"Case Formulation: Strategies and Criteria for
Selection of Negative and Positive Cognitions In EMDR"

Andrew M. Leeds, Ph.D.

October 7, 1995, Adapted and updated from
a presentation at the 1994 EMDR Conference
"Research and Clinical Applications" Sunnyvale, California

Introduction

The process of selecting appropriate negative and positive cognitions can be a smoothly flowing preamble to an EMDR treatment session or it can be a complex search for an elusive quarry. In this paper, I will review the principles and rationale underlying the selection of cognitions in EMDR and will consider procedures to guide us safely past common problems. Most importantly, I will introduce the idea that the process of selecting cognitions can be greatly simplified and enriched when it is integrated into a case formulation approach.[1]

Rationale for Negative and Positive Cognitions

In the EMDR model, negative and positive cognitions and their associated affects are considered the gateway to a hypothesized nodal memory network [2] that holds an incompletely processed information package in dysfunctional form. The information package is intrinsically a blend of perceptions, affect, and cognitions (Shapiro, 1993). EMDR is theorized to intervene directly into the memory networks on a neurophysiological level to help reprocess sensory and affective memory substrates [3] and thereby generate new more appropriate beliefs about self and world. In the EMDR model, negative cognitions are viewed as distillations of incompletely processed affects from (specific) earlier life experiences. In this sense, affect plays a more central role in both the EMDR theoretical model and treatment than in cognitive therapies.

Why Select the Negative Cognition before processing?

Clarifying the NC serves two key purposes. The first is to establish the PC. The second is is to help stimulate the affective component. In addition, an appropriately selected NC enhances generalization to other related traumatic memories.

Why Select the Positive Cognition before processing?

Initially clients lack an affective linkage with the PC. Before processing, clients say they know the PC is true, but don't feel it is true. Asking clients to select the PC before EMDR processing, 1) activates nodal memory elements associated with the PC in short term memory where they (2) represent the potential for a therapeutic outcome and (3) help maintain dual attention to facilitate information processing. (4) It can be diagnostically revealing to recognize and work within the patient's present capacity to "see down the road" and access positive resources. If necessary you can select a more modest PC.

When to Select the Positive Cognition

Develop the PC early in the session before clients are immersed in the affects associated with the NC. Early selection of the PC allows a simpler procedure by not having to elicit the disturbing feelings twice, once before, and once after identifying the PC. A VoC is assessed once the PC is selected. Note: a VoC on the NC is not requested.

Three Characteristics of the Negative Cognition

Three elements make up the NC: 1) A presently held 2) irrational belief, that comes to mind when 3) focusing on the disturbing memory. The NC is: 1) not what was thought at the time of the original event, 2) not a possibly true description, 3) not necessarily believed (or acted on) all the time.

Phrases Helpful in Eliciting the Negative Cognition

Often a useful phrase is: "What words best go with the picture that express your negative belief about yourself or the experience?" Avoid asking "What feelings do you have about yourself in that picture?" since we want to generate a belief not an emotion. 

In eliciting the NC and PC use the sense mode matching the client's description. So if the client describes hearing the sound of an explosion, ask "As you remember hearing that sound, what thoughts do you have about yourself now?" Use corresponding phrases for memories involving body sensations or visual images.

Selection Criteria for Negative Cognitions

The first criterion of a NC is that it be an irrationally negative, self-referencing belief. This self-critical, self-blaming assessment is often based in guilt: "It's my fault;" or shame: "I am disgusting" or "I am inadequate."



NC can sometimes be irrationally negative without being self-critical or self-blaming when dealing with critical incidents. An irrationally negative statement for one aspect of unresolved critical incident could be: "I'm going to die." (self-referencing) or "It's horrible!" (just about the event). These critical incident cognitions reflect the "horrific moment" of past trauma locked in present time. (See Roger Solomon, Ph.D. Critical Incident Trauma, 1993, 1994, 1995 EMDR Conference). After processing these aspects of the trauma, there may be other self-referencing negative cognitions which emerge as also needing reprocessing such as "I should have done something more." When not dealing with the horrific moment of a critical incident, negative cognitions are most effective when they are self-referencing.



Note the NC is a continuing, presently held belief. The NC is derived from prominent or prototypical memories, and it continues to actively influence clients in the present. (Including during EMDR processing!)



Common errors: accepting a negative description of circumstances, events or others; accepting a NC localized only in the past.



