When I first heard about studies using smartphones to treat anxiety with cognitive therapy I was intrigued, to say the least. However, I had a misconception about what that actually meant. My assumption was that the smartphone app would be automating some basic cognitive therapy, a virtual therapist that could give some reflective feedback and also give basic cognitive tools to deal with anxiety. That sounded like it might be useful, at least for mild cases, and I hoped that the app was designed to refer severe cases to an actual therapist.

I had already been very interested in the concept of online, virtual, or computer-based therapy. It seems like this is coming, but of course it needs to be researched to see how it works and for which patients.

But that is not what the smartphone app is at all. Rather it has to do with a treatment technique called cognitive bias modification (CBM). This therapy is based on research that finds that those with social anxiety have a cognitive bias which makes them attend more than others to signs of threat or to negative emotions. Further, they have a cognitive bias to interpret ambiguous social cues as hostile or negative. This raises a cause and effect question – are they anxious because they have these cognitive biases, or does the anxiety make them attend to negative emotions and interpret emotions negatively. Perhaps it is both, in a reinforcing feedback loop.

There is some evidence from prospective studies that cognitive biases predict future anxiety, suggesting (but not proving) a cause and effect. Another way to address this question, and perhaps to develop a treatment for social anxiety, is to target these cognitive biases directly, rather than addressing possible underlying causes of the biases. That is where the smartphone app comes in. There are two computer-based treatments designed to directly treat these cognitive biases: CBM-Attentional and CBM-Interpretive (CBM-A and CBM-I respectively).

For CBM-A subjects are made to look at a computer screen (or now a smartphone screen) with two faces displayed, one neutral and one disgusted or angry. They are told to note and remember the letters that appear on the screen. A letter than appears where the neutral face was located, drawing their attention toward the neutral face and away from the face displaying a negative emotion. This is supposed to train their brain away from attending to negative emotions – modifying their cognitive bias for attention.

For CBM-I subjects are asked to interpret word-sentence associations, and are given positive feedback for benign interpretations and negative feedback for negative interpretations. Again this is supposed to modify their bias away from negative interpretations.

As always, we like to evaluate any new treatment paradigm on two basic criteria – is it plausible, and does it work better than placebo. Plausibility is a little challenging to assess. There are certainly no laws of physics, chemistry, or basic biology involved. Plausibility depends on whether or not you think this top-down approach to brain training can have a significant and lasting effect on our thinking and emotions. Personally I am not convinced that this approach has much value. It all seems like the “magic wand” approach to therapy – rather than addressing a complex behavior with an approach that reasonably addresses that complexity, it focuses on one perhaps superficial aspect of something as complex as social anxiety.

I have not been impressed with the whole “brain training” approach to cognitive therapy, such as EMDR (eye movement desensitization and reprocessing). At best it seems like treating the symptom of a disease rather than the disease itself. But I could be wrong. Sometimes the symptoms are the disease. In some chronic pain conditions, for example, the pain is the problem, and treating the symptom of the pain may be a reasonable and effective approach. Perhaps at its core social anxiety is being driven by some flaw or bias in brain function that can be tweaked by simple training.

My personal bias, therefore, is that this kind of approach has low (but not very low) plausibility but I am willing to be convinced by reasonable clinical evidence. So what does the clinical evidence show?

There are many pilot studies of CBM (both CBM-A and CBM-I and a couple with combined treatments). A review published in February 2011 concluded:

Although the potential clinical utility of CBM is quite exciting, the existing data do not address a number of limitations. First, the majority of the evidence of CBM’s effect on cognitive bias and anxiety relies on analogue samples and brief (one session) experiments. The field is in need of RCTs to test treatment protocols in clinical samples. A related issue is that all existing RCTs represent researchers’ initial pilot studies rather than large-scale RCTs. Therefore, they comprise of relatively small samples (subject numbers ranging from 29 to 44). Effect sizes from small studies are unreliable [93]; thus we await the results of larger, definitive trials.

While positive, the evidence is still preliminary and there needs to be more rigorous trial design before we can conclude that there is a real specific effect here. My concern with the research to far is that we are just seeing the non-specific effects of treatment intervention, rather than specific effects of CBM.

In a recent New York Times article on the topic, one specifically discussing the smartphone app for administering CBM, the results of perhaps the largest study to date are discussed. The study is yet to be published, but the researchers reported their results so far as:

Participants who got the treatment improved their scores on a questionnaire measuring anxiety, dropping by an average of 22 points, compared with an 8-point drop among people in a “waiting list” group, who got no computer games to play. However, a placebo group in the study practiced with a two-face video program not intended to shift the eyes from one face or the other, and their anxiety levels as measured on questionnaires also fell by about 22 points, just as they had for those who got the treatment.

The comparison to a waiting group showed a significant difference, but to a more rigorous and blinded control showed no difference at all. While the article indicated that these results were “confusing” I don’t find them confusing at all – they are dead negative. The only comparison that matters is the blinded one to a reasonable placebo treatment.

Further the article refers to a review from the University of Pennsylvania that found evidence of publication bias:

The authors noted that there was evidence of what scientists call a “file drawer” problem — in which studies finding no effect are filed away or ignored, while encouraging ones are published. “I think in this field the standards for publishing positive studies are lower than for negative ones,” Dr. Van der Does said in an e-mail.

To summarize: preliminary evidence is mixed but tending toward the positive, but has acknowledged serious limitations and evidence of publication bias. Preliminary results from a large and well-controlled study are negative.

I agree with various reviewers who conclude that this treatment is interesting and deserves more research. I would like to see some rigorous large studies – the kind that are really definitive. I sense a great deal of excitement among researchers. For example, from a recent pilot study the authors write:

Excitingly, these procedures have been shown to reduce bias in attention to threat (CBM-A), and to promote a positive interpretive bias (CBM-I) in anxious populations; furthermore, these modifications are associated with reductions in anxiety. We believe that these techniques have the potential to create a real clinical impact for people with anxiety.

The excitement may be premature, however. I hope this therapy is found to have potential, because it can result in a cheap and convenient treatment method. All the more reason to move beyond pilot studies are perform some rigorous studies that can really answer the question as to whether or not the specific elements of CBM-I and CBM-A, alone or in combination, can have long lasting benefits for social or other forms of anxiety.


Posted by Steven Novella

Founder and currently Executive Editor of Science-Based Medicine Steven Novella, MD is an academic clinical neurologist at the Yale University School of Medicine. He is also the host and producer of the popular weekly science podcast, The Skeptics’ Guide to the Universe, and the author of the NeuroLogicaBlog, a daily blog that covers news and issues in neuroscience, but also general science, scientific skepticism, philosophy of science, critical thinking, and the intersection of science with the media and society. Dr. Novella also has produced two courses with The Great Courses, and published a book on critical thinking - also called The Skeptics Guide to the Universe.