Some Musings on Cognitive Behavioral Therapy

 Robby Bensinger told me to finish a post on CBT, so hark! A post on CBT. Many, many thanks also to Miri for critiques and help with the composite example in Part Three. 

Recently, I’ve been on this kick where I read books about therapy and self-help not because I think they’re going to be useful to me, but because I want to know if they’re going to be useful to other people. Like, this marriage manual*, for instance. Books that teach you to do Cognitive Behavioral Therapy on yourself have been a particular focus. When I’m asked for resource recs, people often specifically ask for evidence-based therapy or CBT, and I give the classic suggestions. (If you’re coming up with titles here, you’re probably right.)

And I get feedback! ….which is often mixed. The most common thing I hear is that the advice was good, but the narrator was condescending/annoying/patronizing. In fact, this is what I’ve heard almost every single time. Keep in mind that these are people who sought me out for a specific kind of book rec, then bought or rented the book I suggested, expecting that it would help. With that expectation, they read the entire thing, despite hating the writer’s tone.  (And they described it as ‘hate’ too.) That’s not encouraging.


I’ve been thinking lately about what kind of therapist I want to be. I’m approaching termination with my third therapist in three years. Very shortly I’ll be in exactly the same place my previous therapists were: in an internship, practicing microskills and hoping I sound supportive and trying to take notes and watch the clock and never have distancing body language and not sound overly judgmental and to do it all at once.

I’ve watched them work at it and improve and have off days when the prevailing theme of therapy was “You’re watching the clock and I’m watching you watch the clock.”

At the same time, through some combination of blogging, talking openly about my mental health, and social selection, I’ve spent most of the last few years around people who have tried therapy, are seeking therapy, or are in therapy. These too, are usually people who actively sought out therapists who offered CBT or ‘evidence-based therapy’ techniques. Many of them…maybe even most? that have abandoned therapy as a useful solution have done so because they felt their therapists talked down to them.

“They asked stupid questions!”
“Of course I’d considered that!”
“Yes, I know not literally everybody thinks I’m worthless, hasn’t anyone heard of exaggeration? I just said it felt like that sometimes


So for a very long time I had this model in my head that my friends has just encountered the Bad Therapists, and some rule of conversation meant that people preferred to commiserate over Bad Therapists more than they wanted to tell me about Sparkling and Perfect Therapists. (A version of misery poker or Northwestern’s favorite: homework Olympics.)

And then Part One of this post happened, and I switched from primarily reading instructions for clinicians administering therapy to instructions from clinicians to laypeople.

Sometime later, I noticed how many times my reaction to reading an example was “Wow, this example is painfully heavy-handed, but what’s the principle behind it?” or “I mean, obviously the client in this example didn’t mean that literally, but what’s the skill the therapist is trying to teach?” I was intentionally ignoring all the stuff that might turn clients off therapy entirely, in favor of learning techniques.

An example**

Jason has anxiety, and has been having trouble sleeping because he’s afraid he won’t finish his next project on time. His daughter has been ill, and he’s missed more work than usual, though he has a stellar track record with the company. He might be able to ask his boss for an extension, but every time he considers it, his chest gets tight and he has trouble breathing. He says he fears his boss will fire him for failing to complete a project for the important client.

I, [Therapist X] asked Jason to work through some questions for me.

“Have you ever had a friend who got fired for respectfully asking for an extension?”

“Don’t you think that your boss would rather have your work a little late rather than have to go through the process of hiring someone new to get it done even later?”

Except that if I were Jason, answering No to the first and Yes to the second still leaves me with the crushing fear of “What if I am the exception and I need this job to support my sick daughter, even a five percent chance of firing is still scary.” Working through the questions might give him an idea of what to think about when the fear is crippling, decrease the physiological reaction, or prevent future fears from spiraling out of control, but that hasn’t been made explicit. Jason has to trust Therapist X to have a plan.

