It may seem to go without saying that CT's attempts to modify automatic thoughts, dysfunctional beliefs, and cognitive distortions are responsible for CT's effects and that therefore, that it is important for us to spend time in therapy working on dysfunctional cognitions. However, there are many other aspects of CT that could be responsible for its effectiveness. For example, CT includes "non-specific" factors such as the therapist's empathy and includes many behavioral interventions along with the cognitive interventions. Thus, studies that show that CT works do not necessarily provide evidence that cognitive change is important.

Studies that show that dysfunctional cognitions decrease when depressed individuals are treated with medication alone have led some to conclude that decreases in dysfunctional cognitions are an effect of decreases in depression, not a cause of decreases in depression. In fact, some (Steve Hayes comes to mind) argue that the types of changes in cognition that CT tries to accomplish are unnecessary or are counterproductive.

At this point, we have many studies that show that CBT is effective with a broad range of problems but many fewer studies that examine whether the cognitive changes that CT seeks to achieve have an important influence on outcome. Fortunately, a number of studies have been designed to examine the role that cognitive change plays in the effectiveness of CT.

In a recent study of the multidisciplinary treatment of chronic pain, Burns and his colleagues (Burns, et al, 2003) used a cross-lagged panel design to examine whether cognitive changes were a cause of improvement or an effect of improvement. They found that early-treatment changes in catastrophising and helplessness predicted late-treatment changes in outcome measures but not vice-versa. They also found that early-treatment depression changes predicted late-treatment activity changes but not vice versa. These results provide support for the hypothesis that changes in cognition are a cause of improvement rather than being a result of improvement.

In another recent study, Beevers et al (2003) examined the relationship between degree of cognitive change during treatment for depression and rate of relapse. In a sample of individuals who had achieved at least a 50% improvement in depressive symptoms, the researchers found that both a poor change in dysfunctional attitudes and a poor change in extreme thinking predicted a faster return of depressive symptoms.

Note: This isn't a systematic review, just a couple of studies I've encountered recently, but at least is gives us some support for the idea that achieving cognitive change is important.

Beevers, C. G., Keitner, G. I., Ryan, C. E. & Miller, I. W. (2003). Cognitive predictors of symptom return following depression treatment. Journal of Consulting and Clinical Psychology, 71, 488-496.

Burns, J. W., Kubilus, A., Bruhel, S., Harden, N. & Lofland, K. (2003). Do Changes in Cognitive Factors Influence Outcome Following Multidisciplinary Treatment for Chronic Pain? A Cross-Lagged Panel Analysis, Journal of Consulting and Clinical Psychology, 71, 81-91

You'r discussion of the causal connection between thinking and emotion is revealing. Upon 2nd reading it was clearer you're noted research appears to support the notion that a change of - mind (distorted thinking) may precede changes in emotion and ultimately adjustment. One thing that has impressed me over time is how slow going the use of disputation seems to be. It has occurred to me though, if George Kelley's premise is true that our constructs are layered upon a fairly rigid hiearchy, changing one habit of mind may require something akin to a seismic shift in prioritizing perception and motivation. Your note that some see CT strategies as counterproductive might be involved with this predicament of disputing the wrong layer of one's constructs and having the consequence of firming up resisitance.

As you suggest, there can be a variety of reasons that it is "slow going" when we try to change the client's dysfunctional cognitions. Here are a number of points that come to mind:

1. You refer to using "disputation" to change dysfunctional cognitions. One disadvantage of using REBT-type disputation is that this tends to intensify resistance. One of the reasons that Ct emphasizes guided discovery (Socratic questioning) is that this approach tends to reduce resistance rather than intensifying it.
2. Often, the client's dysfunctional cognitions are strongly held. In addition, the client may have had many experiences that seem to support their dysfunctional cognitions. It is not easy (or quick) to change such cognitions.
3. Sometimes clients fear the consequences of cognitive change (i.e. "If I give up my perfectionism I'll become mediocre") and resist giving up the dysfunctional cognitions until these fears are addressed.
4. Sometimes Cognitive therapists slip into relying solely on verbal, in-the-office interventions. CT is much more effective when verbal interventions are coupled with experiential and behavioral interventions. Behavioral experiments often are much more effective that intellectual discussion.
5. Some dysfunctional cognitions are harder to challenge than others. A dysfunctional belief such as "If I ask for what I want, no one will take me seriously" is easy to test, "I'm unloveable" is harder to test, and "If I don't pray right, I will roast in Hell" may be quite hard to challenge effectively.
6. ...

Obviously, I think that there are quite a few reasons that cognitive interventions may go slowly. One important time when there is "slow going" with cognitive interventions is when some other mode of intervention is more appropriate. Remember that the Thought Sheet isn't the only tool that cognitive therapists have. If the client has OCD and we try to have them generate rational responses to each of their obsessions, this is likely to be ineffective or counterproductive. Exposure and response prevention is likely to be much more effective.

There are some problems, such as depression, where we spend quite a bit of time working directly to change dysfunctional cognitions. However, there are other problems where we spend much more of our time working towards behavior change, helping clients improve their coping skills, working towards acceptance, etc.

James, you contrast REBT-type 'disputation' with 'Socratic questioning'. However, for a good twenty years or more the Socratic method has been recommended by REBT researchers and writers as the most appropriate way to conduct disputation. The term 'disputation' is often misunderstood to mean directly contradicting or arguing with clients, which is usually counterproductive. I suspect that this misunderstanding arises partly from the use of the term 'disputation' itself. REBT continues to use it, though, because it begins with 'D' and thus conveniently fits into the ABC model. The other reason may be that the founder of REBT, Albert Ellis, tends to be rather didactic in his approach to resructuring cognitions, and practitioners who are aware of this - perhaps through seeing a video of Ellis in action - assume that all REBT therapists use the same approach. According to modern REBT, the most effective approach to disputation is Socratic.

