Tony Rousmaniere on Deliberate Practice for Psychotherapists

Tony Rousmaniere on Deliberate Practice for Psychotherapists

by Victor Yalom
Psychotherapist and expert on experts, Tony Rousmaniere, explores the importance of "deliberate practice" (you know, what musicians and athletes do to master their crafts) for psychotherapists.


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


The Other 50%

Victor Yalom: Tony, congratulations on your new book, Deliberate Practice for Psychotherapists: A Guide to Improving Clinical Effectiveness. We’ll get to the deliberate practice part later, and find out what that means, but let’s start with clinical effectiveness, which we as therapists all certainly strive for. You’re very candid and self-revealing in this book, which I think is admirable. And it seems the thing that got you started on your quest towards improving your own clinical effectiveness was the realization early on in your training that you were only helping 50% of your clients. Can you tell us about that?
Tony Rousmaniere: When I initially started training at my first practicum, I was working with high school students and I had a number of the clients respond very quickly. They had a range of different goals and whether it was anxiety, or feeling depressed, or wanting to do better at school, and they showed what is called in the research literature, “rapid gains.”
VY: That’s always nice when that happens. It makes you feel like you know what you’re doing, or you’re doing something helpful.
TR: I went into the field feeling like I could be good at this. I’m good with people, so I was optimistic, and the initial response from clients gave me a lot of optimism. But as time went on,
I gradually realized to my disappointment that a fair amount of my clients were not improving.
I gradually realized to my disappointment that a fair amount of my clients were not improving. And when I started to try to assess overall how many that was, it was about 50%. I call that “my other 50%.” There’s some of them who responded a little, and then just plateaued. There’s some of them who deteriorated—they actually got worse during treatment—and then there are also a fair amount, at least a quarter of them, who just dropped out.
VY: Dropouts are certainly a big problem for almost all therapists. I certainly recall, especially early in my career, I had a file of dropouts that came once or twice, and it was a pretty thick file.
TR: Yeah. It’s something we don’t always like to talk about but it is pretty universal across therapists.
VY: So you took the initiative to take a frank look at this, and what did you find?
TR: Well, I spent a number of years throughout my training trying to figure out what was going wrong and then how could I improve. Specifically, how could I reach the 50% of clients that I wasn’t helping effectively? And I started going back to the traditional method of clinical supervision. I was doing the same clinical supervision that pretty much every graduate student does, where they’re meeting weekly with their supervisor for an hour or two individually, and then also with a group.

I’d often feel like I was getting better, and I was learning the theory better, so I could write better papers about psychotherapy, or I could talk in more sophisticated ways about psychotherapy, but the percentage of my clients actually improving stayed the same.
I was very fortunate that throughout my training I had really good supervisors. I know that’s not always the case, but every year of my graduate training I had supervisors who were very open, very collaborative, very encouraging. They had really good advice and understood psychotherapy theory and technique well, but I found that though I was getting all of this great advice from them and my peers in group supervision, my effectiveness was not actually improving.

I’d often feel like I was getting better, and I was learning the theory better, so I could write better papers about psychotherapy, or I could talk in more sophisticated ways about psychotherapy, but the percentage of my clients actually improving stayed the same.
VY: It sounds like one thing you did was actually track your data, which is something most of us don’t do. We rely more on the second form of feedback you described: Do we feel good about what we’re doing? Can we talk about it intelligently? Do our peers seem to respect us? But that’s not really what we’re in the field for.
TR: Our whole field suffers from a lack of outcome data at the individual therapist level. We have lots of data from randomized clinical trials which show you how therapists do in these tightly controlled circumstances. And we have some data from research collaboratives where they’ll track a large group of therapists over time. But pretty much no therapist individually tracks their own outcome data, or reports it to the public. So nobody really knows how effective they or other therapists are. We know how well we can talk about therapy, or how well we can write about therapy, or how well we can theorize about therapy, but imagine if you could never see a basketball player play, you could only hear them talk about how well they played. Or you could never hear a violinist perform, you could only hear them talk about it.

Imagine if you could never see a basketball player play, you could only hear them talk about how well they played.
This is a real problem in our field. Imagine learning to paint, but you’re never able to show your paintings to anyone. You would just describe them to someone and say, “In this painting I used a lot of green. It might have been too much. Do you think I should have used less?”
VY: When I produced my first video, and then got in the business of producing training videos, what I used to say is, imagine a dental student going to a lecture about dentistry, or about a certain technique like doing fillings, and then going off to perform the filling in a private room, and then meeting with a supervisor a week later to discuss what they did. Would you risk getting a filling from such a person? That’s the problem we’re dealing with. And that was one of the things that motivated me to start producing videos of expert therapists doing therapy.

So you were aware of this problem and used the traditional tools available for developing skills as a therapist: clinical supervision, reading, talking with colleagues.
TR: Going to workshops.
VY: But you still found that your client outcome data wasn’t getting better. How did you track your client outcome data?
TR: I was using one of the simpler outcome measures called the “Outcome Ratings Scale” that as well developed by Scott Miller and Barry Duncan and others, and is part of what’s called “Feedback Informed Treatment.” It’s very accessible—it’s free and can be downloaded from their websites. It lets therapists over time track how well each client is doing, and then if they get enough data, let’s say 30 to 50 clients, they can look at how well are they doing as a therapist overall.
VY: Once you got your data, what did you do then?

