Mardi Horowitz on Psychotherapy Research and Happiness

Mardi Horowitz on Psychotherapy Research and Happiness

by Victor Yalom and Rebecca Aponte
Mardi Horowitz discusses his research on psychotherapy for stress and trauma, his recent book on happiness, and what therapists can teach their clients about attaining it.


Get Endless Inspiration and
Insight from Master Therapists,
Members-Only Content & More


The Interview

Victor Yalom: You had the audacity to write a book entitled A Course in Happiness. I guess this begs the question: as a psychiatrist and therapist, do you really know something about happiness that's teachable?
Mardi Horowitz: I think so. And it took me a few decades to feel that that was the case.
VY: Say more.
MH: Well, I have always had a philosophical bent; I studied Zen Buddhism in my early 20's.
VY: Before it was fashionable.
MH: Well, I think that was the start of the fashion--not with me, but with my teachers.
VY: I guess it's been fashionable for thousands of years, but before it was fashionable in mainstream psychology.
MH: Then Suzuki and Erich Fromm wrote a book on psychoanalysis and Zen. I was also reading Freud at the time—I was reading Freud in high school—so my professors really directed me to the big questions of the human predicament. I'd also always been struck by the line in the Declaration of Independence: "the pursuit of happiness." I'd seen an earlier copy in Washington, D.C., and it said "the right to happiness." There's a little insertion there—probably it was Thomas Jefferson—"the pursuit of happiness." And I sort of pondered that: Well, how do you pursue it? That is, you can't have it—that was the idea. It was the journey, rather than the arrival, that might give you contentment.

That notion persists in my use of the word "course" in A Course in Happiness. It means two things. One: navigating. I'm a sailor, and the practice of sailing teaches you very quickly that you can't sail into the wind, even if that's where you want to go. So if you want to go to San Francisco from Sausalito, you have to hit the winds coming from San Francisco, which, fortunately, it rarely does. You can't just point to the Trans-America Pyramid to get there. You have to go back and forth. But you need to chart your course so you get there with the most economical and speedy means.

The second meaning of "course" is a course that's full of lesson plans and teaching points. My years professing and being a bit of a pedant, I think, have a practical payoff in that I know how psychotherapy trainees learn. And I think those lessons for psychotherapy clinicians, and those lessons learned by psychotherapy patients over a period of time, can be translated so that people can use them on their own if they have the motivation—hence A Course in Happiness.

VY: You're a psychiatrist by training as well as a researcher, but also a therapist. We therapists tend to think we know techniques to help people explore things and understand themselves better, but I'm not sure we're all on board with the idea that we actually have content to teach them.
MH: Yes. I'd say that's been the topic of my clinical research for my career—content can be determined using empirical research. For instance, my 1976 book, Stress Response Syndromes, laid out the information-processing model that then defined the symptoms that became the criteria for PTSD. It wasn't that people didn't know about those symptoms, but there were a variety of conflicting theories of what caused the symptoms. And by doing clinical, field and experimental studies, we could nail it down enough to settle the controversies.

So I think, by using empirical work, we can find that working clinicians agree on how contents change—that's the critical thing. How does the mind's narrative about self and others, for example, change in therapy so the person's able to make more reasonable plans?

That's not how psychotherapists are taught, however, and it took a few decades for me to learn how people learn to be psychotherapists. For example, a young teacher who's really bright and a good clinician will come in and tend to teach theory. Then the trainees will complain because they're not emotionally ready for the theory of how things work. They want to know, how do they even survive with their cases? They want to know how to do it right away. So I think we have to go with what people are motivated to learn. The first thing we teach people so they're less frightened when they're doing therapy—which is scary at first, as you know—is, "Borrow from me these techniques, these rules of thumb. Later on, I'll tell you why you don't always use this rule of thumb, and when this technique can be harmful, or at least not helpful." Then, after a year or two, when they feel comfortable, you can start teaching them how people change.

There tends to be a Y in the road because some therapists feel so confident in themselves, once they're able to establish a trusting, calm relationship with disturbed people, that they just go and do it by intuition. And their patients get better, so they have feedback that they're doing a good job. But they don't understand what's possible for the person.

That's where the content comes in: what are change processes? For example, grieving is a change process that occurs on both conscious and unconscious levels, to change the narrative of life so the person can accept a loss and move on.

