersoug et al. [58]. It appears that the question of timing shows promise but requires further investigation.

Consider the affect associated with the defense when appropriate

This principle refers to the essential role that affect plays in understanding defensiveness and why all human beings are fundamentally motivated to defend. Affect is an intricate part of what is obscured and avoided when defenses are triggered. Furthermore, avoidance of particular affects can also motivate the use of defenses in individuals. Although not all defense mechanisms deal with affect in the same fashion, therapists must understand their place in the patient’s psyche if they are to intervene appropriately in-session. Chessick [59] underscored this therapeutic task by stating the therapist’s responsibility to “find” the painful affect that is being defended against. Naming affects and including them in communications to patients is part and parcel of psychodynamic psychotherapy, especially in the theory of defense. This is similar to Malan’s [53] conceptualization of psychodynamic conflict where defenses and anxieties block the expression of true feeling and are incorporated by many short-term dynamic therapies including Davanloo [7] and McCullough et al. [60].

While it is impossible to identify one particular affect that accompanies each defense mechanism, it is possible to distinguish certain defenses that typically deal with affect in the same manner. McWilliams [44] states that when therapists encounter the defense of acting out, they should get patients to focus on the rising level of anxiety or rage that is present immediately before the employment of this defensive process. Some defenses, such as isolation of affect, have an underlying function intended to diminish the consequence an affect will have on consciousness. For instance, isolation of affect often produces its effect through a general belief that affect and emotions are weak and should thus be avoided; effective interpretation of this defense should take this into consideration. McWilliams [44] also warns against interpretation at an overly cognitive level of understanding for obsessional defenses such as intellectualization, which may only further entrench reliance on the cognitive and negate the affective. For certain immature level defenses such as idealization and devaluation, it may be necessary to use both confrontation and interpretation in order to reach the typical feelings of greed and envy that are believed to be beneath these defenses [61,62].

In the case of passive-aggression, the idea of aggression is interwoven into the understanding of this defense. Although reaction formation does not fit into this category of defense, clinicians would need to uncover the magical thinking that subsumes this defense and challenge the idea that anger is unacceptable. McWilliams [44] again suggests that therapists should interpret the belief about one’s emotional world that leads to the use of this defense, specifically that anger will cause bad things to happen or drive people apart. Vaillant [18] clarifies work with this defense further by indicating that therapists should help patient’s actually vent their angry feelings and help them acknowledge that they are in fact experiencing anger.

Implications for practice and empirical evidence: Across all levels of defensive functioning, clinicians should work with the affective experience that is either transformed or distorted by the defense. Sometimes, this takes the form of interpreting that which is not readily available to the patient as is the case where painful affects are present but not easily acknowledged. At other times, patients should be encouraged to express the emotional material that is being turned into its opposite (as in the case of reaction formation) or expressed in a self-destructive, passive form (as in passive-aggression). The key here may be for the therapist to be attuned to what affect is being avoided by the expression of a particular defense. Each defense will have its own idiosyncratic method of dealing with affect and thus clinicians must be able to identify this process and use it to further the exploration of their patients’ psyche.

McCullough et al. [60] have developed Affect Phobia Therapy (APT), a form of short-term psychodynamic therapy that pays special attention to the interaction of affect and defense processes. Preliminary outcome studies show promise, and indicate that this form of therapy can be effective for treating Cluster C personality disorders [63]. Bhatia et al. [64], in a single-case process study examining APT, found that increased exposure to ward off emotions and decreased levels of inhibition lead to positive changes in patient functioning. This process was facilitated by the patient demonstrating insight into the types of defenses he or she was using and having the motivation to relinquish them. Overall, this single-case process study demonstrates the important connection between affect and defense processes and its importance to therapeutic outcome. This study relied on the Achievement of Therapeutic Objectives Scale (ATOS) developed by McCullough et al. [60] which allows researchers to examine different therapeutic processes and the extent to which a patient is able to assimilate within the course of a session. Further process studies on larger samples using scales such as the ATOS are needed [65] to determine how defensive functioning modulates the patient’s abilities to adaptively process and experience affect.