Deliberate Practice

Honestly, I just started casting about, trying everything I could get my hands on. I went to lots of different workshops, read lots of different books and got supervision from different people. I was in a supervision group with you, as you well know, where we actually used some of the methods of deliberate practice, though we didn’t call them that. In retrospect, I can see that they were, and we can talk about that later on.

But there’s one supervisor in particular I found, Jonathan Frederickson, who was trained as a classical musician, and as a musician he used the method of deliberate practice. He integrated deliberate practice into his supervision and I found that working with him, using those methods, that it really improved my effectiveness more directly.
VY: Can you define what deliberate practice is and where it came from?
TR: Sure. Did you ever learn a musical instrument?
VY: Depends what you mean by learn, but I tried. And achieved a very low level of mastery with a few instruments.
TR: What instruments?
VY: Piano. Clarinet. Banjo. Harmonica.
TR: So imagine you went to your piano teacher and you said, “I want to be really good at piano. In fact, I want to be a professional pianist. But I just don’t have time to practice. I’m hoping you can assign me some books so I can get better. We’ll meet once a week, and then in a few years I’d like to have some performances.” What do you think your piano instructor would say?
VY: If I could say that with a straight face, I’m sure I’d be laughed out of the room.
TR: Exactly. As part of learning piano you did deliberate practice. Did you ever learn a sport in school or college?
VY: Sure. I played tennis and I’m engaged in some deliberate practice of tennis these days. I have a weekly lesson and am playing during the week and trying to get better, but it’s very difficult.
TR: Well, imagine you went to your tennis coach and said, “I want to play tennis at a professional level, but I just don’t have time to practice. I mean, who has time for that? So let’s meet once a week. Give me some books I can read and I’ll make it work.” They would, again, laugh you out of the room, right?

Most people have experience with deliberate practice, they just do it in other fields. Many fields use deliberate practice as a core part of training—not just to be a professional, but to achieve basic competence, to achieve moderate expertise, and then to achieve full expertise.
VY: So what does that mean in a psychotherapy practice?
TR: Deliberate practice is a term invented by K. Anders Ericsson and colleagues in the early ‘90s. They were trying to figure out how experts achieved their expertise across a broad range of fields—musicians, athletes, chess masters, pilots, you name it—and they isolated only one variable that predicted expertise: solitary deliberate practice.

Deliberate practice is based on five principles. The first is observing your own work. So in psychotherapy that would be watching videotapes of your own work, or having an expert observe your work.

Second is getting expert feedback on the work. So that’s supervision or consultation.

The third is setting small incremental learning goals just beyond our ability. In tennis, that is turning your wrist a little to the left, or in piano it would be just working on this one note.
In psychotherapy, we talk about broad things like trying to improve the working alliance, but there are a hundred skills that fall under that broad umbrella.
In psychotherapy, we talk about broad things like trying to improve the working alliance, but there are a hundred skills that fall under that broad umbrella.

The fourth component is repetitive behavioral rehearsal of those specific skills. So when you’re playing tennis you’ve got the ball machine shooting balls at you and you’re just hitting the balls again and again and again. That’s your repetitive behavioral rehearsal. It lets you move the skills that you’re learning into behavioral memory, procedural memory, so that they can begin to happen automatically, which frees up your mind to think about more complicated parts of the game.

The fifth component of deliberate practice is continually assessing performance. That’s something we do subjectively in psychotherapy, but there’s a lot of research to show that our subjective assessments of client outcome are not terribly accurate.
VY: One thing you say in your book, which I find quite refreshing, is, “I am not a master clinician. I am not a master therapist.” Why did you write that?
TR: Well, I wanted to be very clear. This is not a book by an expert therapist and this wasn’t me imparting my wisdom about my therapy techniques. I am a beginner. I am relatively new to the field. However, I am obsessed with becoming a more effective therapist. I might not ever become an expert therapist. I might not ever become a master therapist. That’s okay. As long as I keep getting better, I feel really good about that. So I really wanted to frame this book from the very beginning as one about just trying to improve.
VY: How did you start learning about deliberate practice and then implementing it for yourself?
TR: Well, I should say that I actually found out about deliberate practice when I interviewed Scott Miller for In that interview, Scott Miller talked about deliberate practice for psychotherapists, and it was the first time I had ever heard of it. So he should get credit. He is the first psychologist to consider this for our field and he worked on this from the ground up.

My supervisor at the time only would supervise therapists who videotaped their work. He said the reason was that there’s so much nonverbal communication going on. A lot of it is totally unconscious. Unless we can see what’s happening in therapy, as well as hear it, we just don’t really know what’s going on. And as I showed him videotapes of my work, there were multiple instances where the transcript of the session looked like good therapy. It read like good therapy. But the nonverbal communication showed that the client wasn’t progressing at all.
VY: You give several examples of that in your book. Can you give us one now?
I found that I had a bunch of clients who were basically complying with me. They were pretending to go along with therapy.
I found that I had a bunch of clients who were basically complying with me. They were pretending to go along with therapy. They would answer my questions. They would think about themselves, but they weren’t really struggling within themselves. They were appeasing me and kind of assuming, “If I give Tony what he wants, somehow magically I’m going to feel better.”