When the client offers a description, incorporate the description into the eliciting question. "As you think about [description], what do you think about yourself?" Or "What does that say about you?" "It probably starts with 'I am'."



The second criterion of a NC is it accurately focus the client's presenting issue. Clients should indicate accord when the therapist restates the NC. The clinician often must listen carefully and go through several steps to determine the client's core issue.



The third criterion of a NC is an affective resonance. When saying or hearing the NC, the client shows or feels more shame, fear, anger, or other emotional arousal.



The fourth criterion of a NC is that it is generalizable to other, related areas of concern. Belief in this NC would affect the client in different settings or times up to and including the present. The most common error related to this criterion is a NC too specific to the incident. As in: "I should have locked the door." Clearer is: "I didn't do everything I could have." or "It's my fault."

Phrases Helpful in Eliciting the Positive Cognitions

To elicit the PC ask: "What words would you rather have when you think about that incident?" or "How would you rather think about yourself now when you think of that incident (or memory)?" For the same reason, don't ask: "How would you rather feel about yourself?" When this does not suffice, other frequently useful follow up phrases are: "What does that mean about you?" and "Could you express that as an 'I am' statement?"

Selection Criteria for Positive Cognitions

The first criterion of the PC is that it affirms a positive, self-referencing belief. The PC is a self-validating belief generally reflecting self-confidence or self-acceptance. However, not all PCs are self-referencing. A PC may indicate safety or transformation of a horrific present to the remembered past - "I'm safe now." "It's over." or the attribution of responsibility - "She/he had a problem giving love." This second type of PC is often combined with or followed by a self-referencing PC: "I am deserving." "I'm strong now."



A common error is selecting a PC which merely negates a negative belief - "I am not disgusting anymore." In EMDR, we do not accept this "denial" form of a PC. We look for a PC that organizes thinking, feeling and behavior in new ways: "I am fine as I am."



Another common error is selecting a PC that contains magical thinking about changing actual past events or other's actual attributes. Context = "Dad abused me." NC = "I am unlovable." Magical PC = "My father loves me." Appropriate PC = "I am lovable now."



The second criterion of the PC is it accurately focus the client's desired direction of change. A common error is pushing for a PC that is too big a step for the client's present frame of reference. The most positive PC may not seem possible. So, a diluted version in the desired direction may be selected initially. "I am learning to love myself now." After reprocessing, the stronger form of the PC may be accepted. "I love myself now."



The third criterion of the PC is that it is generalizable to other, possibly related loci of concern. Shorter PCs of three to eight words are often best. "I am competent." "I learned from it." "I'm free."



The fourth criterion of the PC is a positive, affective resonance. Initially, this may be hope, embarrassment or a doubtful acknowledgement of the PCs desirability. After reprocessing, the PC usually resonates with a stronger, more positive affect.



When you are struggling to have clients generate cognitions offer a possibility or use a pre-written list of metacognitions.



When clients have excessive difficulty generating the negative and positive cognitions, it's ok to offer a possible cognition to see if it fits for them or to ask them to review a pre-written list. Different clinical applications of EMDR can benefit from alternate sets of cognitions. Clinicians should generate their own lists appropriate to their clinical populations. As a reference for starting such a list see Landry Wildwind's article "Creating Positive Cognitions" (EMDR Network Newsletter, December 1991, Vol. 1, Issue 2, p.11) or the list at the back of the 1995 EMDR Level I and Level II manuals.

Be alert to unmatched negative and positive cognitions.

Finally be alert to unmatched negative and positive cognitions. There should be a thematic relationship between the negative and positive cognition. If the negative cognition deals with the issue of safety ("I am not safe."), the positive cognition should not be focused on the issue of competence ("I am competent.") When cognitions are unmatched, use your clinical judgment and/or ask the client to clarify the thematic issue to be addressed. Don't assume you have the wrong PC. Sometimes you will work backwards from a good PC to a good NC. Remember an appropriate NC will be linked into the disturbing affect. To help select the most appropriate cognition you can use a phrase such as: "Which words are most connected to the disturbing feelings you have when you think of the incident (state NC #1 i.e., 'I am not safe') or (state NC #2 i.e., 'I am inadequate.')"

Clinical Examples

All examples listed below are from the practicum section of EMDR trainings. They were chosen to illustrate typical issues that occur in case formulation and selecting cognitions.


Key. PNC = Preliminary Negative Cognition. PPC = Preliminary Positive Cognition.


SNC = Selected (before EM) Negative Cognition. SPC = Selected (before EM) PC.