Defining Happiness

Rebecca Aponte: Getting back to happiness, how do you define this? What is your definition of happiness as something we could train people toward?
Mardi Horowitz: Very often, the really big concepts that have been around since words were first written on tablets are very hard to define. Justice, truth, happiness are those kinds of words. So it has to be kind of broken down into its components. The components that I deal with in A Course in Happiness are pretty long-range components like contentment, satisfaction with yourself articulated in your life—rather than joy, which might be when you open a birthday present and it's what you wanted.
VY: So that's shorter term.
MH: That's pretty short term. You can say, "My dog is happy if I give him a bone," but it's a state of mind rather than an enduring life skill.
RA: I see.
VY: Martin Seligman takes the stance that, as therapists and psychologists and psychiatrists, we've tended to focus over the years on psychopathology, on the negative emotions—stress, anxiety, depression, and the like—and the assumption was that if you get rid of the negative emotions, what you're left with is happiness. He's taken the stand that that's actually not the case—that's really more like neutrality—and happiness, as he's researched in positive psychology, is a whole other set of things. I'm wondering what your stance is on that.
MH: Well, A Course in Happiness is, in a way, taking that stance and going pretty well beyond it. I think the stance is correct as far as it goes, like Norman Vincent Peale's The Power of Positive Thinking. There is the power of positive thinking, and I think the positive psychology theory, like evolutionary psychology and self-psychology, are all really excellent additions to theory. But it's very hard for people to inhibit attention to negative topics. That's the essence of the critical symptoms for PTSD that we have studied experimentally as well as in clinical subjects, which is that they have intrusive thoughts. So you can say, "Don't have intrusive thoughts." And, as you know from other research, that tends to increase them rather than decrease them. So a big message in A Course in Happiness is to pay attention to where you're paying attention, and that there's a lot of work in addition to focusing on having more positive experiences—for example, developing more reflective self-consciousness and reducing harsh self-criticism, a source of negative feelings.
But you don't get into positive emotions by telling yourself that you're just a grand, creative wonderful person who's always compassionate, when that isn't true.
But you don't get into positive emotions by telling yourself that you're just a grand, creative wonderful person who's always compassionate, when that isn't true.
RA: Right.
MH: Reality is the enemy of an enduringly positive frame of mind. The Dalai Lama's Art of Happiness, Seligman's research in positive psychology, or Daniel Gilbert's book Stumbling on Happiness—I think it's really good research, and it's really good philosophy, and it's really good spirituality. But along with being positive and doing all the things that are in those writings, people also have to review memories of traumatic experiences. They have to recover from losses. They have to encounter grievances that have endured since childhood and given them a chip on the shoulder. They can, in a realistic way, focus their attention on positive things. That's good. But they have to have times when they focus their attention on the negative things in the right state of mind—calm, often alone, maybe with a trusted confidante—and then review these memories so as to bring their life narratives into more harmony with what's approaching in the near future, so they have plans. So A Course in Happiness deals with a systematic approach to that, derived from our studies of change processes in psychotherapy.

An Integrative Approach to Case Formulation

MH: One of the things in psychotherapy that our group has done is we've developed an empirical basis of case formulation, which allows an integration across different brand names in psychotherapy.
VY: Now, case formulation is an old concept, but I think you have a particular way of approaching that.
MH: Yes—standing on the shoulders of not only the old psychoanalytic and psychodynamic concepts, but also of people like Aaron Beck and Albert Ellis and Bugental, who were taking out of the 1960's psychoanalytic mode of formulation those things that were changeable. I don't think they disrespected the idea of unconscious dynamics, but they were saying, "Well, what can change?" If we really clarify it, change is going to take place through the use of consciousness as a tool.

We know from psychotherapy research that the relationship is the most important factor, but in our research studies we examined some additional variables.
The reason that specific techniques didn't seem to come out in many different psychotherapy studies is that it's different strokes for different folks
The reason that specific techniques didn't seem to come out in many different psychotherapy studies is that it's different strokes for different folks--a technique that's focusing on deeper emotional values may be good for some people, but actually may be even harmful and disorganizing for other people. If you don't get into the dispositional variables, then you get a washout.
VY: It seems like you always hear those questions in research: what techniques are good for what clients in what circumstances? But you never really hear the answers to that. You always hear, "It would be good if we could tailor treatment to people, but..." You hear things like, "CBT is good for depression." But then you look at studies that say it's no better than anything else.
MH: That doesn't mean it's not effective.
VY: Sure.
MH: And there's a huge fallacy out in the field that people don't even acknowledge. Once I say what it is, everyone will say, "No, no, no, no, no, of course we don't believe that." But there still seems to be a prevailing fallacy, which is that more studies of effectiveness means the therapy is more effective. It's simply not true. I mean, everyone knows that's not true. Psychotherapy has been very well established to be effective in general. But that doesn't mean it's effective for every case, and certainly we see negative therapeutic outcomes in some people. Some people start psychotherapy and you end up having to hospitalize them. So there's a lot to the technique; it's not that they have a toxic therapist.

A Case Study: Clone One and Clone Two

VY: Can you give an example of how a case formulation for a specific client may give an indication of certain techniques or approaches for them?
MH: Actually, right now I'm writing a paper for the American Journal of Psychiatry on exactly this topic.
VY: Okay, great. Good timing.
MH: So I'll give you the case example. It's a young woman whose mother has recently died. But the patient is in her 20's—she's been very dependent on her mother for guidance. She would probably diagnostically fit into a category of major depressive disorder a year after her mother's death, along with dependent personality disorder. So let's say she's put into therapy. It would be a focal therapy aimed at her in relation to her mother's death, and why she was not depressed beforehand, and why she's now depressed. Let's say she goes into therapy with a female therapist of an older, warm, trustworthy nature. So she sort of has a replacement, and her symptoms get a little better right away. But she comes in and starts expecting guidance from the therapist on what her decisions should be. And let's just leave out the issue of antidepressants and overmedication, which tends to occur with the simple cases.