And I was going along with this. In fact, sometimes I was even cutting them off. I was talking over them. That’s another thing you can’t see in a transcript. Sometimes my tone of voice was very strong. Theirs was very meek. You can’t catch that in a transcript. Sometimes I would be sitting forward, with a lot of intention in my seat, and they would be sitting back kind of passively. In psychodynamic therapy, we call these “transference dynamics.” Each model of therapy has a different way of discussing the relational dynamics between the client and the therapist, but I found that by watching video I was able to identify all kinds of mistakes I didn’t realize I was making.
VY: It takes courage to look at yourself and have someone else observe you.
TR: Thank you, but it felt more like desperation than courage. I got into this field because I really wanted to help people, and I had a lot of clients that I really cared about. I really wanted to help them but I wasn’t. Sometimes they’d drop out and sometimes they’d deteriorate, and that really pained me.

I could give you another example. Role-plays are another great way of getting direct observation of your work and we would do role-plays in the consultation group you and I were in together. You observed while I was role-playing with one of the other group members that my voice was kind of forced.
VY: Yes.
TR: Do you remember?
VY: I do remember it, yeah.
My voice sounded like someone trying to be a therapist rather than just being a real person.
I was trying to be a therapist. And my voice sounded like someone trying to be a therapist rather than just being a real person.
VY: Right.
TR: That would have never shown up in a transcript. What you advised me to do is to work on this specific skill. We isolated the specific skill. You said, “Just try talking naturally, Tony. Just try saying whatever you’d say naturally.” And if you remember, it was hard. It took a lot of practice for me to do that. I don’t know if I ever told you this, but I went back after that group and I watched video after video of my clients and I practiced just talking naturally to my clients in the videos.
VY: You just sat by yourself and practiced saying the words aloud?
TR: Yeah.
VY: Wow. So that’s an example of solitary deliberate practice. You were just sitting by yourself with a video and practicing speaking.
TR: Exactly. In most other fields, the bulk of the training actually occurs during solitary deliberate practice. So a professional musician might get coaching a few hours a week, but then they’re spending 20 hours a week practicing on their own. The same with an athlete. Same with a master chess player. And that is something that we do not have in our field. We spend time reading about psychotherapy a lot. But we don’t spend time practicing skills ourselves, so the skills don’t move into procedural memory, and then we’re often left floundering in session.
VY: I remember that term procedural memory from graduate school, but I don’t remember what it is. Can you refresh our readers about what it means and why it’s important?
TR: When you ride a bike you are using procedural memory. When you drive a car you’re using procedural memory. It’s when your body just remembers automatically how to do something, because you’ve done it so many hours. It’s automatic. So you can think about other things while you’re driving—like how to get to your destination—because your body knows how to make turns and yield and stop at the light.

Now, that can be a double-edged sword. My wife points out quite frequently that my driving is not always so great. But it’s in procedural memory, so I do it automatically. We want to get the skills into procedural memory, but then we want to also keep refining them throughout time, or else we stay stuck at the same plateau.
VY: Getting back to deliberate practice, so the first step is observing your own work, and one way to do it is through video. Getting expert feedback is step two, and you were getting some feedback from your supervisor about your work via video. The next step is setting small incremental learning goals just beyond your abilities. How do you do that?
TR: Ideally that’s done by the supervisor. In the group supervision we were in, you identified my voice being forced, which was something I couldn’t hear in myself. You showed me how to improve that and then let me practice it. In the group, you gave me little tweaks here and there. Try a little of this, a little of that. And then I took it home to practice on my own with the solitary deliberate practice. Ideally we’re getting that kind of corrective feedback that focuses on specific incremental skills throughout our careers. That’s how you learn pretty much any other skill.
VY: In any other field you’re getting constant feedback. If you’re a lawyer, you’re observing your senior try a case and you’re sitting next to him and maybe you’re getting up and doing some things and they’re observing you. If you’re in plumbing, you’re an apprentice plumber, you’re going to watch a master, they’re going to watch you. We’re about the only field that I can think of where that doesn’t happen on a regular basis.
TR: I think we actually work in one of the most secret fields on the planet, though not intentionally so.
I think we actually work in one of the most secret fields on the planet.
I mean, obviously there’s confidentiality rules and that kind of thing, but even CIA agents in deep cover every few years get some kind of performance review. But I could go the next 30 years without ever having anyone give a meaningful look at my work. We’re required to do continuing education units, but that’s generally about cognitive learning, which is valuable for learning new laws or new theories, but a lot of research has shown that it doesn’t translate to improved skills or effectiveness with clients.
VY: You cite a lot of evidence in your book that even years of clinical experience don’t lead to improved performance.

The Audience Can Tell the Difference

Researchers have been looking into this for decades. There’s literally decades of research and they’re trying every which way to show that experience improves performance. But except for isolated cases here and there—for example, experienced clinicians can do better with severely psychotic clients—experience is not associated with improved performance.

I think this can be possibly explained by the fact that we do not as a field engage with ongoing deliberate practice. You could take a professional basketball player and if you tell them that they’re not allowed to practice anymore, and then ask them to play 10 years later, they’re not going to be as good.

