[Register for Next Event][Sampling of Past Meetings][Member Benefits & Activities]
[Articles on Cognitive-Behavior Therapy][Join NJ-ACT][
Member Directory][Renew Membership]

Behavior Therapy vs. Cognitive Therapy
For Depression: Here We Go Again

A Review of “Depression in Context”
By Christopher Martell, Michael Addis, and Neil Jacobson

 By Milton Spett

In the mid 1970s Peter Lewinsohn developed the behavioral theory of depression. Lewinsohn argued that the essence of depression is a low rate of behavior, and this low rate of behavior causes all the other symptoms of depression. Lewinsohn also hypothesized that the low rate of behavior resulted from a lack of reinforcement from the environment. This was pure Skinnerian theory: teach the depressed patient to elicit higher rates of reinforcement, and the depressed patient’s rate of behavior will increase, causing the depression to lift.

Meanwhile Aaron Beck was developing the cognitive theory of depression. Beck argued that the essence of depression is not a low rate of behavior, but three core schemas: 1) The self is defective or inadequate; 2) All experiences are defeats or failures; and 3) The future is hopeless. Beck also asserted that depressed people use faulty information processing, like selective attention, to maintain their negative core schemas in spite of evidence to the contrary. According to Beck’s cognitive theory, therapists should use both cognitive and behavioral techniques to modify the depressed patient’s core schemas. Once the core schemas are modified, the symptoms will remit and will not return.

Like the Gulf War, the war between the behavioral and cognitive views of depression was a rout. Within a few years virtually everyone agreed with Beck that depressogenic schemas cause depression, and both cognitive and behavioral techniques should be used to change these depressogenic schemas. The ensuing peace was so amicable that the former adversaries joined forces and called themselves “cognitive-behaviorists.” And there, for 25 years, the matter rested.

But now, as the 21st century rolls in, along come Neil Jacobson and his colleagues, brandishing new research data, and arguing that Lewinsohn was right after all: cognitive interventions to modify core schemas are inefficient and unnecessary for treating depression. According to Jacobson, all you really need are the behavioral techniques he calls Behavioral Activation (BA), interventions which are remarkably similar to those proposed by Peter Lewinsohn a quarter century ago.

What is Behavioral Activation?

BA focuses on the consequences of behavior. Behavior that helps patients achieve their goals is encouraged, and behavior that interferes with achieving goals is discouraged, a true operant conditioning paradigm. For example, suppose a patient believes that everyone hates him. A cognitive therapist would try to convince the patient that this wasn’t true. But a BA therapist would ask the patient, “What are the consequences of this belief?” The consequences of this belief are avoiding people and not having friends. But the patient wants friends, and recognizes that avoiding people prevents him from having friends. So the BA therapist encourages the patient to interact with people in order to achieve his goal of having friends. The BA therapist may also assert that believing everyone hates him is counter-productive because this belief leads to avoiding people and not having friends. But note that unlike a cognitive therapist, a BA therapist would not try to convince the depressed patient that he is wrong in believing that everyone hates him.

BA emphasizes idiographic interventions, encouraging each patient to perform those behaviors that will improve the patient’s mood and achieve the patient’s goals. If a depressed patient stays in bed in the morning in order to avoid the pain of getting up, going to work, and facing failure, guess what intervention the BA therapist makes. That’s right, “What are the consequences of staying in bed?” Of course the patient reports feeling worse, and the BA therapist suggests that she try out a new behavior, such as getting up and going to work, to see if that makes her feel better. The BA therapist will also point out that staying in bed may cause her to lose her job and feel even more like a failure.

BA also asserts that patients must keep struggling until their old avoidance habits are replaced by new active habits. BA uses traditional behavioral techniques including mastery and pleasure activities, graded task assignments, activity charts, covert rehearsal, problem solving, and goal setting. The authors illustrate their concepts with some cute acronyms, such as ABC, TRAP, and ACTION, the details of which need not concern us here.

Let’s Go to the Research

Jacobson and his associates compared “up to 20 sessions” of BA with cognitive-behavioral therapy (CBT) that included BA techniques plus cognitive interventions aimed at modifying core schemas. (A third treatment group included BA plus modification of automatic thoughts, but no attempt to modify core schemas.) The results are reported in Jacobson, et al. (1996) and Gortner et al. (1998). At post-treatment, and at 6, 12, 18, and 24 month follow-up, there were no significant differences between BA and CBT. The authors convincingly demonstrate internal validity as well as lack of experimenter bias. Jacobson and his colleagues cite this study as evidence that their BA can achieve results that are equal to CBT, and because their BA is simpler and easier to learn and administer, it is preferable to Beck’s cognitive-behavioral approach.