It is not easy to define terms precisely and then get everyone to agree to the definition (no matter what one's theoretical orientation is) and it certainly would be a good thing if there were more of a consensus on how important terms are defined. However, I think CT is fairly clear about what we mean when we talk about "Socratic questioning" or "guided discovery."

In Cognitive Therapy of Depression (1979, pp.69-71) Beck and company say:

"Use Questioning Rather Than Disputation and Indocrination

A well-timed, carefully phrased series of questions can help the patient to isolate and carefully consider a specific issue, decision, or notion. A series of questions may open the patient's thinking around a specific issue, and thereby allow him to consider other information and experiences - either recent or past. The patient's curiosity can be raised by a series of questions. His apparently rigidly stated views become tentative hypotheses. In this way, questions are used to relieve depressed, constricted thinking.
-snip-
Questions constitute an important and powerful tool for identifying, considering, and correcting cognitions and beliefs. As with other powerful tools, they can be misused or artlessly applied. The patient may feel he is being cross-examined or that he is being attacked if questions are used to 'trap' him into contradicting himself. In addition, open-ended questioning sometimes leaves the patient in the defensive position of trying to guess what the therapist 'expects' for an answer. Questions must be carefully timed and phrased so as to help the patient recognize and consider his notions reflectively - to weigh his thoughts with objectivity." (They also give three good examples of how we use questioning on pages 68-70.)

A more contemporary discussion of how we use questioning can be found in the summary of the princilpes of CT that Dr. Beck and I included in our chapter for the second edition of Major Theories of Personality Disorder (Lenzenweger & Clarkin, 2004):

"The approach used in Cognitive Therapy has been described as "collaborative empiricism" (Beck, et al., 1979, Chap. 3). The therapist endeavors to work with the client to help him or her to recognize the factors that contribute to problems, to test the validity of the thoughts, beliefs, and assumptions which prove important, and to make the necessary changes in cognition and behavior. While it is clear that very different therapeutic approaches ranging from philosophical debate to operant conditioning can be effective with at least some clients, collaborative empiricism has substantial advantages. By actively collaborating with the client, the therapist minimizes the resistance and oppositionality which is often elicited by taking an authoritarian role, yet the therapist is still in a position to structure each session as well as the overall course of therapy so as to be as efficient and effective as possible (Beck, et al., 1979, Chap. 4).

One part of this collaborative approach is an emphasis on a process of "guided discovery." If the therapist guides the client by asking questions, making observations, and asking the client to monitor relevant aspects of the situation, the therapist can help the client develop an understanding of his or her problems, explore possible solutions, develop plans for dealing with the problems, and implement the plans quite effectively. Guided discovery has an advantage over approaches in which the therapist unilaterally develops an understanding of the problems and proposes solutions in that it maximizes client involvement in therapy sessions and minimizes the possibility of the client's feeling that the therapist' ideas are being imposed on the client. In addition, since the client is actively involved in the process of developing an understanding of the problems and coming up with a solution, the client also has an opportunity to learn an effective approach to dealing with problems and should be better able to deal with future problems when they arise."

Beck, A. T., Rush, A. J., Shaw, B. F., & Emery, G. (1979). Cognitive therapy of depression. New York: Guilford Press.

Pretzer, J. L. & Beck, A. T. (2004). A cognitive theory of personality disorders. In: M. Lenzenweger & J. Clarkin (Eds.) Major theories of personality disorder (2nd Edition). New York: Guilford Press.

An idea that seems implicit, or "between the lines" in the notion of collaborative empiricism is that the therapist is helping the client achieve more coherent cognition - that is, thoughts that agree with more stable beliefs. ONce the client recognizes that s/he is thinking things s/he doesn't believe, there is motivation and progress...

Has this notion - of therapy as promoting coherence / reducing contradictoin - been developed explicitly by anybody in the CT literature?

This is an interesting concept that I haven't seen developed in the literature. However, there is a lot that I haven't read so I may have missed it.

The idea that the coherence (or lack thereof) of one's cognitions is what's important sounds like the sort of concept that the constructivists would like a lot. More traditional CT authors (among whom I must count myself) would argue that the point to collaborative empiricism is to test cognitions against external reality so that the client ends up with cognitions that are more realistic and incidentally are more coherent.

My bet is that some of the constructivists would argue that external reality is unknowable, that all meaning is constructed by the individual, and that coherence is what's important.

If you're looking for a dissertation topic, this may have potential.

I´m afraid it´s not OK the reference of the first article. I have found at
Beevers, Christopher G.1,2,4; Keitner, Gabor I.1,3; Ryan, Christine E.1,3; Miller, Ivan W.1,2. Cognitive Predictors of Symptom Return Following Depression Treatment. Journal of Abnormal Psychology 112(3) August 2003 p 488–496

According to currently burgeoning "third wave approaches" such as Acceptance and Commitment Therapy (ACT) it might be more important to change the client's dysfunctional cognitions and responses to dysfunctional cognitions. Something, of course, which has been done implicitly, within CT of the traditional "Beckian" type, anyway. You cannot really do any thought record without "standing back" from your cognitions. A closer look at Haye's et al. work might be helpful in this debate .