My friend plays for the symphony in Washington, DC, and she practices two hours a day, six days a week. She’s at the very top of her field and she still practices. She’s getting close to retiring. She still practices. I asked her why she still practices and she said, “If I go a day without practicing, I can tell the difference. If I go two days without practicing, my peers can tell the difference. If I go three days without practicing, the audience can tell the difference.”
VY: The evidence is compelling, but it flies in the face of what we as clinicians think. Most of us feel a lot more confident ten or twenty years into our practice. We feel like we know so much more, not only from our clinical work, but from our life experience. We can empathize with a broader range of clients because we have a broader range of experiences ourselves. We’re not so anxious in session, worrying about how clients are going to think of us, and whether they are going to see how young and inexperienced we are. So it just feels like we are much better therapists. Yet you’re saying that the evidence does not bear that out.
TR: Well, the evidence shows that there’s a lot of variability. Some therapists do improve in time. But some get worse over time. And because we’re typically not tracking our outcome data from an empirical perspective, it’s hard for us to know. We have a lot of cognitive biases, not because we’re bad people, but because it’s the way our brains were built. So it’s risky to trust your own private perception of your work over time without ever getting feedback.

Unfortunately relying on our clients’ opinions is not entirely reliable either. There’s been many studies showing that clients will routinely not tell their therapists when they’re not doing well. In fact, Matt Blanchard and Barry Farber at Columbia University did a study of over 500 clients and found that 93% of them reported having lied to their therapist. Negative reactions to therapy was one of the most common topics they lied about, including pretending to find therapy effective, and not admitting wanting to end therapy.

Now, almost every client I have in my practice has been in multiple previous therapies that they found to be marginally effective or not effective at all. They probably did not tell their previous therapist this. I can tell you, I have a lot of dropouts. I’ve had an overall 25% dropout rate across my career.
Almost none of my clients tell me that I’m not helping them before they drop out. They just leave.
Almost none of my clients tell me that I’m not helping them before they drop out. They just leave. These are the clients we need feedback from the most. Clients who are like, “Oh, this is helping so much!” are not as helpful with their feedback.
VY: Are you still using the same forms to get feedback from your clients?
TR: I use a variety of forms—the session rating scale and some others. I’m always experimenting with different ways of getting feedback from clients and also from experts—but what I do most now is record all of my sessions through video and then get expert feedback on the sessions.
VY: And when you have dropouts, if you look back on those rating forms, do you see warnings signs?
TR: Yes. There often are, but not always. Many clients feel pressure to be nice to their therapist. Look, when I’m at a restaurant and I don’t really like the food, and they come around and ask me how’s the food, I don’t often say, “It’s kind of crappy.” I usually say, “Oh, it’s fine.”
VY: So let’s get back to the final two steps of deliberate practice: engaging in repetitive behavioral rehearsal and continuously assessing performance. How have you gone about doing that?

Jazzing it Up

So the first three steps we’ve covered are usually pretty easy for therapists to understand, but I often lose them when I talk about repetitive behavioral rehearsal. They’re like, “Psychotherapy is a relational art. Every session is different. Every relationship is unique. This isn’t just playing chess and moving pieces around. It’s not football or basketball where the net is always in the same place. Our clients change their goals every session. We work in an infinitely complex field. So, how can we repetitively practice behavioral skills?”

A metaphor I like to use is jazz. Jazz is the kind of music that utilizes improvisation as an inherent part of the craft. But jazz musicians don’t just sit down and start randomly doing whatever they want on their instruments. To become a jazz musician, you actually go through very rigorous training where you’re learning standardized ways of playing your instrument. You’re learning the same notes as everyone else. You’re learning the same theory as everyone else. You’re practicing the same way as everyone else. And when all those musical skills are moved into procedural memory, you’re then able to improvise with other performers.
VY: That’s why I never got too far with clarinet, because I wanted to improvise. I just wanted to be able to improvise like jazz, but I wasn’t willing to spend the hundreds or thousands of hours playing the scales.
TR: There’s been a lot of research that shows that slavishly adhering to psychotherapy models, kind of following them cookbook style, or doing exactly what’s in the manual with every client, actually leads to worse outcomes. So that doesn’t help either.

There’s a tricky balance where on one hand you know the skills, you’ve internalized the skills, you’ve practiced the skills. But then on the other hand, you’re very adaptable and reflexive to the client.
VY: I think what you’ve pointed out is not obvious to therapists at all, because we just don’t have that in our professional culture, in our training. As you said, so much of the focus is on theory, on reading books, on writing papers, on being able to sound intelligent in class or seminars or group supervision. What are the actual skills to practice?
TR: Many people assume that since they’ve gotten lots of face-to-face hours with clients that that should count as practice. To get a degree and get licensed, typically you have to have hundreds or thousands of hours with clients.

It only counts as practice if there isn’t a real client in front of you.
Something K. Anders Ericsson and the other researchers on expertise found was that it only counts as practice if there isn’t a real client or real engagement in front of you. So a basketball player playing a game doesn’t count as practice. A musician performing doesn’t count as practice. A chess player playing a match doesn’t count as practice. That’s all considered performance. And the reason is that during performance you can’t isolate a specific skill, and you can’t repeat it again and again and again while getting feedback.
VY: I see that in tennis. I’ve spent years trying to learn a top-spin backhand, and yet when I play matches, I’m worried about winning the point. I default to hitting a slice. I don’t do what I’ve learned.
TR: Well that takes us back to procedural memory. When we’re in moments of what we call emotional arousal, your brain immediately goes to procedural memory. That is why it’s important to practice these skills behaviorally and repeat them hundreds and hundreds of times until they’re moved into procedural memory—so you can perform them in those moments of emotional arousal.