Now, the therapy techniques that would be optimum for this patient will focus on helping her stabilize her states of mind, develop new relationships, modify her sense of identity, and develop better plans for the near future. This is kind of simple and obvious. That's what the patient would say she wanted, if she could articulate it.

Now, in the condensed, teaching form of this article, I start with Clone One and then go on to Clone Two of this exact story.
VY: What do you mean?
MH: Clone One is the person who, before the death of the mother, had a relatively coherent and well-developed sense of identity, but had role relationship models requiring guidance from her mother. She'd grown up in that container, but now the death has occurred and the container is broken. She feels more fragile, has a regression, and hasn't replaced those functions either by her own growth or in relationship to another person.

Now, let's say the techniques in Clone One's case are successful: they involve just being clear that that's her life story in a way; that she has, for the time being, the safety of a container with a good therapist; that in this container she's going to work through any sense that she's been shattered or abandoned; and that she's going to be helped to develop near-future plans in being more assertive, going out and forming relationships, and not being so frightened, hopeless and helpless. She gets better and lives happily ever after, because those techniques were very helpful and just what she needed, from just the right person, at just the right age milestone for that kind of development. So she's gone through a maturational path. And those techniques tend to be pretty interpersonal in discussion; we're looking at the repetitive, maladaptive interpersonal patterns, like excessively needing guidance from another person, being exploited by another person because she's seen as a sucker, and so on.
RA: Right, she's sort of handing over control.
MH: She's handing over control and someone says, "Okay, you do this and this and this and this for me, and I'll tell you what to eat for dinner."

On to Clone Two: this patient has not had a chance in her previous development to develop a coherent self-organization, so she has dissociative fragments of identity—not only in conflict, but segregated in terms of memories. She may even have different memories of a relationship with her mother in different states of mind. So when the therapist is interpreting something in one state of mind, the patient may shift to another state of mind and be misinterpreting the interpretations.

States of Mind

VY: You refer to this idea of states of mind a lot. Can you briefly state what you mean by that?
MH: States of mind is one of the big concepts I refer to in formulation. And the reason for it has to do, again with the training of psychotherapists, which in the last 25 years has emphasized diagnosis.
Diagnosis is an actuarial thing: it's good for accountants and insurance companies, and for questions like, "How many cases of schizophrenia did we have in Africa in the last decade, and how does that compare with China
Diagnosis is an actuarial thing: it's good for accountants and insurance companies, and for questions like, "How many cases of schizophrenia did we have in Africa in the last decade, and how does that compare with China, and what does that indicate about..."

Also, diagnosis stemmed out of research: the DSM in 1980 was a drastic revision saying, "Okay, we don't have a theory of mental disorder and what causes symptoms, so let's just describe it."
VY: "Let's just categorize the symptoms."
MH: "Let's categorize by what we can find out in maybe a half-hour interview." So that's all that is, but of course the students think it's something real. I was on the committee for PTSD , anxiety disorders, and borderline, narcissistic, and histrionic personality disorders. And I'm the world expert on at least two of those things. They're my criteria—they're the best I could do at the time—but they're not etiological entities, and they're treated as if they were.

And the worst thing about the use of our product in making DSM III and then IV, and now V—the same arguments, by the way, are taking place—is that they're committee judgments. The committee knew there was a dilemma. Ultimately it came down on static descriptions, in part for some forensic reasons. So now you have to have five of these eight depressive symptoms for three months in order to qualify for major depressive disorder—something like that.

But if you have the passionate aim of teaching therapists, then after you say, "Here are the diagnoses, here are the rules of thumb," you have to say, "Now let's go back to the symptoms. What causes each symptom? Where do those different causes converge? And of those causes, where can we change things?"

So the states-of-mind concept was a way of dislodging the rigidity of static memorization of the diagnostic criteria. The idea is that
people aren't always despairing if they have despair. Or if they have a phobic symptom, they may not always be phobic.
people aren't always despairing if they have despair. Or if they have a phobic symptom, they may not always be phobic.
VY: Or dysthymia says you're kind of blue most of the time, more days than not—so you can be quite depressive, but not blue all day long.
MH: Right. What are your other states of mind? And then the critical issue around states of mind is: how much in control are they?

The Dissociative Patient

RA: Right—which goes back to your second subject, the dissociative woman.
MH: Right. She was not in conscious control of where she was focusing her attention, nor was the therapist of the second woman able to draw her attention and keep it in a state of mind. She was flip-flopping in different states of mind.
RA: Was the therapist able to see it, at least?
MH: Well, with my fictional therapist and for the journal article, of course! But she uses a different technique from the first case. She observes that there are shifts in states of mind, and that this person is a very dysregulatory one, and begins to say, "Now, what's happening here?" Then the technique shifts more to helping the patient focus attention on her sense of self, her bodily self, her sense of self in the room with the other person, her sense of what was happening, and learning a kind of reflectiveness on these things that the person had not acquired before. And developing that skill helped the patient get a sense of pride that they were able to do that. So it's a different set of techniques.
VY: So in the second case, it's much less focused on the disruption from the death of her mother. You deal primarily with the organization of her self that was a problem beforehand, but was exacerbated when her mother left the picture.
MH: Exactly right. So instead of coming back relatively swiftly from her regression to where she'd been in terms of her identity structure, in Clone Two it's going to be a longer therapy and a larger growth, ending up maybe five years later, where Clone One and Clone Two can sort of converge—they both have the capacity for intimacy, for interdependence rather than dependence, and they have integrity as well as control over their states of mind to a larger extent. But it may take longer and require different techniques—not totally different, because they overlap to some extent.