FPC = Final (after desensitization) Positive Cognition. SEM = set(s) of eye movement.

When deciding on which of two Positive Cognitions to focus, let the client choose.

Example 1: The presenting issue was a painful memory of having polio as a little girl and not getting very many needs met by others. The initial issue focused on trust in others.


PNC: I can't trust anyone to take care of me.


PPC: I can take care of myself no matter what happens.

I questioned the cognitions based on a mismatch between PNC and PPC. The client stated her primary concern was personal efficacy.

SNC: I can't take care of myself.


SPC: I can take care of myself.



Comment: the mismatch suggested both trust in others (risking dependence) and trust in self (self-reliance) were issues. Trusting to depend on others (including spiritual trust) may be a deeper issue. Here, the client said self-efficacy was her preferred focus.

In chronic depression and character disorders, the clinician may offer a graduated positive cognition.

A more dilute PC will be more readily accepted and can help the client achieve a more noticeable shift in VoC.



Example 2: Presenting issue: "No one will help me through my financial and emotional problems." Hypothesized core issue as stated by the clinician: Dependent personality.
PNC: I am always a victim of uncaring people.


SNC: I can't take care of myself.


PPC: I can take care of myself.


SPC: I can learn to take care of my needs.


Comment: Due to a low VoC (1), the clinician offered a more graduated SPC (often appropriate with a character problem) which was more believable (VoC 3) to the client.

Be alert to magical thinking in the PC

A good PC will have a crisp sense of capturing positive core affect.


Example 3A: Presenting issue: Depression following a car accident 1 year ago.


PNC: Everybody is trying to control me. (anger)


Somebody is always trying to hold me down.


SNC: I am powerless. (shame) I am not good. (face flushed).


PPC: I'll be cheerful, not as worried about what I can't control.


SPC: I'm in charge of myself. I am able to be in charge of myself.


FPC: I can stand up for myself.



Comment: The PPC expresses some magical thinking and future orientation. The SNC avoids these problems. But notice the FPC. A good PC usually has that sense of crisply expressing a core affect, here self-efficacy and assertiveness.





Example 3B: Presenting Issue: A female professor's perceived gender discriminations in academic politics.



PNC: I don't deserve that.


PNC: If I were a boy it wouldn't happen.


PNC: It's not fair.


PNC: I don't have a choice because I'm a girl.


PNC: I'm not good enough.



SNC: I can't compete.


SPC: I'm capable.


FPC: I'm a contender.



This all woman practicum group knew the client's preliminary, other-directed negative cognition was not appropriate. Note the magical thinking in "If I were a boy, it wouldn't happen." The group patiently elaborated the selection of a self-referencing cognition. The selected positive cognition addresses the negative statement, but the final positive cognition which emerged during processing captures that crisp core affect which expressed the client's grit and strength.

Be alert to a tendency of the client and clinician to revert to magical thinking rather than accept a normal level of risk.

Example 4: A client's troubling memory spontaneously shifted from a "freeze frame" with a perceived lack of safety, to a resolved issue and a return to an appropriate sense of safety. 
At a Level II training, the client, of Asian descent, reported intense fear associated with the memory of an Ed Sullivan show in which the image of the atomic blast was shown.



1st PNC: This could happen to me and I couldn't do anything about it.


1st PPC: It will never happen.



2nd PNC: I felt powerless.


2nd PPC: Whatever happens I can handle it.



SNC: I'm not safe and my family is not safe.


SPC: I am safe and my family is safe.



VoC: 3-4. Emotion: fear. SUDS: 6. Location: neck, shoulders, chest, hard to breathe.



I assisted this group to select appropriate metacognitions. Both of the observers had trouble accepting the SPC because: "But she's not safe, really. It still could happen." I asked one observer when she was last aware of being afraid of an atomic blast? Did she live with a daily fear of nuclear holocaust? When she thought of the image of the atomic blast was she afraid? She said, "No." I pointed out that we all live with a certain healthy sense of safety in spite of all the risks we face every day, many of which are much more likely than atomic explosion. These include auto accidents, random violence, and cancer.



The client's physical sensations decreased over several sets of eye movement. As the sensations became weaker, the picture spontaneously changed to an image "before the explosion" with no distress associated with the picture. Returning to the target (in living room with family, image of atomic blast on TV) the client reported sadness and anger in the pit of the stomach. With further SEM, the anger subsided quickly and the sadness decreased gradually until it too cleared. After reaching a 0 SUDS, and doing the installation, the client reported, "I feel stronger" and had a pleasant feeling of tingling in the head and torso. This was further installed with SEM.