In psychotherapy, we work in states of very high emotional arousal. We help clients who are suffering intensely. And we feel that suffering while we’re sitting with them. So we will go almost immediately into procedural memory.
VY: We don’t have a lot of experience or knowledge about how to practice skills that are fundamental in the psychotherapy enterprise. How did you figure this out since there wasn’t a manual for you?
TR: Most fields have taken hundreds of years to figure out models and methods for deliberate practice. I’m hoping that we can start this. Because there wasn’t already a model or method for doing it, I focused on what’s called “facilitative interpersonal behaviors.” These are behaviors that have been shown by research to be effective in therapy across a wide range of models. You can think of them as the basics of psychotherapy. Many of them have to do with attunement with the clients in session, components of the working alliance.

A lot of research shows again and again that the quality of the working alliance in therapy contributes ten times more to outcome than the model or anything else. Bruce Wampold has written a lot about this in his books. He calls it “the contextual model for psychotherapy,” where he focuses on facilitative interpersonal behaviors. An example of that would be tone of voice. I’ve noticed that if I’m not careful I can start speaking louder than my clients. I can talk over them. I can basically overpower them with my voice. This is sometimes due to my own anxiety that goes up in session due sometimes to what they’re presenting, or my own counter-transference.
VY: How do you work on that?
TR: I sit with my own videos, especially videos of clients that I find stir up my own anxiety, and I will practice talking to the video in a level voice. I want to be engaged.
VY: You’ll literally be watching a video and just practice speaking?
TR: Yes. If someone saw me doing that, they would think I was crazy. But think of it like a basketball player shooting, practicing free throws. They’ll just sit there doing it again and again and again, and they might do a hundred a day. So I’ll spend 15 minutes just practicing speaking to videos of clients who I find I have some anxiety with when in session with them.
VY: So you’re experimenting with different tones of voice, and kind of get that into your body, into your procedural memory.
TR: Yes. Another thing I’ll do is I’ll watch videos where there’s clients who are stalled, deteriorating, something’s not going well. I’ll watch the sessions with the volume off. And I will take notes about everything I see in terms of their body language. And as I watch that, I’ll also notice my own anxiety. Does my own anxiety go up or down based on their body language?
VY: Your anxiety in the session, or your anxiety as you’re—
TR: Watching the video.
VY: Your anxiety as you’re sitting there watching the video?
TR: Yeah. I found this very surprising at first, but just watching my own videos was incredibly mobilizing of my own anxiety, my own feelings, and my own defenses. Every therapist I’ve talked with who watches videos of their own work also finds it to be quite challenging emotionally.

It’s exposing ourselves to ourselves, and in a way that we normally aren’t. And that’s one of the reasons it’s difficult to videotape and then watch your own work. So if I can sit there watching the video and noticing the body language and noticing my own anxiety, those are two different skills I’m working on. If I can do enough of that so it moves to procedural memory when I’m sitting with the real clients in front of me, it’ll be that much easier to do those skills in the background, so I can focus on something else.
VY: And what impact did that have on your work? How did you know or notice that that was actually helping you?
TR: Well, one thing I noticed is that I have a butt-load of anxiety with a lot of my clients. And I was shocked to find out. At first I was incredibly embarrassed. I didn’t want to tell anyone. And then I realized that some people could tell anyways when I talked about it with them. And then I thought, keeping it secret is not going to help anyone.
One thing I noticed is that I have a butt-load of anxiety with a lot of my clients. And I was shocked to find out.
And then I realized most therapists have some degree of emotional reaction. I’m a psychodynamic therapist; we call this “counter-transference.” But I also found that there’s a certain level of anxiety that’s kind of universal working with all of my clients. So I don’t know if it’s individual counter-transference from a certain client, or it’s just me.

Some of it might be a sympathetic reaction to what the client is bringing up. Some of it is just my own material. Some of it is wanting to do a good job. And there’s just a certain level of anxiety always going up and down within me during a session. I’m still not really good at this, but I’ve gotten better at tracking that in the background during the session. I can use it psychodiagnostically. So if a client is talking about something that really bothers them, but they’re good at hiding it in their words or even nonverbals, I can often feel their anxiety within me. A sympathetic reaction to their anxiety within me. There’s a clue there.
VY: Using yourself as a tool.
TR: Exactly. When I talk about deliberate practice, people often assume I’m talking about CBT or behavioral therapy, but that’s not accurate. The most benefit I’ve gotten from the deliberate practice methods has been with the more dynamic interpersonal/intrapersonal aspects of therapy.
VY: What do you mean by that?
TR: The more subtle, intuitive sense of myself and the transference roles being played out between me and the client, what I feel pulled to do with the client, how that might be repeating old problematic patterns from the client’s life. How my own counter-transference might be getting stirred up, and I might be guiding the client towards or away from material in ways that are unhelpful. How I might be retreating.

I’ll give you another example. A supervisor once pointed out that I was being critical of a client. I was horrified by this. Horrified. My job is to be empathic, not critical. And if you read the transcript, I was not coming across as critical. In the transcript, I was coming across as very supportive. But he said, “Listen to your voice. It’s critical right here.” I was embarrassed to admit it, but I actually had a sharp edge in my voice. And that was due to my own counter-transference.
VY: Whether you use the term counter-transference or not, or whether you work with a model that has transference or counter-transference or intersubjectivity, or as an important part of a theoretical model, those things are happening anyway.
TR: Yes.
VY: There are feelings between client and therapist that you’re feeling drawn orcompel us to do compelled with certain thing with certain clientsclients to do certain things, whether you act on them or not, whether it’s to support them, whether it’s to tell them what to do. Whether you feel detached or bored. Or whether they pull on your anxiety in one way or the other. Those types of dynamics are always occurring, whether you’re paying attention to them or not.
TR: Many of us know this from reading the theory, but we haven’t practiced actually noticing it in the moment. We practice it with real clients, but that doesn’t count as practice. So, one of the ways that I have addressed this is I’ll sit and watch videos of clients where, again, they’re stalled or deteriorating. And I will just write down what do I feel pulled to do. Do I feel pulled to save them? To criticize them? To support them? Or what role do I kind of want to be in with them?