Configurational Analysis

VY: How do you teach your method of case formulation to psychotherapy trainees?
MH: For some reason, early trainees often come in with a kind of pseudo-psychoanalytic, excessively deep idea of what formulation is, and it's all based on projecting theory into whatever clinical material comes into the room. And it's often whatever theory they read that they thought applied to themselves. So they say, "Oh, this is what it all is," and then they just see this everywhere. Like spots in the visual field, they're illusions about patients. In fact, even seeing experienced therapists on videotapes with different cases, you sometimes see what I would frankly call errors, because they're applying the same segment of theory to every case.

So I developed a system called configurational analysis—which is based on four formal categories or levels of formulation—in part to help both students and colleagues think about cases. Here are the categories. One: Just describe what you observe, and select the phenomena you're trying to explain. Not everything—it could be one, two, or three symptoms, for example.
VY: So depression, anxiety, or disorganization, something like that.
MH: Right, exactly. So if the phenomenon one's trying to explain is depression, the second category is: what are the states of mind? What do you mean by depression? You're saying the person has the same prevailing mood that, if you were to generalize, is "depressed for weeks." What are the person's states of mind? The person may have the state of mind of piercing sorrow with pangs of yearning, and illusions that a divorced person is now coming back into the door.
VY: Much more specific descriptors of how the client experiences depression in that moment.
MH: Right. So that might be a state. It would probably be only a minute or two. And it might uncontrolled, too; it might be undergoverned. Then the person might have a state of kind of apathetic boredom with some tinge of restlessness and aimlessness, and feeling just kind of gray. And they might be able to rouse themselves from that, so it's a little bit more in control. Then they might have a state of agitated, restless urgency in which they engage in frenetic and fruitless activities. They might also have a state of irritation and anger. And then they might have a state of relative repose.
VY: And they might have several hours a day where they're at their job and be very competent and feeling good about themselves.
MH: Right. And then you say, How do they shift in cycles of these states?

What triggers each state? "Well, when I get absorbed in my work, I get into a state of relative less-depression." What triggers the pining and yearning? And so on. So it's only one level down, but it's still observational.

What's more, you can share this language with the patient, so the patient can begin to examine their states of mind and look for the triggers, just like in positive psychology. You can say, "Well, how can I feel a little bit better right now? Maybe instead of criticizing myself for being lazy and having screwed up all my relationships, I should look at my achievements: I've done the architectural plans for three new buildings. I've made a living somehow. I've not gotten in car accidents. I'm taking care of my parents"—or whatever the person might say. So that's states of mind.

And even at the states level, you get a psychodynamic configuration right away with the patients. "What states are you frightened of entering that you can't prevent yourself from? What states would you like to enter and can't get into? And what states are you using to avoid the dilemma of trying to get into a good state but then you're afraid of a bad state?" So, you might hear, "I don't ask people out for coffee because they might reject me." You're then getting into the next level of formulation, which is: what are the themes that are related to these state transitions? And the themes are certain topics like, "Do people like me?"
VY: Fear of rejection.
MH: Yeah, and so forth and so on. So the topic might be impoverished relationships. And when they're on this topic, does that trigger them getting into the sorrowful state when they're thinking about a lost relationship, and a hopeless state when they're thinking about the possibility of avoiding rejection because they've been repeatedly rejected? Then, also, when you're talking about these topics, that's where you get into content: What are the topics of concern? What's unresolved? People may have big events but they've sort of reached resolution on them, so you don't talk necessarily about the biggest event. You may be talking about some little, trivial insult.
VY: Okay, so just clarify the third box again, it's...
MH: It's the topics of concern. And it's what operations the person's deploying in order not to progress adaptively to a resolution on a topic. What are the obstacles to actually thinking that through in a realistic way and making good plans for the near future? So it's looking at what, in psychodynamics, would ordinarily be called defenses. But all therapy models recognize obstacles. A person paradoxically wants to inhibit, avoid or distort the very topic they're there to discuss. Once you recognize how are they doing that, then that's where a therapy technique will be deployed.

But the question will be, what will happen if you counteract their inattention and focus attention?
What therapists do, mostly, is tell patients where to pay attention.
What therapists do, mostly, is tell patients where to pay attention. And part of that is paying attention to their own attention, so this system of formulation helps. Really, micromoments of therapy decide what to do next, once the person has learned it.

But the fourth level is often what beginning therapists plunge down to with their theory prematurely, which is the self-and-other configurations. That's why this system of formulation is called configurational analysis: it gets down to the level of the self-and-other attitudes and beliefs, but then organizes state of mind. So when you have a patient who's flip-flopping to different states of mind, even in the relationship with you as the therapist, you often can then see, once you're looking at it, the difference of states, the different topics, the obstacles. You often can say, "Ah, here is a recurrent attitude—the patient's flip-flopping. Either they're the aggressor and I'm the victim, or I'm the aggressor and they're the victim." Once you see these role relationship models and each person as having a repertoire of role relationship models, of different self-images, then you can see a recurrent pattern.

