An Interview with Michelle Craske, PhD, on “Maximizing Exposure Therapy for Anxiety Disorders”

An Interview with Michelle Craske, PhD, on “Maximizing Exposure Therapy for Anxiety Disorders”

In my ongoing series of posts about exposure therapy, I’ve written several times about the work of UCLA professor Michelle Craske, PhD. Dr. Craske has been on the cutting edge of exposure research, and her work has undermined the traditional notion posited in in the emotional processing theory that habituation to a feared stimulus is important in exposure work.

The Society for a Science of Clinical Psychology has posted an interview that Dr. Jacqueline Persons conducted with Dr. Craske on “Maximizing Exposure Therapy for Anxiety Disorders.” This approximately 47-minute interview is an excellent introduction to and distillation of Dr. Craske’s work. The webpage also includes a pdf of a 2014 paper Craske authored summarizing this model.

Dr. Craske’s model focuses on using exposure to cultivate inhibitory learning to create new learning that overrides the fear-based association called excitatory meaning. The idea is that through repeated exposure, new learning occurs that contradicts the unhelpful more fear-based associations that result in anxiety and avoidance behavior.

3 considerations in the inhibitory learning model for conducting exposure with a client

In her interview, Dr. Craske emphasized three important things to consider in using exposure:

  1. What does this person really need to learn (i.e., inhibitory learning) in order to be less afraid of the fear stimulus (i.e., excitatory meaning) that will lead to improvement?
  2. How can the exposure exercise be designed to maximally benefit that learning; that is, how can the exposure exercise “violate” what that person expects will happen (e.g., usually something bad) based on the excitatory learning?
  3. New learning will occur both during the exposure exercise and long afterward through a period of consolidation. Consequently, whether someone’s distress diminishes (i.e., habituation) during the actual exposure exercise is not as important as the actual learning that may occur during and following the exposure.

Role of cognitive restructuring

Dr. Craske also had some interesting things to say about the role of cognitive restructuring in exposure work.  In traditional CBT, exposure work may be preceded by cognitive restructuring. According to Dr. Craske, preceding exposure work with cognitive restructuring may actually undermine the exposure work by softening the violation of the person’s expectations—the expectations based on the excitatory meaning of the feared stimulus. Consequently, Dr. Craske stated her lab will do cognitive restructuring after the exposure work as a means to help consolidate the new learning (i.e., inhibitory learning).

Check out the tables on the linked research paper!

Dr. Craske’s work can be a bit dense to read. Even if you don’t intend to read it, I encourage you to download the 2014 paper that is linked to the interview. Following the Reference section are 6 tables that offer concrete examples for using inhibitory learning to conduct exposure with specific anxiety-related problems and deepening the impact of the exposure. The paper also includes clinical examples and can serve as a useful reference for the interview.

In conclusion

If you have any interest in exposure at all, I encourage you to listen to this interview with Michelle Craske on “Maximizing Exposure Therapy for Anxiety Disorders.” She speaks very clearly and elegantly about her work on inhibitory learning in manner that is very accessible.

The link also allows you to download a research paper summarizing this work. Download it. Even if you’re not up to reading a 30-page research article, it provides a nice reference and includes clinical examples and tables with suggested wording for setting up exposure exercises.

A special “thank you” to Dr. Jacqueline Person for conducting the interview, and to the Society for a Science of Clinical Psychology for posting it!

Brian Thompson Ph.D.

Author: Brian Thompson Ph.D.

Brian is a licensed psychologist and Director of the Portland Psychotherapy Anxiety Clinic. His specialties include generalized anxiety, OCD, hair pulling, and skin picking.

More Avoidant Therapists Are Less Likely to use Exposure in OCD Treatment

More Avoidant Therapists Are Less Likely to use Exposure in OCD Treatment

In a previous post, I linked to a blog post about therapist reluctance to use exposure therapy. Exposure therapy is one of the most effective interventions for anxiety-related problems but, sadly, remains under-utilized by clinicians. It is an intervention that has been a major focus on my posts on this blog.

A recent study out of Drexel University looks at the role of experiential avoidance in therapist reluctance to use exposure therapy to treat OCD. Experiential avoidance (EA) refers to a tendency to avoid uncomfortable thoughts, feelings, and bodily sensations, and is a cornerstone in the model of psychopathology in Acceptance and Commitment Therapy.


The researchers recruited 172 clinicians who identified as cognitive behavioral therapists. Each participant completed self-report measures of EA, attitude towards evidence-based practice, thinking style, and treatment approaches. These therapists watched 2 of 4 possible video vignettes featuring actors portraying people with OCD based on scripts that were vetted by OCD experts. The therapists then rated how much time they would allot for different therapy techniques in treating the case example.


The researchers found that therapists who scored higher on experiential avoidance (EA) reported they would devote less time to using exposure for treating OCD clients in the vignettes they watched. This is striking as exposure is the gold standard treatment for OCD. Additionally, therapists who scored higher in EA showed a lower preference towards evidence-based practice. An interesting gender finding was that women exhibited a greater preference for evidence-based practice than men.