Consider the degree of emotional “activation” associated with the defense

The concept of “emotional activation” is based on the idea that a defensive process usually contains some form of affective material that is defended against, compartmentalized, or expressed. As such, therapists are in a position to observe this affective level when patients are defending. The defense of splitting usually contains a great deal of affect directed at an object in which a distortion of the object is readily noticeable to the observer (e.g. strong feelings of anger towards a significant other), however this can be experienced in different ways depending on how much emotional activation is attached to the defense. The patient can be enraged and thus extremely “charged”; as a result, very little interpretation would be possible because the sheer emotionality of the situation makes it inappropriate to use interpretation. On the other hand, the patient could be using the same defense (splitting) in such a way that no such extreme activation of affect is present and thus interpretation by the therapist is possible. Lowenstein [42] was the first theorist to elucidate this process and warned therapists against interpreting when defenses are too emotionally activated and therefore not amenable to interpretation. He also stated that interpreting when the emotional level seemed “cold” and “detached” was of little use.

This principle is consistent with the “mutative interpretation” proposed by Strachley [66], which states that an interpretation must evoke a specific level of anxiety to be effective. When too little anxiety is inculcated by the interpretation it may be forgotten or dismissed by the patient and fall short of its intended target. On the other hand, interpretations that provoke too much charge or anxiety may be overwhelming and could not be useful to generate true insight.

The defense of acting out is considered to be a maladaptive defense mechanism usually expressed as uncontrollable impulsive behaviour. This version of acting out is not to be confused with the broader conceptualization of acting out seen in the dynamic literature that has to do with “acting out” the transference outside of the therapeutic hour. Specifically with acting out, Freud pointed out that [67,68] therapists should always interpret when the defense is cold or no longer emotionally active; this way the defense would not be too emotionally charged and would thus be more susceptible to interpretation. Furthermore, this will make it less likely that patients will use defenses such as denial or other disavowal defenses to guard against examining the acting out.

More recently, McWilliams [44] suggested another approach with this defensive process that helps patients to focus on anxiety as a means of counteracting the defense. She advises therapists to encourage their patients to focus on the raising level of anxiety before the act so as to limit the probability that anxiety will lead to uncontrollable impulsive discharge.

Implications for practice and empirical evidence: The implication of this principle is that psychotherapists undoubtedly need to pay special attention to the level of emotional activation during sessions as a gauge for intervening with defenses. This of course will vary from patient to patient and from defense to defense. Clinical acumen plays an important role regarding optimum levels of activation required for effective interpretation of defense mechanisms.

There is however limited empirical support for this principle. Previous research by Salerno et al. [69] explored the relationship between therapist intervention, patient defensive functioning, and affect. Their findings support the link between these concepts by showing that confrontation elicited more patient defenses than did clarification. Salerno et al. [69] also found that patient affect and patient defenses are associated with one another regardless of the intervention used by therapists. In recent years, research has begun to show the link between certain forms of psychopathology and affect regulation, or the ability of individuals to regulate their own emotional states [70].

Avoid using technical language in interpretations

There is little debate in the literature regarding this principle as authors agree that therapists should refrain from using overly technical language in their verbalizations to patients regarding defense mechanisms. Not surprisingly, no sources were found that endorsed the use of lengthy or technical terms in interpretations. Glover [8] was the earliest source found to explicitly state that “official objection to the use of technical terms was nearly unanimous” in the psychoanalytic community. The author argues that technical terms have “no meaning” for patients and are ultimately useless in psychotherapy.

Langs [6] went one step further and hypothesised that the use of overly technical language by therapists may in fact promote the use of intellectualization and isolation defenses by patients. These defense mechanisms share the function of distancing patients from the experience of affect. As a result, therapists who use overly verbose and technical sounding interpretations may in fact be promoting the use of defenses that are counterproductive in psychotherapy.