On each of these levels, we've shown that you can get empirical, reliable, and valid predictive agreement between clinicians if you define the labels—so configurational analysis is an empirically based system of case formulation. It is psychodynamic in that it deals with wish, fear, defense, unconscious processes and stuff, but it's integrative in that you could take a cognitive behavior therapy clinician and see if they formulate their cases this way (we just published a paper on this; they do), if you enable them with a system. They're making the same observations. And the systems of cognitive behavioral formulation and configurational analysis and psychodynamic—they're all containable under the circus tent of these formal properties. But the stories they focus on tend to be different.

Focusing on Now

RA: How has all your research influenced or informed the way you think about happiness and about how happiness can be attained?
MH: Over my lifetime as a psychoanalytic psychotherapist, I shifted from what I was taught to focus on—which was mostly the developmental past and how it led to the character of a person, including character distortions and layers of the onion and that sort of thing—to seeing that as being important only if it's related to the near future. So my time frame as a therapist is: What's going to happen in the next minute with me? What's going to happen in 10 minutes? What's going to happen in two or three weeks with this patient? And what's going to happen to this patient over the next year or two? That's why the focus is on what can change. The questions in my mind, using the states of mind and other concepts, is: what's happening right now?

So the patient's telling me some story about some grievance that they have or a stressor event that's coming up that they're trying to prepare for, and I'm listening for how they're processing it in their mind.
I'm paying attention to the transactional music between us: are we getting closer, or are we in union of some sort, or are we getting further away?
I'm paying attention to the transactional music between us: are we getting closer, or are we in union of some sort, or are we getting further away? What's the state of mind of us as a pair? What's the state of mind of the patient? What's my state of mind? Am I getting bored? Why am I getting bored? Am I getting scared? Why am I getting scared? If I'm getting eager to make an interpretation, why am I so eager? Should I keep my mouth shut? Should I open it up? Should I be intuitive? Should I not? So I'm thinking about those things. But I'm also going to the past if it's going to help us understand why the patient's about to make the same mistake again.
VY: If you think that's going to be helpful to them.
MH: If I think that's going to be helpful. Because I'm thinking, how can this patient change?

A Calm, Rational Approach

VY: Some patients who come into outpatient therapy are already very intellectualized and use intellectualization as a defense. I notice your work tends to take a fairly intellectual approach to analyzing everything. In the Course in Happiness, for instance, you advise a lot to people kind of step back a bit and take a look at their life and make some rational decisions. But I'm wondering, with some patients who are already trapped by their own overrationalization, whether...
MH: Yes, but often you find with the kind of patient you're talking about—it really is a very common obstacle—the person says, "Life is so full of predicaments," or, "How does this relate to what Nietzsche said in Fundamentals?" And of course, that's getting away from the heart of the matter. So with different patients, I might say different things. To one patient, I might say, "What do you think's happening between us?" Or to another patient I might say, "Seems to me this isn't the heart of the matter. We're talking about your decision whether to quit school or stick with your very delayed graduate thesis, which I know makes you feel either ashamed or scared and confused. And here you're talking about... What do you think's happening here?" And the patient would say, You know, it is a little scary," or "I'm a little confused." And I may say, "I am too, on your behalf!" That's what I mean by focusing attention.

Also, there's a difference between what I'm encouraging the reader to learn to do in A Course in Happiness and what the reader's going to do. I'm calm about the reader's pain. And I'm trying to say, "Try and be as calm as you can, which doesn't mean go write a philosophical essay on your predicament. Try and be as calm as you can, and allow yourself, in a safe moment, to consider your emotional distress." That's the difference between A Course in Happiness, which takes on a stress mastery approach, and a book on happiness that says, "Don't worry, just be happy"—like the Bobby McFerrin song.

I say worry, but have productive worry, and learn to stop worrying when it's not productive.
I say worry, but have productive worry, and learn to stop worrying when it's not productive. That would mean paying attention to states of mind. Is your state of worrying like going through the rosary beads of your worries? Are you repeating it and repeating it and repeating it, which is only etching in a source of negative feelings? Or can you get into a different state of mind where you're able to look at this catastrophic view of your life, and you're able to look at your excessive feelings of entitlement and expectation that life will shower you with an ever-expanding stock market? And can you get in a state of mind where you can begin to realistically look at it between these two extremes? I'm saying, "Don't avoid these things, but have tolerance for the negative feelings. Feel your feelings." But you don't get through mourning by crying ten thousand tears.
VY: But if you don't shed any tears, that's usually a problem.
MH: And you're going to cry, or feel like crying, when you examine some of the aspects of what you lost that got you into this stressful thing. But you have to tolerate it. The point is not only to feel anger or sorrow or shame or guilt or fear or all the negative feelings. Your aim is not to be so frightened of them, so that you can use consciousness for what it's really best at: it's a special tool for resolving problems. If it ain't a problem, we don't have to be too conscious of it. It's like driving a familiar route—you sort of find you got there and you didn't remember, "Turn left and turn right and turn left. Watch out for cars." That's automatic after you learn to drive.
VY: But if you spent hours driving circles getting lost, that's the time to pull over and look at the map or GPS and chart a new course.
MH: Right. And sometimes you have to note when the GPS is wrong and you have to pay attention, yourself.