And over time, doing this again and again, and again, I’ve built my ability to observe that as it happens in session.
VY: So the final idea in deliberate practice is continuously assessing performance. Usually we think that most of our training belongs in graduate school or early in our careers, when we’re interns or psych assistants, accumulating our hours. But you believe that if we want to achieve our maximum proficiency, we should be like other professions and keep doing whatever is necessary to get to the top of our game.
TR: In pretty much every other profession, professionals have to engage in continual deliberate practice throughout their entire career. And if they don’t, they stall, and then gradually decrease in effectiveness.
VY: Let me just challenge you on that. If you’re a professional athlete or musician, yes, you’re going to spend hours a day practicing. Most other professions, I think, you don’t do that. If you’re a surgeon, you do surgery. If you’re a lawyer, you do legal work. You’re not setting aside time to actually practice being a lawyer or a surgeon.

Competency vs. Excellence

Surgeons actually do set aside time now, and they engage in repetitive behavioral simulations. For other fields, including psychotherapy, it is possible to stay at a level of competence without deliberate practice. So I believe most therapists are competent. In fact, by the end of graduate training, most therapists are competent. Overall, the outcome data for psychotherapy is pretty good. It compares favorably to medicine in many ways. Our deterioration rate of 5 to 10% is actually not horrible. The rate of complications or side effects is very low. The rate of legal problems, people suing us, is relatively low. Overall, we perform a competent service, right? And you can stay an absolutely competent therapist your entire career without using deliberate practice.

Now if you’re an accountant, you might not need to get better. Being competent might be totally fine for your livelihood. Or if you’re a lawyer, being competent might be totally fine for your livelihood. And I’ve met musicians who don’t engage in deliberate practice. They’ve found a level of competence which works for them and they’re totally happy with that. That’s totally fine. For me, it’s not satisfying. It wasn’t satisfying. And it still isn’t satisfying. But that doesn’t mean that it has to be appropriate for everyone.
You can stay an absolutely competent therapist your entire career without using deliberate practice.
I know that for several years your wife got a job at the University of Fairbanks and you were up there with a lot of darkness. And you used that time productively by learning about deliberate practice and some of these exercises you’ve just described. For therapists that are reading this and are intrigued, and do have that desire to up their game, in addition to reading your book—which is wonderful and well-written and also very funny at times—what would you advise them to do in terms of utilizing these principles?
TR: I’d recommend a few things. One is record your work. Video is really the most effective way of doing that. Using video for consultation supervision is now becoming more and more recommended across the field, and I have advice in the book about how to start videotaping your work. I want to emphasize that this is especially true for psychodynamic therapists, who are traditionally the most resistant to reporting their work.
VY: A lot of therapists worry that their clients will be put off by that.
TR: There’s been a bunch of research on this, and they’ve found that clients in general don’t mind. The client wants to get better. That’s really what the client is thinking about. I don’t mandate recording video for all my clients. I always ask them and it’s always optional and 10 or 20 percent say they don’t want to do it. I don’t argue with them about it.
VY: So you think it’s the therapists who are more uncomfortable about it?
TR: The research shows that, absolutely. Mark Hilsenroth, a psychodynamic researcher, and colleagues did a study recently where they gave the clients questionnaires about using video, and most of the clients were like, “fine, no problem.” They just want to feel better. When I go to the doctor, I’m like “do whatever you got to do.” I want to feel better. That’s what I’m thinking about. However, they also gave the questionnaires to therapists, and they found that when the therapist was uncomfortable with video, the clients were more likely to be uncomfortable with video.

I almost got fired from one of my first supervision jobs because other supervisors were uncomfortable with me using video. Therapists can be very uncomfortable with it, which I find to be quite ironic. Because the clients don’t seem to mind much.
VY: How do you introduce it to clients?
TR: I’m very upfront with the client. I say, “ I’m a human being, I make mistakes like everyone else. And if we record the session, and I can look at the videos later, or show them to experts for consultation, I have a much higher chance of spotting my mistakes. And then we can address them and then I can help you more.”
VY: It makes so much sense. And as you say it now, I recall early in my career, maybe in my internship when we audio recorded our sessions, the idea that I might make mistakes, or that I was getting supervision or consultation, filled me with a lot of anxiety. I think that’s more reflective of the state of anxiety that many beginning therapists feel. And as you mature you realize you’re not perfect, that you don’t help everyone, that there’s always more to learn. Certainly a maxim in psychotherapy is that there is no end to what clients can learn about themselves. There’s certainly no end to what therapists can learn about themselves, including how to be a better therapist.
TR: I’ve found through watching years of my own tapes that if I work with a client for two or three sessions, I’ve already made a mistake. Honestly, I probably made a mistake in the first session, which sometimes can take two or three sessions for me to see. So if I’m not seeing my own mistakes by the third session, it means I’m missing something. And I’m okay with that.
I don’t think being an expert means never making mistakes.
I don’t think being an expert means never making mistakes. It means knowing how to spot your mistakes and correct for them in a timely way.
VY: All right. So you’d encourage therapists first to start video recording their sessions. And then what?
TR: To get expert feedback from someone that they trust. It’s got to be someone you feel good about it. A good supervisor is able to get under your skin. You were able to notice something in my voice. And that’s personal, that’s intimate. And it was okay because I trusted you. We had a good relationship. Without a relationship like that, it’s going to be hard to get the necessary feedback. Ideally it’s a long-term relationship. A lot of our trainings are these one-off weekends or series of two or three weekends, where you’re getting a big knowledge dump, but no one is looking at your work. You’re not getting individualized feedback. And then you’re not getting ongoing long-term feedback. But that’s what’s necessary for the skills to improve.
VY: I think that may be changing. Some of the approaches that we’ve just been making videos of—motivational interviewing and emotionally-focused couples therapy— actually have a lot of that integrated into their ongoing training, where you have to submit samples of your work and get feedback on it. But what you’re saying makes a lot of sense.