This issue for this client was the perceived sense of lack of safety with the focus on this single image of the atomic blast and rising mushroom cloud. The freeze frame spontaneously changed once the affect, expressed initially as fear and tension, was reprocessed. The issue of powerlessness could have been addressed, but it is more complex since there is little a small child (or even one adult) can do to prevent nuclear war. Safety seemed more directly linked to the issue since in fact the client, like the rest of us actually survived the cold war. The magical thinking of both the PPCs "It will never happen" and "Whatever happens I can handle it" showed how clinicians can get drawn into magical thinking when they forget that minimization can be a normal, healthy defense mechanism.

EMDR is a complex method. It is not just eye movements.

You can often achieve significant shifts through appropriate case formulation using the EMDR model.

Example 5: At a Level II training, the practice group lacked time to do a second full practice session so the decision was made to do only the assessment, without eye movements.

Presenting issue: A Jewish therapist described terror in groups. Initially there was no specific memory. When the student clinician was instructed to find a memory in which this dysfunctional response manifested itself, the client generated a memory from High School senior year being in an enormous gymnasium to register for college.



PNC: I am not good enough.


PPC: I am acceptable the way I am in groups.


VoC: -1. SUDS 10: terror.



The group was asked what red flags or clinical cautions, if any, they could identify in this initial work up. They could not identify any. The student clinician guessed wildly that perhaps the client was a survivor of ritual abuse??? What concerns, if any, would you have with this assessment?



The extremely low VoC, combined with a 10 SUDS in a functional therapist suggests this is not an appropriate NC. Also the level 10 SUDS suggests that a much earlier memory is likely involved.



As facilitator, I asked whether the response to groups went back to junior high school? The client didn't go to school much in junior high and mostly cut classes. I then asked the client if the response went back to grammar school? Yes. Did it go back to kindergarten?



At that point a memory emerged of being three or four years old. The family had moved into a tenement in an Irish, Puerto Rican, and black neighborhood in a dangerous part of NYC. They moved in the middle of the night because they were Jewish. They didn't want anyone to know they had moved in because they didn't fit in to the neighborhood. The client grew up with some family neglect, living on the streets and learning how to survive taunts and frequent beatings. At this point the selected NC emerges.



SNC: There's something wrong with me.


SPC: I'm fine as I am.


VoC: 3. SUDS: 8.



Commentary: The client was pleased to discover the source of the current social anxiety that had been never connected with the childhood experiences of being beaten and shamed by peers. The client felt empowered to address the present anxiety with a rational source for the formerly irrational distress. The case formulation and assessment did not resolve the problem, but it did start the reprocessing and make the issue more easily treatable.

How Much Farther Down the Road Can You See After Many SEM or a 'Completed Session'?

Example 6A: Helping the PC evolve near the end of the session with cognitive interweave. (This example courtesy of Roger Quillen, Ph.D.)


SNC: I have nothing of value to give.


SPC: I have something of value to give.


FPC: My feelings can flow freely.



At a Level II training the client demonstrated his presenting issue by being angry and resistant with his therapist in the practice session. After many SEM, issues with his mother emerged. It appeared he had developed anger, shame and inhibited expression to defend against her excessive intrusions. These issues were targeted and after many more SEM the SUDS were still not coming all the way down. Finally the facilitator asked "What would it be like if you could let go of this stiffness and resistance?" The client said, "Why then my feelings could flow freely!" He showed delight and surprise in his face at this thought. He was then asked to "Focus on that" and the session continued to completion.



Comment: The evolution of this client's PC shows the shift from a defended stance, in which his sense of self worth is still based on his sense of other's valuing what he has to give, to a more open stance in which his experience of himself is the value.



The client starts with the question of externalized self-worth: "Is what I have to give of value?" With the stimulus of the cognitive interweave he moved from valuing what he had to give, to how he had diminished his capacity to 'let his feelings flow freely' to defend against his mother's intrusions. Note how in this case the client was able to spontaneously generate a more appropriate PC after many SEM in response to a 'what if' cognitive interweave question.



Example 6B: Trust what you see in the client's face not just what the client reports.
At a Level I Training, the patient was a psychiatrist whose family had failed to provide as much support as she remembered needing when she developed a serious visual handicap in adolescence. The issues involved a sense of loss and perceived lack of family support in a currently high functioning individual.



PNC: I can't rely on others.