Research on Stress and Trauma

VY: I want to shift gears a bit. You've spent a great deal of your career researching stress and trauma. What got you interested in that?
MH: Well, I had my own traumatic experiences, which I remembered more and more as I began to study trauma. But what really got me started was dissatisfaction with the theory I was taught as a psychiatric resident. I kept asking my teachers, "What's the evidence for that?" I didn't want randomized clinical trials. What I wanted was to have them tell me a case where they saw that to be true, and what they observed, and what made them think that was what was going on.
VY: What were you taught that didn't make sense to you?
MH: I was taught standard ego psychology and psychoanalysis, and the emphasis was on people who were repeating aspects of an Oedipus complex. Now, I had cases and I saw them pretty frequently, and I listened very carefully, I think. It's not that I didn't see any cases with triadic conflicts—it's that I saw a lot of other stuff too. I said, "Well, what about this, what about that?" And they kept saying, "Pay attention to the Oedipus complex. Interpret defense, interpret defense, interpret defense." It wasn't wrong; it just wasn't complete. It seemed to be applied by my supervisors to some cases where, in retrospect, I would say, for example, they had borderline personality disorder, and that caused fundamental distrust in the transference—not necessarily competitive rivalry.
VY: So when you were taught, psychoanalysis was still the dominant model.
MH: Back in the ‘60's.
VY: Right. And it was before the pendulum swung in psychiatry to be all about the brain and medication.
MH: Right.
Now we're in the decade of the brain, which seems to have gone on for 30 years!
Now we're in the decade of the brain, which seems to have gone on for 30 years!

One of my colleagues calls me an in-betweener: I don't seem to accept the biological approach and I don't accept the psychological approach. Well, I'm a scientist. I'm a scientist, physician, clinician, psychiatrist—I want to understand how it works. And it doesn't work just biologically, and it doesn't work just psychologically, and it doesn't work just socially. It's an interaction of complex patterns, and we need research methods that focus on complex patterns. That means an uphill fight with study sections that give grants, because they want homogeneous groups by diagnoses. And since I contribute to the diagnoses, I'm entitled to say they're too static. I'm trying to work to redefine post-traumatic stress disorder, even though the criteria are right out of my book on stress response syndromes. And I'm at work to see us go beyond brand names in psychotherapy towards an integrative approach, which I've tried to simplify in my books States of Mind, Understanding Psychotherapy Change, and Cognitive Psychodynamics. But economics is what drives a lot of the field. So it's big pharma; it's simplified randomized clinical trials with very simple, cheap, inexpensive treatments that can be done by people who don't have much training.
VY: This is good to hear from an insider, from a psychiatrist who's done a lot of research.
RA: Yes, it is.
MH: Yeah. Psychiatry is a complex field. And there was that big hope for a single gene for every major mental disorder.
VY: It's always on the first page when they find it, and then six or nine months later there's a little article on page 20 that says that the gene for schizophrenia or alcoholism wasn't confirmed. "The Norwegians weren't able to replicate the study…."
MH: Right. And negative studies, even those little paragraphs, are usually rejected. It's very hard to get a negative study published. Everyone likes positive studies. It's understandable because everyone wants solutions to really big problems. But the big problems are complex, so we probably need a methodology that deals with the interplay of five or six variables, not the correlation between two variables. But if you want your PhD, you'd better correlate two variables, because you'll get it done.
VY: It already takes long enough to get a PhD. We obviously don't have time to even scratch the surface on all your research, but what are a few of your findings on stress and trauma over the years that have really stood out?
MH: Well, I think the information-processing model really holds up for stress and trauma, which is that the catastrophic event, in a way, shatters expectations. If we were all like good boy scouts, truly prepared, we would just enjoy stressors like a rough and tumble game, because we knew what to do. When we're tackled in football, or a fly ball is coming to us in baseball, we know how to handle that. We may lose, but we aren't traumatized by the loss. But an unexpected event, or even an expected event—to the extent that any expected event still has unexpected aspects—leaves an active memory in mind that is stored and has to be processed, and will come back intrusively, even if we don't want it to be processed.