Research shows that most therapists think they’re well above average, which statistically is impossible. How do we then go about choosing a supervisor, a consultant, who is good?
TR: This is tricky because I don’t know any supervisor who tracks their outcome data or reports it to people who are approaching them for supervision. At this point all we can really go off of is our gut sense, and occasionally we can watch videos of our supervisor’s work. I found you because I met you and had a good feeling about you. And then as we did supervision together I found it was helpful. But ideally we’ll have a more empirically rigorous way of assessing that in the future.
VY: I tell therapy clients to meet with a therapist a few times. If it doesn’t feel helpful, you may want to discuss with them what feels good, what doesn’t feel good, and see if they’re open and receptive to hearing that. If they’re not, or the therapy doesn’t feel helpful, try someone else. It’s too important not to.

So get a coach, supervisor, a consultant. And then what?

Track Your Outcomes!

Another thing I recommend doing is tracking your own outcomes, and then using some kind of empirical measure to do that. The outcome ratings scale is a great measure to use. It’s free. It’s easy to use. There are dozens of other measures available. There’s the Outcome Questionnaire. There’s the Behavioral Health Measure. There’s measures made for different settings, like universities, or working with children. And accumulate your own outcome data over time. And over years you’ll start to get a picture of how effective your practice is.

One of the reasons I started doing this is I had a supervisor look at my work and she thought I was doing horrible work. In fact, she said, “You want to kill your clients.” I was shocked. I knew I had made mistakes but I didn’t think I was that bad. But I didn’t have any data; it was just one opinion versus another. This is one of the reasons I doubled down on collecting my outcome data. After a year I had enough outcome data to look at my practice and see that overall I was helping the majority of my clients.

I definitely still have dropouts and deteriorations, but it helped my self-assessment be more level. Before then, there were some weeks I felt like Superman. I felt like everyone was getting better. And then some weeks where it seemed like everyone was getting worse. Of course, neither was ever true.
VY: But we certainly have days like that. If you’re in private practice and you have a few dropouts, or a few no-shows, it’s hard not to feel like something is wrong with you. So getting long-term outcome data is kind of a buffer for that.
TR: I found that my outcomes at my private practice in San Francisco were pretty good. The outcomes at the university counseling center in Alaska were not as good.
The outcome data never looks all good. And it never looks all bad.
Maybe that was due to the setting, the clients, maybe it was due to the darkness. Maybe it was because I was on the edge of being depressed because I was in the middle of Alaska. I mean, it could have been any number of things. Back here in Seattle, the outcome data is looking a little better. But importantly the outcome data never looks all good. And it never looks all bad.
VY: So it’s not so bad that you think you should hang up your shingle. And it’s not so good that you think, “I nailed this. I can coast.”
TR: Yes. Correct.
VY: So people start recording their sessions, getting a consultant in a long-term relationship, but the rubber meets the road with deliberate practice. What would you recommend to help people get over the initial hurdle, because I imagine it’s a big hurdle to actually sit down and do some of these solo exercises that you recommend.

“It’s the thing I look forward to least in my day”

It is a big hurdle. It’s the thing I look forward to the least in my day. It’s the thing I put off the first in my day. I would rather go to the gym, pay my taxes. In the recent election I was making get out the vote phone calls, which is a very stressful thing to do, and I found that I would do that before my deliberate practice. So it is very, very stressful. And unfortunately in our field it’s not recognized. It’s not rewarded. You’re not compensated for it. Your clients don’t know you’re doing it. Your peers don’t know you’re doing it, or don’t care. A licensure that never asks, or doesn’t care if you do it.
VY: Your spouse may prefer that you go wash the dishes, rather than sit and talk to yourself on video.
TR: Exactly. And to add to that burden is the fact that there are not immediate payoffs. They call deliberate practice short-term effort for long-term gain. So here’s what I do: I think of the therapists who are really, really good who I want to be like. And I know from talking with them that they got that good by engaging in hundreds or thousands of hours of watching their own videos. I’m not smarter than them. I’m not more talented than them. If I ever want to be that good, I’m going to have to put in that time.