And while there’s an argument to be made that books have to construct examples that are larger than life to have clear cut principles, I started noticing it in therapy myself. It was actually, once I was paying attention, very common for me to suspend my annoyance, as it were, in favor of expecting the paradigm of therapy to succeed.

So when my therapist asks “You said you wanted to restrict your calories after eating dinner two days in a row. Does that seem like it’s part of your plan for recovery?” I don’t say “Of course it’s not how I want to do recovery, but I’m struggling with anorexia, dammit!” I trust that she has some goal that involves me learning something new about triggers or a coping mechanism, and that that goal is served by the conversation her question is starting. So I say something like “No, it’s not, but I don’t know how to feel okay with eating so much more than I’m used to” and we go merrily on from there.

The problem is, this only works if I expect that therapy is going to be beneficial to me, trust that I might not be the best judge of what is succeeding, and believe that what sounds like stupid questions or silly roleplaying has an end goal I support.

Guys, I want to be a therapist, and I still have to work on suspending my initial, irritated reaction to CBT techniques.

Then how do we prevent the feeling of condescension? The image of the snooty psychotherapist, making judgemental notes and pontificating on childhood repression is still very much part of the public imagination. (Someday, I want a world where “penis envy? *snicker*” is not the rejoinder to my career plans…) “Here’s let’s try some roleplay to practice this social skill!” “Yes, but can you think of another way of looking at that?” …they don’t exactly make people disengage from that mental picture.

What about taking the opposite tack? Closing the distance a lot, laying many more cards on the table.

We say “Look. This is CBT. It’s been shown to help quite a bit for some people, and I think it could help you. Sometimes I’m going to ask you questions that sound condescending or silly or obvious, and I’d like you to step back and just answer them. Play along, as it were. Imagine that you’ve got to choose between sometimes being annoyed with what happens here, and not trying a therapy that’s evidence based. I’m asking you to pick the one that means you’ll sigh in frustration occasionally. And in return, I’ll do my best to frame questions well and make this a good experience of getting better.”

And then we really do see if there’s another way to see that situation.

*It says relationships. It’s 95% about marriages.

**composite, not from any of the books/writing I’m referencing in this post


  1. says

    I think your solution is good. I think any philosophy of therapy that involves the assumption that a patient should just trust the therapist to know what they’re doing regardless of whether or not that therapist explains themselves or gives any indication that they know what they’re doing is doomed to fail. A patient-therapist relationship is a relationship between two different types of expert — collaborators on a project with different specialties to contribute, as it were (my specialty is me, and my therapist’s specialty is psychology). Anything that would seem condescending or inappropriate in any other sort of collaborative endeavor is probably going to seem that way in therapy as well. A good coworker doesn’t do things on your project without telling you why or tell you to do things without telling you why.

    Related thoughts:

  2. Ysanne says

    Excellent article and good suggestion at the end. I couldn’t find a reason not to be open about the strategy of making people actually think through assumptions that seem too obvious and trivial to even mention.
    On a slightly different note, a friend tried to recommended Cognitive Behavioral Therapy to one of his students, and in the process googled “CBT” to explain a bit about it… ended up hilariously awkward, since most of the internet seems to use these 3 letters for something less therapeutic and definitely more NSFW.

  3. J B says

    The missing ingredient, I often find as a client, is empathy. There have been therapists who demonstrate they understand my feelings, and others who come off like scolds who are correcting me. The difference is whether they show empathy or not.

    I get to informally counsel plenty of people, just as a friend, and I never start off with the Socratic questioning (well, sometimes I do, and my anecdotal observation is, it doesn’t work as well). Instead I check to make sure I have caught whatever they are feeling and try to identify what need brings that feeling up. Usually it takes a few rounds of that, and my friend shows some sign of relaxation and then, if they want me to help alter their thinking (I ask), then maybe I work on talking about the patterns they are repeating that don’t help so much. So I’d alter your example this way:

    I, [therapist Y], started by trying to understand Jason’s needs and feelings.