The interesting thing in starting to focus on intrusive thinking is: when does it occur? I would get calls from mental health professionals who'd say, "You're an expert on trauma. I was just in an automobile accident and a passenger was injured, and it's three days later. I'm not upset. Is that okay?"
VY: And what would you say to them?
MH: I'd say, "Too bad you asked, because the fact that you're troubling to call me up and ask means you have an intuitive sense it's not processed yet. Just wait. But don't then be frightened that you're going crazy when all of a sudden, three months from now, you have a bad dream. Very often, paradoxically, you start processing a difficult experience you've had only when you feel safe. You're too close to the accident to feel safe, so you are restoring your equilibrium by waiting. But it's still there, it's in your mind, it's unconscious, and it will come back to you when you're ready. And if you have trouble with it, call me again. But, in other words, it's not abnormal to know you're in denial and numbing, which is why you're calling. If you were really okay, you wouldn't call."
VY: So your advice might be, "Wait, and when it's a problem, that's the time to deal with it"—not to rush in with the critical incident stress debriefing and have everyone talk about something they experienced, whether they want to or not.
MH: Right. Well, critical incident stress debriefing was really oversold, as are certain other techniques. And the word I want to emphasize is "sold." It's the economic driver that makes people want to stay within their brand names of psychotherapy, because that's how they think they're going to attract patients—because they've got the gold dealie that says, "I trained in, you-name-it, ear-twitching therapy." And probably almost anything can be helpful. In fact, therapists wouldn't do it if they didn't know it was helpful.
VY: For some people, sometimes.
MH: For some people sometimes. But they don't want to leave their economic niche until there are no patients for it.
VY: Right! Who does?
MH: Exactly.
VY: You've done research for decades on this topic. Were there any findings that surprised you or were counterintuitive, or that therapists, don't know or get about stress and trauma?
MH: I think clinicians tend to underemphasize the patient's potential for growth. And the growth is going to be in terms of identity coherence and harmony. So when a person is coming out of a loss—the loss of a job or home, for example—they have to work through the meaning of that loss to themselves and their loved ones. That's top priority. They have to sustain the negative feelings. And there are sources of positive feeling that they can get, like pride and the respect of others, for handling a loss with courage and stamina—and that, itself, can change negative attitudes about identity. So instead of the person feeling, "This happened to me because I'm so worthless, or I'm so incompetent, or because I can't cope, or because I'm dependent," they can now feel, "I'm a human being. I got through this dark passage. This is a sign of real, authentic strength. I made some poor decisions, but then, who am I to predict the future? If I made a poor decision, it doesn't mean that what Uncle Charlie said about me being so stupid is how I need to see myself."
VY: So one thing is to see stress or trauma as a potential for growth; the goal is not just to return to baseline.
MH: Right.

Where Therapists Get Stuck

VY: You run a second-opinion clinic for psychotherapists, where therapists bring cases that they are feeling stuck with. Obviously every case is different, but in terms of dealing with stress or trauma, are there ways that you see clinicians get stuck or make mistakes, other than not seeing the potential for growth?
MH: Clinicians get stuck in their own attitudes.
VY: For example?
MH: For example, they've made an initial formulation of the case. They've been treating the case. And they didn't reformulate. At our second-opinion clinic, we give them a written report, sometimes a dozen single-spaced pages long. We go through the phenomenon, we go all the way through states, and then we end with technique, and we buttress this with the empirical literature where we can. So there are concrete suggestions like, "Why don't you say this?" Then we get the response from the patients and clinicians. It's extraordinarily successful.
VY: How do you know it's successful?
MH: Well, they say so. But how we really know is that the clinician then sends another case.
VY: Could you give an example of some of the types of suggestions? Therapy is so complex and so personal that I'd think a lot of therapists would be skeptical that you can get enough accurate information. How do you really know what's going on in the room so that you, as an outsider, could be helpful?
MH: We do two-hour interviews with the patient—you can do quite a different interview when you're a consultant than you can as a therapist. Where we have permission to, we record the interview and go over it again afterward. Then we discuss it with five senior faculty and a bunch of presidents, and then we boil it down. The patient's not paying for all that—they're paying for about 90 minutes of it, and we're spending six or seven hours as an intellectual and teaching enterprise. Then we give the written report to the therapist.

When we interview the therapists afterward, They say, "I kind of knew that—but I didn't know I knew it." They say, Yeah, now I see it!" So they had bits of it, but they didn't see how it fit together, and they didn't see where to go with it as a practical suggestion.
VY: So one way they get stuck, you'd say, is they don't reformulate the case. How else?
RA: It sounds like what you were just speaking to is that they're not taking that little blip of intuition seriously enough to truly consider it and to use that as a starting point to reformulate their original opinion.
MH: Right. One example (I'm fictionalizing, of course) is a case who was chronically suicidal to the point where they would get hospitalized—just from suicidality, not for psychosis. And yet the patient in therapy sessions was rational, presenting emotional topics. And the therapist, by the therapist's report and by the patient's independent report, was sort of hammering away at structuring current time, because the therapist felt that was disorganizing for the patient…
VY: Helping them structure the time in their life.
MH: Right. "What are you going to do this week? What did you do last week? Did you do your homework? Didn't you do your homework?" Giving them homework to do. Having phone calls: "If you don't call me by five o' clock, I'm calling the police," and that kind of thing. The patient definitely felt the therapist was very caring, no question. (In our second opinions, by the way, we're not referring the patient to another therapist.) But they were feeling stalemated, because while that was a little stabilizing for the patient—
VY: They weren't getting better. They were still chronically suicidal.
MH: Right. So in our formulation, we put together a number of pieces of evidence and said, "Look: This patient has two forms of confusional states. Even though they're not manifesting their confusional states in the therapy hour, we can infer that they are having confusional states when they're not with you. And here's what's happening in those confusional states." We were specific about it, but I'll be general: They're confusing thought and action, so they're weighing, in terms of their deeply held emotional values, certain things critical to the self, when they were thoughts, not actions, and they're treating the thoughts as if they were actions. And they're confusing self and other—so they don't always know whether you said something or they said something, or you think this about them or they think this about themselves.