As the authors note, it’s impossible to derive a causal relationship between these variables. For example, they speculate that therapists who are more experienced in exposure may develop lower EA because of they themselves get used to doing exposure. In support of this view, the authors note that therapists who spent a larger portion of their time doing clinical work exhibited lower EA.

Overall, this study suggests that higher experiential avoidance may be a barrier to using exposure therapy and evidence-based practices.

Check out the article!

If you’re a member of the Association for Contextual Behavioral Science, you can download a copy of the article here if you log into your account.

Cognitive Behavioral Therapy Superior to Antipsychotics for OCD

Cognitive Behavioral Therapy Superior to Antipsychotics for OCD

A practice I’m seeing more often that concerns me is the addition of antipsychotic medications on top of antidepressants when the antidepressants aren’t working. If someone isn’t showing improvement on an antidepressant alone, a prescriber may add an antipsychotic medication—the idea being it will increase the effectiveness of the antidepressant. The research for this is a little questionable, especially as the side effects for antipsychotics can be pretty bad. I’ve felt strongly enough about this issue that I wrote an editorial about it that the Oregonian published in 2012.

Antipsychotics and obsessive-compulsive disorder

In previous post, I wrote about a study that found that giving an antipsychotic in people with posttraumatic stress disorder (PTSD) provided no additional improvement. A recent study looked at whether adding an antipsychotic medication would be helpful to people with obsessive-compulsive disorder (OCD). Results are extremely clear that the answer is, “No!”

As I’ve written before, the most effective treatment for OCD is cognitive behavioral therapy (CBT) with exposure and response (or ritual) prevention (EX/RP). (Note: in other posts, I abbreviate exposure and response prevention as “ERP” but use “EX/RP” here to remain consistent with the article.) There is some research that suggests that antidepressant medication can have a small impact on OCD-related problems, but EX/RP remains the gold standard treatment

Another study showing that CBT does the best with OCD

A 2013 study in JAMA Psychiatry examined a group of people with OCD who were already taking an antidepressant but were still experiencing moderate or worse OCD symptoms. These individuals were divided into 3 treatment groups.

  1. Some received psychotherapy—cognitive behavioral therapy with EX/RP.
  2. Some received an antipsychotic—Risperidone.
  3. Some received a placebo (i.e., inactive) pill.

CBT with ERP was much more effective

The results were striking. The researchers found that only 23% of people showed improvement on the antipsychotic; moreover, this result is even less impressive given that 15% showed improvement on the placebo (e.g., sugar pill). In fact, statistical analysis suggests there was no difference between the antipsychotic and the placebo—this means that the antipsychotic and a sugar pill performed about equally.

By contrast, 80% of people who received cognitive behavioral therapy with EX/RP improved.

80% vs. 23% is a big difference, especially since the latter is more of a placebo effect than a response to an active treatment.

One quibble

In the Conclusion section of the abstract, the writers make a subtle statement that really bothered me:

Patients with OCD receiving SRIs who continue to have clinically significant symptoms should be offered EX/RP before antipsychotics given its superior efficacy and less negative adverse effect profile.” [bolding is mine.]

This statement implies that, even though EX/RP is superior to antipsychotics, that antipsychotics are still a viable treatment. This seems a bit disingenuous, however, as the researchers’ own analyses indicate that whatever improvement antipsychotics demonstrated was likely a placebo effect.

If anything, a sugar pill should be offered before an antipsychotic since they are equally effective, and the former has fewer side effects.

Concluding thoughts

Although I think this is an important study because it makes it clear that adding antipsychotic medication is unlikely to be of much help for someone with OCD, the superiority of ERP over medication for OCD isn’t new information.

There’s already a solid base of research that suggests the EX/RP is superior to antidepressant medication for OCD. Giving an antidepressant to someone receiving EX/RP for OCD neither helps nor hinders treatment. This study makes it pretty clear that antipsychotics should not be considered for people with OCD.

What Kind of Exposure Do You Want, Goldilocks?

What Kind of Exposure Do You Want, Goldilocks?

Understanding factors that contribute to outcome are crucial as we continue to refine treatments and revise the theories that underlie them. In a study published in the Journal of Anxiety Disorders, Norton and colleagues (2011) examined the role of activation and habituation in exposure therapy. The rationale for study was based on the emotional processing theory, which I’ve written about previously.

By activation, the authors mean how distressed the person becomes during the exposure exercise. Habituation refers to the reduction in distress when someone is confronted with a fear inducing stimuli. I’ve written in greater detail about habituation and some problems in using it as a marker of change. In this study, the focus was on within-session habituation—the degree to which distress reduced during a particular exposure session. In this study, the authors appear favorably disposed towards the emotional processing theory while still acknowledging research that does not support some of its proposed mechanism of action.

The Set-up

The sample consisted of 106 people who were enrolled in studies of a transdiagnostic protocol for a cognitive behavioral treatment for anxiety developed by the first author. The 12-session protocol included cognitive restructuring as well as exposure. According to the article, the protocol can be adapted to a variety of anxiety-related problems. Initial sessions begin with psychoeducation, self-monitoring, and cognitive restructuring, before shifting into 6 sessions of exposure therapy. The remaining sessions shift back to cognitive interventions before ending with relapse prevention. Because of missed session and treatment incompletion, the researchers focused their analyses on the first three exposure sessions.