The authors also note that the BA group exhibited as much change in core schemas as did the CBT group. They speculate that perhaps BA is just as effective as CBT in its ability to modify core schemas. But I have another explanation. I believe that “up to 20” sessions of BA is just as ineffective as CBT in modifying core schemas. Perhaps 20 sessions of any therapy just isn’t enough to make significant changes in maladaptive core schemas that have been reinforced over decades. I believe that 20 sessions of BA or CBT can alleviate a current episode of depression, but do little to prevent future depressions because neither treatment modifies the core schemas that Beck cites as the underlying cause of depression.

And the data support this interpretation. 50% of the BA patients were listed as “recovered” (Beck Depression Inventory scores of less than 9) at post-treatment, and half of this 50%, had relapsed 24 months later. This suggests that only 25% of the BA patients recovered and remained undepressed for at least two years. The figures for the CBT group are similar. The lack of a no-treatment or placebo control group prevents us from determining if either therapy was more effective than spontaneous remission in preventing future depressions, but the landmark NIMH depression study did have a placebo control group as well as an 18 month follow-up. The NIMH study (Shea et al., 1992), found that 28% of the CBT patients and 18% of the placebo patients were free of depression after 16 weeks of treatment and remained free of depression without additional therapy for at least 18 months. In other words, 16 sessions of CBT alleviated depression and prevented a recurrence for 18 months, over and above spontaneous remission and placebo effects, in about 10% of the patients. The conclusion of the NIMH authors (p. 782) is similar to my critique of the Jacobson study: “…16 weeks of [CBT and other treatments] is insufficient for most patients to achieve full recovery and lasting remission.”

In my clinical practice I have successfully treated a number of patients for depression, using cognitive-behavioral and other techniques. I assume these treatments were successful because by the end of treatment, my patients were able to remain undepressed after experiencing situations which had previously provoked depressive episodes. Several of these patients have contacted me again, months or years later, but always to say “hello” or ask for guidance in dealing with some new problem. As far as I know, none has had a recurrence of their depression. All of these patients had been in therapy for a minimum of two years.

What About Changes in Core Schemas?

Jacobson’s study found significant improvement in measures of core schemas for both the BA group and the CBT group. But several other studies (see, for example, Gilboa & Gotlib, 1997) have demonstrated that when depression remits without treatment, the schemas that were apparent during the depressive episode are neither experienced nor apparent in paper and pencil measures. However, these negative schemas can be evoked by putting the formerly depressed subjects into a sad mood. So it is quite possible that short-term BA or CBT do cause depressive symptoms to remit in some patients, but do not affect the core schemas beyond causing the schemas to become dormant and inaccessible to experience, measurement, or psychotherapy. But even though dormant, the schemas remain, making these patients susceptible to future depressions.


If psychotherapy must be short-term, the work of Jacobson and his colleagues, if replicated, suggests that BA may be just as effective (or ineffective) as CBT. BA does appear to be a powerful technique which should be emphasized in any treatment of depression, especially during the initial phases. But most depressed patients require longer term treatment if they are to recover and remain recovered, and there is currently no research demonstrating that behavioral activation or any therapy can accomplish this. In the absence of evidence to the contrary, we must follow common sense and clinical experience, which both suggest that long-term CBT is the treatment of choice for most depressed patients.


Gilboa, E. & Gotlib, I.H. (1997). Cognitive biases and affect persistence in previously dysphoric and never-dysphoric individuals. Cognition and Emotion, 11, 517-538.

Gortner, E. T., Gollan, J. K., Dobson, K. S., & Jacobson, N.S. (1998). Cognitive-behavioral treatment for depression: relapse prevention. Journal of Consulting and Clinical Psychology, 66, 2, 377-384.

 Jacobson, N. S., Dobson, K. S., Truax, P. A., Addis, M. E., Koerner, K., Gollan, J. K., Gortner, E., & Prince, S. E. (1996). A component analysis of cognitive-behavioral treatment for depression. Journal of Consulting and Clinical Psychology, 64, 295-304.

 Shea, M., Elkin, I., Imber, S., Sotsky, S., Watkins, J., Collins, J., Pilkonis, P., Beckham, E., Glass, D., Dolan, R., & Parloff, M. (1992). Course of depressive symptoms over follow-up. Archives of General Psychiatry, 49, 782-787.

2004 NJ-ACT. All rights reserved.
Reproduction of any documents, related graphics, or any other material from this World Wide Web site is strictly prohibited without permission from the organization. For permission, click here to email us.