And those are two things that you can tell the patient about in a sympathetic way, that they do this. Then the focus of the therapy becomes: "What's the difference between thought and action, and what's the difference between you and not-you?" And, You have some vulnerabilities here, and we need to address them, very patiently, very slowly, very repeatedly."

Then the patient would say, "This is terrible"—there would be obstacles to hearing that. But once the patient realizes that you're really sticking with them like you have stuck with them, and that you are examining this together, then when they're having these confusional states outside the therapy, they can say, "Oh, I'm going to talk about this with Dr. So-and-so. I don't have to do anything about it right now."
RA: And they can know what it is, at least.
MH: Yeah. And we said, "Well, this is going to be scary for you because you think maybe if you talk about confusional states, they'll get more confused. But states are unlikely to get worse. So this is an experiment; see if they get better."

The Near Future: Research Directions in PTSD

VY: We've covered a wide range of topics because you've had a wide-ranging career with many accomplishments and contributions. What's of interest to you now? What are you working on these days?
MH: Well, I'm trying to deal with what you might call personalized or individualized choices of psychotherapy techniques in PTSD. I don't think PTSD is treated as optimally as we can do it. And I don't think some of the manualized treatments, while they're effective, are effective enough.
VY: Say a little more—what do you mean by personalized?
MH: Decision trees. We're trying to write up a fifth edition of Stress Response Syndromes. Everything has held up pretty well in that book and successive editions, but the fifth edition will have more on how you make decisions at critical moments in therapy—like when to use exposure techniques, and when not to use exposure techniques because they're likely to retraumatize the person rather than desensitize them. So I hope that will be helpful, because a lot of people are just taught, "In Session One, give them education for 20 minutes. Then get the story of the stress event for 20 minutes. Then assign homework. In the next session, review the homework for 10 minutes, then do a gradated exposure treatment, then assign more homework, then give more education. Then in the third session..."
VY: That sounds like bad therapy.
RA: Listening to that, it's very easy to see how so many therapists would end up underestimating the potential of their clients.
MH: Yeah. But if you want to hire somebody with one year of training and pay them a little less than you'd pay an experienced clinician, and have them be helpful to people, that will be helpful. It's just that it won't be as helpful as that patient might need. So you could start with that, and if the patient has a remission of their disorder, fine. "Come back if you have trouble." But if they don't have remission or if they've dropped out, then you have to make some new decisions. Or if you have an experienced clinician, you can make decisions all along and decide when to do what.
VY: Well, I think this has been a great discussion. Thank you so much for coming and talking with us.
MH: You're welcome. It was a pleasure.

Copyright © 2009 All rights reserved. Published August 2009.
Order CE Test
$30.00 or 2.00 CE Points
Earn 2.00 Credits
Buy Now

*Not approved for CE by Association of Social Work Boards (ASWB)

CE Test
Mardi Horowitz

Mardi Horowitz is Professor of Psychiatry at UCSF and President of the San Francisco Center for Psychoanalysis. He has advanced integrative psychotherapy theory in his book Cognitive Psychodynamics and is author of the defining book on PTSD, Stress Response Syndromes, now in its fourth edition.

He has written 17 professional books and over 280 scientific articles in the fields of psychology, psychiatry, and psychoanalysis. His recent books include Treatment of Stress Response Syndromes, Understanding Psychotherapy Change, and for a popular audience, A Course in Happiness. Psychotherapists may be especially interested in Personality Styles and Brief Psychotherapy, Nuances of Technique in Dynamic Psychotherapy, Formulation as a Basis of Planning Psychotherapy Treatment, and Person Schemas and Maladaptive Interpersonal Patterns.

In addition, he has provided forensic, media and governmental consultations in the fields of PTSD, stress, personality, and psychotherapy.

To contact Mardi, order his books, or to see more of his artwork and his complete list of publications, please visit

Victor Yalom Victor Yalom, PhD is the founder and resident cartoonist of He maintained a busy private practice in San Francisco for over 25 years, but now sees only a few clients, devoting the bulk of his time to creating new training videos for He has produced over 100 videos, conducted workshops in existential-humanistic and group therapy in the US, Mexico, and China, and currently leads consultation groups for therapists.  More info on Victor and his artwork and sculpture at

Rebecca Aponte Rebecca Aponte was the Operations Manager for from 2008-2012. She then left California for graduate school, earning a PhD in Psychology from Colorado State University - an experience that only deepened her appreciation for the experience she was exposed to during her time with Rebecca now works for the California Department of State Hospitals.

CE credits: 2

Learning Objectives:

  • Describe the relationship between Horowitz's interest in art and research on stress and trauma
  • Explain how manualized therapy for trauma can limit patient growth
  • Describe Horowitz's perspective on how new psychotherapists learn

Articles are not approved by Association of Social Work Boards (ASWB) for CE. See complete list of CE approvals here