Wolberg [10] distinguishes between “authoritative” and “tentative” interpretations. Tentative interpretations are presented as “hunches” or “best guesses” by the therapist so that the method of presenting unconscious material is used without provoking excessive anxiety. However, in certain cases when this approach is not providing the desired effect in the therapy, the author claims that a more authoritative language may be used to overcome forces of repression [10].

Implications for practice and empirical evidence: The clinical implication of this principle is to avoid jargon-filled interpretations and stay as close to the patient’s words as possible. Also, it would be important for the therapist to use language that is not authoritarian in nature but rather that is collaborative and inclusive with the patient. However, Wolberg [10] acknowledges the potential usefulness of more authoritative interpretations. Although each therapist will have his or her own unique style and choice of words, the key point of this principle is that a therapist can include the patient in the interpretation and not present interpretations as absolute fact. Using the same or similar words spoken by patient’s in-session is one way that therapists can stay as close as possible to the patient’s experience of reality.

To the best of our knowledge, no study has empirically examined the length of defense interpretations made by therapists and whether there is an optimal length of interpretive material a patient can handle at any given time. Given the convergence of clinical wisdom surrounding the use of appropriate and non-technical language with patients it would be important to empirically determine what impact, if any, adherence or non-adherence to this principle would have on the psychotherapeutic relationship and psychotherapeutic outcome.

Balance between supportive and interpretive interventions

While usually not considered to be as important or curative as the interpretation, supportive techniques also make up a large part of what dynamically oriented therapists do in-session. They differ from interpretations in that supportive interventions do not confront or make mention of unconscious material; instead they aim to support patients’ behaviours and generate practical solutions to problems. McWilliams [44] proposes that when using supportive techniques therapists should identify feelings and life stressors as opposed to interpreting defenses. McWilliams [44] indicates that this is especially true for patients who are more disturbed. This may, for example, require the therapist to be tolerant while listening to the patient’s frustrations without interjecting to point out defenses that arise during the process. Additionally, supportive techniques such as these sometimes require the therapist to not confront the patient’s distortions and resistances; however, it does not mean that the therapist agrees with the patient’s understanding of events but also does not mitigate or devalue their experience. Haven explains how, with the defense of projection, therapists should support the distress associated with the defense but not the projected content. Vaillant [18] goes one step further and suggests that the therapist can mitigate the difficulty caused by the use of certain maladaptive defenses (e.g. acting out) by gently nudging patients toward mid-level defenses (e.g. displacement). These suggestions are not interpretive in that they are not geared toward awareness and insight but rather make use of suggestion and therapist approval.

Implications for practice and empirical evidence: One useful clinical suggestion to emerge from the debate on support versus exploratory interventions is that clinicians should use both support strategies in addition to interpretive interventions as a means of achieving a strong alliance and favourable outcome. Although there currently is no definitive method for determining what this technical mixture should be, the defensive profile of patients based on the immature-mature continuum as well as the disorder they suffer from, appears to be a promising avenue of study empirically as well as clinically.

The idea of combining supportive and interpretive interventions has been studied empirically by Despland et al. [71]. They proposed that “at each level of a patient’s defensive functioning there appears to be some specific range of more exploratory (interpretative) interventions that will be optimal to facilitate growth of the alliance”. Although they stated that support alone was not enough in psychotherapy to form a strong alliance, the correct mixture of support and interpretation by therapists was considered necessary for an optimal therapeutic alliance. Despite the fact that Despland ret al. [71] were interested mostly in the alliance, due to the strong link between alliance and outcome in psychotherapy research [72,73] it is clear these findings also have implications for outcome variables as well. In that study, 12 patients seen in ultrabrief (four sessions) were assessed for alliance and defenses. Therapist interventions were also examined and then placed on a continuum referred to as the Expressive-Supportive Intervention Level (ESIL), on which there are various techniques that are interpretative in nature, or non-interpretative or supportive techniques. The research group used this continuum to calculate a ratio between the average technique level (supportive versus expressive) and defensive maturity level, by such indicating to what extent the therapist’s level of support-interpretation is adjusted to the patient’s level of defensive functioning. The results indicated that adjustment scores at session one predicted alliance scores at sessions three and four. This result was independent of initial defense scores. Those patients who started off with lower defense scores were still able to form strong alliances when therapists were well adjusted.