SNC: I'm helpless.


SPC: I can be strong.


Post-processing PC: I can get my needs met now.


As the facilitator, I gave the clinician the go ahead to work with the selected cognitions without knowing the background of the selected issue and in spite of the limitations of the SPC due some time pressure.



"I can be strong" [hopeful tone] is a weaker version of "I am strong." [definite statement] The work progressed well but with a mild sense of a 'mock' session. I noticed the patient seemed to have more affect than she reported to her clinician. When I returned from observing other groups they had finished treatment with 0 SUDS, 7 VoC and a "clear" body scan. I noted the client continued to have a trembling jaw, obvious tears and emotion in her face.



I reviewed the issues, the NC and PC with the clinician and the client, and then asked the client if she were willing to work a little more? "Yes." "Ok, then would you like to believe: 'I can get my needs met now'?" She immediately burst out crying. Holding Kleenex to her face she nodded that she would. I asked the clinician to continue the installation from that point. Within three SEM the client reported a VoC of 7 on the new PC. Her eyes looked much clearer and her face looked more hopeful and less clouded.



I pointed out that because family members were not there for her in her teens "being strong" was adaptive for her then. I noted that perhaps, being strong enough not to need others was not the issue which remained unresolved. As someone who had completed medical school with a visual handicap and established a private practice, she knew she could be self-reliantly strong. This had suggested to me perhaps the unresolved issue had more to do with being able to receive from others and permit herself the vulnerability that entails. Thus I suggested the modified PC: "I can get my needs met now." In fact she learned she was strong enough now to permit herself dependency feelings.



The issue in case formulation for generating the positive cognition is sometimes how to balance the perceptions of the client, which are often limited by "how far down the road they can see right now" and the perceptions of the clinician, who must assess the client's life issues and then focus on the metacognitions that can inform more options and freedom for the client's current life. After installation of the client's selected positive cognition, further installation of a more apt metacognition can sometimes allow the client to resolve a current version of the earlier issue.

Notes

[1] The theoretical foundation for this paper is Francine Shapiro's accelerated information processing model as described in her journal articles (1989a, 1989b), in her book (1995) and as presented at EMDR Level I and Level II trainings (1991-1995). The major selection criteria in this paper are elaborations of principles enumerated by Dr. Shapiro.



My first article on "Selection Criteria for Negative and Positive Cognitions in EMDR" was published in the Winter 1992 EMDR Newsletter . That article followed a facilitator training in September 1992 where presentations were made by Landry Wildwind, LCSW; Eirin Gould, MFCC; and myself. The staff asked me publish my handout in the newsletter. I am indebted to Landry Wildwind and Eirin Gould for the discussion during that initial presentation and for subsequent conversations and case presentations in which I benefited from their insights. Roger Solomon also made significant and very helpful contributions to my understanding of selection criteria when working with critical incidents.



Much of what I have learned about the problems clinicians encounter in EMDR case formulation and about how to clarify these issues is the result of supervising numerous EMDR training practice sessions and in helping to train new members of the facilitator staff. I remain grateful to the many practicum participants and facilitators-in-training who have shared their work and their questions with me.



[2] Morton Reiser (1990) discusses the nodal memory network model and its relationship to neural structures in Memory in Mind and Brain, What Dream Imagery Reveals which I reviewed along with Ernst Poppel's Mindworks in the May 1992 EMDR Newsletter.



[3] For an introduction to sensory and affective memory substrates in postttraumatic stress see Bessel A. van der Kolk (1994) "The Body Keeps the Score."

Bibliography

Reiser, M. (1990) Memory in Mind and Brain, What Dream Imagery Reveals, Basic Books.

Shapiro, F. (1989a) Efficacy of the eye movement desensitization method in the treatment of traumatic memories. Journal of Traumatic Stress, 2, 199-223.

Shapiro, F. (1989b). Eye movement desensitization procedure: a new treatment for post traumatic stress disorder. Journal of Behavior Therapy and Experimental Psychiatry, 20, 3, 211-217.

Shapiro, F. (1993) The Maintenance of EMDR Treatment Effects, EMDR Level I Basic Workshop Manual, p4.

Shapiro, F. (1995) Eye Movement Desensitization and Reprocessing, Basic Principles, Protocols and Procedures, The Guilford Press, New York.

van der Kolk, B. (1994) The Body Keeps the Score: Memory and the Evolving Psychobiology of Posttraumatic Stress, Harvard Review of Psychiatry, January/February, 253-265.