How Do Activation and Habituation Relate to Dropout and Treatment Outcome?

As the authors examined habitation and activation across three exposure sessions, I’ll step back and focus on the broad findings of the study.

1. Clients were more likely to drop out of treatment if their anxiety was too high or too low during the first exposure session. Dropouts were also higher among those who didn’t show a decrease in anxiety (i.e., habituation) during the first session. (These trends weren’t found in the second and third exposure sessions.)

What does this mean? It’s hard to say for certain, but the findings suggest people with really high anxiety during the first exposure session who don’t experience a reduction in anxiety during the exposure may be at a greater risk for dropping out of treatment. In my mind, it might be worth spending some time preparing clients for this possibility, perhaps even normalizing it.

2. Contrary to what the emotional processing theory predicts, people with lower anxiety overall during exposure tended to have better outcomes at the end of treatment.

This is tricky to interpret, too. The authors suggest that too much anxiety may inhibit treatment, which is not inconsistent with the emotional processing theory.

Should Activation Be “Just Right”?

With so many variables (e.g., three exposure sessions, high/low activation, habituation, dropout, symptom reduction), we should be cautious in interpreting these findings. Should anxiety during exposure like Goldilocks and the Three Bear—not too high, not too low, but just right? Or alternatively, people with high anxiety the first time may be frightened off by treatment, and people with low anxiety may not have been engaged to begin with. Unfortunately, these variables were not manipulated experimentally, so it’s difficult to know what causes what.

Painting in broad strokes here, it does appear that how a client experiences the first exposure session is important. It may be worth processing a client’s experience afterward, responding to any questions or concerns, and renewing commitment to treatment. This study also illustrates how complicated it is to try to understand potential variables when conducting exposure therapy.

Treating Panic Disorder with Comorbidities: Why Focusing on the Panic May Be the Best Option

Treating Panic Disorder with Comorbidities: Why Focusing on the Panic May Be the Best Option

A common reason psychotherapists give for ignoring research is: “Research studies usually focus on one problem—the people I see often have multiple problems.” Whether this is reason enough to abandon evidence-based practice is debatable. The statement, however, does touch upon a very real concern: how do we choose a treatment focus for clients with multiple problems? The answer to this question isn’t always clear, but it is the responsibility of practitioners to see whether there is research that can guide a decision. In the case of panic disorder, research suggests clients will get the most bang for their buck by focusing on the panic.

A Little About the Study

It’s not a new study—it was published in 2007—but I thought it was really interesting when I originally read it and haven’t had a chance to write about it yet.  The first author is Dr. Michelle Craske, a professor at UCLA, and one of the most renowned anxiety researchers.

In this study, sixty-five people with panic disorder were randomly assigned to one of two groups. Everyone in the study received 12 weekly sessions of a manualized group cognitive behavioral treatment for panic disorder as well as 6 adjunct individual sessions spaced every two weeks. For half the participants, the individual sessions reinforced what was taught in the panic disorder group; for the other half, the individual session provided tailored cognitive behavioral treatment for a co-occurring disorder. The most common co-occurring conditions were major depressive disorder, generalized anxiety disorder, social anxiety, and specific phobia.

The individual sessions were scheduled every two weeks in order to simulate therapist “straying” from treating one condition to another. For example, someone with comorbid depression in the experimental group would receive cognitive behavior therapy for depression in the individual group, whereas someone in the control group would receive additional panic disorder treatment in the individual group. As a consequence, the control group only received panic disorder treatment regardless of comorbidities, and the experimental group received panic disorder treatment and individual sessions targeting a specific comorbidity.

The researchers assessed participants before treatment, post-treatment, and at 6- and 12-months following treatment.

“More of the Same” Wins Out

The results are by no means a slam dunk, but they suggest that people who only received panic disorder treatment did better—both for panic and the comorbid condition—than people who were treated for panic and a comorbid problem. Some of the effects were small, and superiority for panic treatment-only wasn’t across the board; however, the results suggest that not only can focusing on panic disorder result in improvements in comorbid conditions even if the co-morbid conditions aren’t directly addressed, but that it may be a better option than trying to address both. As the authors put it, “the results raise the interesting possibility that staying focused is superior to straying” (p. 1106).

Take Home Message

For therapists who are skeptical that conclusions drawn from research studies are applicable in their own work, this study probably didn’t win anyone over. The attempt at mimicking how an actual therapist would deal with comorbid conditions—devoting a session to it every two weeks—is a little contrived (although very creative!). I can see how someone might dismiss this single study.

My take? Given that panic disorder is one of the most treatable conditions in the research literature, I think this study offers some compelling evidence for treating the panic first. Once clients have learned to manage their panic, other problems may resolve on their own. If any problems remain after successful resolution of panic, then these problems may then be addressed. In a nutshell: for people with panic disorder and co-occurring conditions, it’s probably best to treat the panic first.

From Our Blog