Siefert et al. [74] echoed the sentiment of the Lausanne group years later when they also concluded that therapists did in fact adjust their supportive and interpretative techniques to patients’ defenses early on in Short-term Psychodynamic Psychotherapy (STPP). Siefert et al. [74] found that overall defensive functioning predicted the use of both cognitive behavioural and psychodynamic interventions (supportive, expressive, etc.) indicating that therapists are using patients’ defenses as a guiding principle in these forms of psychotherapy even if implicitly. However, they were not able to reproduce the results of Despland et al. [71] with respect to defensive functioning and therapeutic alliance. Furthermore, Hersoug et al. [24] confirmed this latter finding when they found that initial defensive functioning did not predict either alliance or outcome on its own.

In another study, Hersoug et al. [75] questioned the earlier notion by Despland et al. [71], which assumed that therapist supportive and expressive interventions could be placed on a continuum and then compared to the defense hierarchy. They concluded that what was assumed to be a “poor” adjustment ratio, that is therapist interventions and patient defenses that are not congruent, was actually correlated with a stronger alliance score in some cases. They also found that when support strategies were given to patients with more adaptive defense scores, alliance tended to improve. This is counterintuitive when we consider that support strategies match with the lower end of the defense continuum to form a more “well adjusted” dyad. Hersoug et al. [75] explain this finding by suggesting that because Despland et al. [71] studied an ultra-brief form of therapy, it was not necessarily comparable to their naturalistic design, which examined Sessions 7 and 16 of a 40-session treatment.

In a follow-up study Hersoug et al. [76] found that interpretations, but not support strategies, were associated with a decrease in maladaptive defenses over the course of therapy. This relationship was not replicated with respect to adaptive or mid-level defenses. Although adaptive defenses did increase in the sample, neither the use of support nor the use of interpretive techniques explained the change. In a study of ultra-brief psychodynamic psychotherapy using sequential analysis, it was determined that therapists typically use supportive interventions to “prepare” patients before making defense interpretations [77]. That study also found that there are predictable ways in which psychodynamic therapists structure and use therapeutic interventions.

As a result, it appears that the relationship between defensive functioning at the beginning of therapy and alliance and outcome is dependent on a therapist’s ability to understand and use defenses as part of treatment planning. For example, all of the above mentioned studies did not find a direct relationship between Overall Defensive Functioning (ODF), or the average maturity level of the patient’s defenses, and the therapeutic alliance. Only the Despland et al. [71] study found an effect when the concept of adjustment was added. Therefore, it seems that the relationship between defense, alliance, and therapeutic technique is determined at least in part by the therapist’s ability to tailor the treatment to patients’ characteristics but the role played by supportive interventions is still open for discussion. These studies raise questions regarding how therapists structure their use of techniques in psychotherapy.

Accurately identify defense mechanisms used by patients

The final of the ten principles assumes that an important aspect of a therapist’s therapeutic competence has to do with correctly addressing the type of defense employed by patients; put differently, for example, when the therapist believes that he is addressing a patient’s use of the defense of intellectualization, is the person actually using that defense mechanism, Glover [8], Greensen [5], and Langs [6] sustain that the therapist must accurately address the “process” by which the patient is defending. The word process in this case refers to the psychological process by which the mind makes use of one mechanisms of defense (e.g. denial) over another (e.g. repression). Langs [6] suggests that inaccurate interpretations, which address the incorrect process, could conceivably damage the alliance or adversely affect outcome.

Brenner’s [78] thinking was somewhat different from the others in that he conceptualizes defense as any cognitive process that can be enlisted by the mind to serve a protective function. While Brenner [78] does not specifically argue against accuracy in defense interpretation per se, his writing suggests that it would be somewhat misguided to dedicate time and energy to accurately identifying specific processes if any of a multitude of cognitive processes could interchangeably serve this function.