The same way that if I wanted to be a really good basketball player, or a really good anything else. It might not make me as good as they are, but it will definitely move me in that direction. I have a reminder that pops up on my computer every day that says, “How good do you want to be in five years?” Now, if that day I don’t really care how good I am in five years, I won’t do it. And that might be fine. I might feel like I’m good enough, and that’s totally fine. But as of today, I still want to be that much better in five years.
VY: Well, I admire what you’re doing. And I’m gratified that I was able to impart some wisdom that was useful to you. It’s lovely to have this conversation and to have been able to read your book and have the tables turned and to be able to learn some very valuable things from you, Tony.
TR: Oh, thank you. To be interviewed by you for your website, it’s a great honor.
VY: I would encourage anyone who finds these ideas interesting to go out and grab your book and read it. Although it is chock full of research citations to back up what you’re saying, it’s not just idle theory. It’s also chock full of funny stories, humorous anecdotes, and I guess I’d like to just leave our audience with one of them. Can you tell the story about the job at the university?

Professional Identity Politics

Sure. My wife was applying for a job at a university in the West that really wanted to hire her. It was a very small town, and it was full of therapists, so I didn’t think I could just start a private practice there. She’s a wildlife biologist and the ecology department at that university that wanted to hire her were trying to arrange what’s called a “spousal hire,” which is something traditionally done in academic circles when they want to hire a person and there’s a spouse. They call it the “two body problem.” So they went to the university counseling center and they said, “We will give you money, we will pay for the salary if you hire Tony for three years. Part-time. Just so we can get his wife. We don’t care about Tony, but we want his wife, and Tony comes with the wife.” In other words they could have had me as a part-time therapist for three years for free.

We’re more like religions than any kind of public healthcare service.
They asked me to submit videos of my work as part of the application process and I thought, “This is great. I’ve been videotaping my work for years now.” So I sent in some videos and went in for the interview and they were horrified by my work. The style of therapy that I do is short-term psychodynamic psychotherapy. It’s a bit more active and engaged and I work actively with the client’s feelings and defenses. They were doing a more traditional long-term, reflective approach of psychodynamic therapy. When we were watching the video they kept asking, “Do you think this is appropriate for the client?” I kept saying to them, “Why don’t we look at the client outcome data. Why don’t we look at how the client responded?”

It’s like we were having two different conversations. They weren’t really concerned with how the client was responding. They were concerned with the model of therapy I was using. It made me realize that we’re more like religion than any kind of public healthcare service.
VY: You wrote in your book that they weren’t interested in your outcome data any more than a church would want to see how many meals a Buddhist monk had provided to the poor!
TR: Exactly. If we don’t collect our outcome data, if we don’t look at our work, we get unmoored from the outcomes, and we get stuck in professional identity politics where have all these debates about obscure theory because we don’t have actual outcome data to look at. They actually liked me as a person. They said, “You’re such a nice guy. It’s a shame it’s not going to work out.” But they didn’t accept me, and so we couldn’t move there, she didn’t take the job.
VY: The interesting thing is you were both in the general rubric of psychodynamic therapy where oftentimes the clashes are most intense.
TR: Yes.
VY: I had a college roommate who was a Leninist and he would go to some Communist convention. Probably less than a very, very small percentage of the population consider themselves Communist. And instead of coming back with a Kumbaya feeling, he would come back and report to me the big clashes between the Stalinists and the Leninists.

And even now with this emphasis on evidence-based treatments, or so-called evidence-based treatments, there’s a clash often between modalities, not taking into account that the data finds that modalities and theories do not explain outcome.
TR: If anyone ever talks to you about evidence-based treatment, ask them whose evidence. If it’s someone else’s evidence, it is not correlated with your personal outcomes as a therapist. There’s been study after study after study showing that though the models are proven very effective in clinical trials, when taught to therapists they don’t improve the outcomes of individual therapists.
Until evidence means our individual evidence, we’re not really doing evidence-based therapy.
Until evidence means our individual evidence, we’re not really doing evidence-based therapy. We’re working from someone else’s evidence.
VY: Well thank you so much for taking the time to share your journey and your expertise with our readers. Even though you humbly claim not to be an expert therapist, you certainly have accumulated a great deal of expertise on how to become an expert or master therapist.
TR: Yeah, I am learning a lot about how to become one. And thank you for having me. It’s been wonderful.

Note: Tony’s latest article, "What Your Therapist Doesn’t Know," has been been published in The Atlantic.

© 2017, LLC
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
Tony Rousmaniere Tony Rousmaniere, PsyD is Clinical Faculty at the University of Washington and has a private practice in Seattle. He hosts the clinical training website, and is the author/editor of four books on clinical training: Deliberate Practice for Psychotherapists, The Cycle of Excellence: Using Deliberate Practice to Improve Supervision and Training, Using Technology to Enhance Counseling Training and Supervision: A Practical Handbook, and the forthcoming Mastering the Inner Skills of Psychotherapy: A Deliberate Practice Handbook. In 2017 Dr. Rousmaniere published an article in The Atlantic Monthly, “What your therapist doesn’t know”. Dr. Rousmaniere provides workshops, webinars, and advanced clinical training and supervision to clinicians in the United States, the United Kingdom, Europe, Asia, and Australia. He was previously Associate Director of Counseling and Director of Training at the University of Alaska Fairbanks Student Health and Counseling Association. More about Dr. Rousmaniere can be found at
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

CE credits: 2

Learning Objectives:

  • Discuss the importance of outcome data for psychotherapists
  • Describe different methods for acquiring data from clients
  • Explain the importance of deliberate practice of psychotherapy

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