    “So it sounds like you’re just feeling super worried about your livelihood. You want to make sure you can do a good job, so you can keep your ability just to have the basics of life that employment allows, is that right?”

    “Yeah, I’m terrified! What happens if I don’t have my job, what about my daughter, who’s sick, to say nothing of my prospects if I get fired!”

    “OK. Yes, that’s really scary territory. You want to know your boss could understand that things are hard for you?”

    sighs “Yeah. I just don’t feel like I can risk it.”

    It would go on for a few more rounds after that, probably hitting on self-respect, the bonds of family, and more, unique for each person even if there are similar themes for such a situation. The dialogue as written asks the client to put himself in the boss’s shoes. That’s very hard to do when you’re in fear. Once that and other feelings are addressed to my own satisfaction, then I can trust that a therapist gets where I’m coming from, and then I can take advice, because then trust has been established. Asking me to think about how the boss might think can even be interpreted as a betrayal of trust, which I’ve just brought a raw pile of feeling to be addressed.

    That’s it, but it makes a huge difference. I got it from Nonviolent Communication (which Google will tell you plenty about), and it’s very helpful. Unfortunately I’m unaware of any rigorous clinical trials or studies with NVC as a treatment modality, but they might be out there. Let me know if that makes sense.

    • Kate Donovan says

      I am familiar with NVC! For some reason I am thinking there’s some trials–let me poke about and see if I can find them.

  4. Robert B. says

    Huh. I never felt that reaction. My therapist asks like one or two REALLY GOOD questions every session. The rest is basically just pseudoquestions that invite me to keep talking and empathetic statements that validate my feelings. I wonder if he’s not using the type of therapy you’re talking about? I know a bit of psychology, but basically nothing about therapy technique.

    • Kate Donovan says

      It’s also possible that you two have some combination of personalities and techniques that prevents the issue I’m talking about. It’s not that I think CBT necessarily ends up that way, but because of the framework, it seems like there’s a higher likelihood that could be turning people off therapy entirely.

  5. Jason GL says

    Interesting post — I like that you’re thinking hard about how to be a more effective therapist.

    The pattern I’m hearing you describe is that you want to ask your patient a preliminary question to guide her toward a useful thought process, but the answer to that preliminary question is so obvious that asking the question comes off as condescending or tedious.

    As you point out, asking the patient to trust you (and the process) and apologizing for the fact that there will be some silly-sounding questions along the way can be helpful. Acknowledging that the questions are silly is certainly better than pretending the questions are normal.

    But what if you re-frame the questions so that they explicitly acknowledge that you both already know the answer? Instead of “Have you ever had a friend who got fired for respectfully asking for an extension?”, what if you said something like “I suppose it’s very unlikely that you’ve ever had a friend who got fired for asking for an extension. That hasn’t happened to you, has it?” Instead of “You said you wanted to restrict your calories after eating dinner two days in a row. Does that seem like it’s part of your plan for recovery?”, your therapist could say “You’re feeling an urge to restrict your calories, which seems like it might interfere with your plan for recovery. Does that worry you?” If pretending not to know an answer is annoying, the easiest solution might be to stop pretending not to know an answer — depending on your confidence level, you could downgrade an open-ended question to a request for confirmation, or you could even just answer your own question and then ask the patient how they feel about that answer. In my opinion, questions are most effective when they’re saved for occasions when you really do have some genuine uncertainty. If the only point of the question is to prompt the patient to ask herself the question (as a tool for reframing her thoughts), then you can explicitly say “When you find yourself feeling this way, I want you to ask yourself: ‘have any of my friends been fired because of the situation that’s worrying me?’ ” That way the patient knows that you’re teaching a tool, rather than pretending to ask a question.

    I also second what JB says about empathy — that’s good advice.

Leave a Reply

Your email address will not be published. Required fields are marked *