Accuracy was largely ignored in the empirical literature until recently when more emphasis was placed on studying the effect of accurate versus inaccurate interpretations [79]. An accurate interpretation is defined as one that correctly recognizes the type and function of the defense used by patients in session, whereas an inaccurate interpretation fails in one or both of these aims.

Implications for practice and empirical evidence: The accurate identification of defenses used by patients in psychotherapy is an implicit assumption inherent in all training programs that teach psychodynamic psychotherapy and it speaks to a universal characteristic of any good therapy: if the patient’s experience is being accurately identified by the therapist it should in theory be positively related to the therapeutic alliance and to therapeutic outcome [80]. Junod et al. [79] examined this concept empirically within the context of strong and weak alliance dyads. Their results indicated that poor accuracy scores were more typical of the weak alliance dyads and higher accuracy was associated with a stronger alliance. However, they also found that overadjustment (interpreting more mature level defenses on average) was associated with the strong alliance group as well. Thus, it is difficult to determine whether it was actually the accurate identification of defenses by therapists or some other aspect of defense interpretations that accounted for the difference between these two groups. Moreover, this study only examined the average defense used by patients and the average level of interpretation made by therapists, thereby overlooking the moment-to-moment interaction of the therapeutic process that would be of vital importance in an investigation of this type.

Petraglia et al. [81] carried out another investigation in order to account for the interactive nature of interpretation accuracy. They found that higher adjustment scores, which involved either interpreting the defense that came immediately before the interpretation or a higher-level defense, were associated with a significant increase in the maturity of the defenses used by the patient immediately following the interpretation. Although these results seem promising, they should be interpreted with caution because this investigation had a very small sample size (n=6) and was exploratory in nature. Still, it appears that some evidence supports the notion that accuracy of interpretation is an important aspect of dealing with patient defenses. Perry et al. [80] also addressed the issue of accuracy in interpretation by examining three case studies in which the therapy was delineated along whether or not therapists could accurately identify the defensive process at work in the therapeutic treatment. The authors highlight the role therapists play in correctly recognizing important defense mechanisms as they change with successful treatment.

This study attempted to synthesize the available theoretical and empirical literature with respect to technical suggestions for interpreting defense mechanisms in psychotherapy into ten overarching principles. Additionally, from this synthesis of theory and research, an outline of clinical implications was described to help guide clinicians’ thinking when interpreting a patient’s defenses in-session. This list of principles is by no means exhaustive, nor are the clinical implications expected to serve as the “gold standard” for practice. Rather the goal of this study was to pragmatically integrate both classic psychoanalytic texts on technique and modern empirical studies into a guiding framework that can be used by theorists, researchers and clinicians to spawn future investigations and advancements.

This study had several limitations that are worth mentioning. First, although every effort was made to avoid missing important sources, it is possible that certain authors were overlooked either by omission or error. Second, due to the sheer volume of sources examined in the study it is also conceivable that certain sources were mistakenly discarded during data retrieval and analysis (Figure 1).

One potential future purpose of these principles is to use them as a means of first, generating new research and second, as research becomes increasingly available, steering psychotherapy training programs for evidence-based psychodynamic psychotherapy. Programs could eventually use these principles to measure levels of adherence to evidence-based principles for therapists learning to make interpretations of defense mechanisms. In this manner, psychodynamic psychotherapy training programs could integrate the teaching of theoretical material with principles that show empirical validity as well.

It is important to note that six of the ten principles (Principles 1, 2, 3, 4, 5, 8) discussed in this paper have never been researched to date. Thus, much work remains to be done in the field. The remaining four principles (Principles 6, 7, 9, 10) have been investigated and have guided clinical work already. The novelty of this paper is the suggestion that these different areas of research be integrated to form part of a whole, which could then be used in a cohesive and unified way.

Moving forward, what is desperately needed in psychodynamic psychotherapy is a more thorough empirical evaluation of the rich clinical tradition that this approach rests upon. The next step in the process of refining defense technique would be to investigate these principles more thoroughly by means of empirical analysis and translating this into useable guidelines that are both empirically robust and clinically relevant.