This post is based in part on my thoughts regarding Doidge's book on neuroplasticity.
Psychoanalysis is a type of psychotherapy in which patients usually attend sessions almost every day (3-5 days per week, 50 minutes each time). The details of theory and practice vary, but in general psychoanalysts tend to believe that early childhood events and memories are very important to examine and understand, and that these events (e.g. relationships with mother) have direct causal links to adult personality traits and psychological symptoms. Also psychoanalysts tend to believe that the relationship with the therapist is a setting in which prior relationship dynamics recur, in the form of "transference." Most psychoanalysts assume a relatively quiet or passive stance, tending not to have active conversation or "problem solving" dialogs with patients. Also most psychoanalysts would tend to interpret various types of phenomena, such as dreams, behavioural habits, etc. as laden with meaning. A course of psychoanalysis might take years, and in general the model would be that the patient would "work through" various childhood conflicts, including as they might be transferentially manifest in the therapy, and that the patient might come to understand the various themes at play in their lives, as manifest in dreams, habits, and interpersonal behaviour. This process of understanding and "working through" is thought to lead to symptom relief and life change.
Doidge himself is a psychoanalyst. One of the chapters in his book describes psychoanalysis as a "neuroplastic therapy." (chapter 9, Turning our Ghosts into Ancestors). Part of the support for his claim comes from a case study (a type of evidence characteristic of psychoanalytic thinking). And part of his support comes from briefly describing the life and work of Eric Kandel, the great nobel laureate neuroscientist.
Kandel's work brilliantly demonstrated some of the specific anatomic and molecular changes that happen in neurons as memories are formed.
Kandel himself has been an advocate of incorporating recent biological scientific knowledge into the practice of psychiatry and psychoanalysis (see: http://www.hhmi.org/bulletin/kandel/), and had apparently planned to become a psychoanalyst himself.
I consider it not to be particularly relevant to mention Kandel at all, other than to quote someone important who probably considers psychoanalysis a good thing. It is a common sales tactic to mention an important person's name while trying to convince someone of something. Also it is common in medicine and psychiatry--but especially in alternative medicine--for there to be some mention of something that sounds "scientific" to bolster the public opinion of a product, while the science itself, if looked at closely, is only obliquely related. For example, many questionably effective naturopathic remedies, sold at quite a profit, include advertising laden with some kind of biochemical jargon, much of which, at close examination, lacks substance, but which sounds impressive.
I believe that psychoanalysis can be a powerful and transformative experience. However,I also strongly suspect that there are elements of dogma contained within the theory which are irrelevant to its beneficial effects, and which at times could make it an inefficient therapy.
Consider this thought experiment:
Suppose the beneficial effects of psychoanalysis are due to the following factors:
1) meeting with someone for an hour per day, who will listen and try to understand life problems
2) finding an "explanation" for symptoms. In the case of psychoanalysis this explanation tends to come from an examination of early life events.
Suppose that it is the belief in the explanation that causes symptom improvement, therefore that if some alternative "explanation" for symptoms could be developed, then it would lead to the same symptom improvement. Therefore, suppose that the psychoanalytic theory of character and symptom development is actually a fiction, akin to a dogmatic religious belief system, but that adherence to this belief system, and the resultant faith and conviction, would be the causes of symptom relief and character change.
A way to test this would be to conduct a randomized study of two types of intensive, long-term psychotherapy. Both would be 5 sessions per week, 50 minutes per session, lasting 5 years.
Group 1 patients would have psychoanalysis.
Group 2 patients would receive the same intensive, empathic, sessions, with intelligent and thoughtful, well-boundaried therapists. But let us imagine that some other belief system would underlie the therapy for group 2. For example, astrology. Or some form of religious fundamentalism (of any variety). Here, interpretations would be based on the positions of stars & planets, or on passages from religious texts.
A condition for this type of experiment would be that the patients in both groups would have to lack any differences in bias for or against the style of therapy. So, for example, patients in group 1 would have to have a similar level of belief that psychoanalysis is a valid and culturally-accepted system of thought, and have similar respect for the therapist, compared to the beliefs about therapist and therapy style of patients in group 2 (regarding astrology or fundamentalism, etc.).
In both groups, I suspect that subject matter would come up in the sessions, which would require the therapists to respond either empathically or interpretively. There would probably be dreams that would come up, probably interpreted quite differently--or not at all-- in both groups. The process of therapy, dream interpretation, feelings of closeness with therapist, etc. might well be experienced similarly between groups.
My hypothesis is that both the groups would show similar improvement in a 5 year course of therapy, with only a slight advantage for group 1. I believe this is because the core effect of such therapy is not from the theoretical belief system, but from the process, which is caring, consistent, empathic, understanding, and interpretive. Failed therapy experiences may happen in both groups, some of which because the patients do not like the style or belief system which is being introduced, some of which because life problems can be treatment-resistant at times, some of which because the patient did not feel well-matched with the therapist. I think group 1 would do very slightly better than group 2, because despite the dogma involved in psychoanalytic theory, the underlying process is more intellectually open (at its best).
Unfortunately, I think there is a substantial risk for people in both groups to come out of the experience with stronger dogmatic beliefs, irrespective of any therapeutic improvement. In a more mature psychoanalytic frame, I think this risk would be diminished, as the process would hopefully be more intellectually open.
I do believe that we as intelligent creatures should always seek the "truth" as best we can know it, and therefore we need to challenge our dogmas. The best therapies, in my opinion, need to seek such truths without being restricted by dogma. This is consistent with the underlying theme of psychoanalysis, which I think is about liberation (liberation from symptoms, liberation from past harms or traumas, etc.).
I am reminded now of Joseph Campbell, the comparative mythologist, who might argue that the different styles of therapy are something like different mythologies, none of which are literally "true", but perhaps all of which might contain core aspects of wisdom about the human condition. He might also argue that dogmatic, literalistic adherence to any system of belief could obstruct its underlying message. But he would also agree, I think, that one has to have "faith"--a sense of trust, engagement, and belief--in order to have a transformative experience from anything.
In psychoanalysis, I think it is immensely valuable to seek meaning by examining early childhood events, and by searching for meaning and themes in dreams and nuances of behaviour. But I think it can be can be obstructive to believe, literally, for example, that specific non-traumatic events or patterns of engagement with one's mother at the age of 2, are the causes of specific adult symptoms. I consider the greatness of psychoanalytic interpretation to lie in its focus upon a human life as though it is a great novel or work of art, and that the therapy is partly an experience of understanding, analyzing themes, interpreting, looking at context, in order to enrichen the experience of the art.
A weakness in psychoanalytic practice can, in my opinion, be due to its passive approach at times, which can render it less efficient. Another weakness can be due to a dogmatic or literalistic over-absorption with the theory, causing the therapy to digress--sometimes for years--into an examination of early childhood events, when the core elements of therapeutic need lie solidly in the present, or in the more recent past. I think modern psychoanalysis needs to much more actively incorporate ideas from cognitive and behavioural therapies, from social psychology, as well as from behavioural genetics, etc., and to actively question its dogma.
From a "neuroplastic" point of view, I think the immense advantage of psychoanalysis is in the frame, which is intense (5 days per week), long-term (over years), intellectually open (anything that passes through one's mind is encouraged to be spoken), and consistent. If one was taking language or music lessons, we would see MUCH more "neuroplastic change" in the brain (and, much more importantly, we would see much more language or music learning), if the lessons took place 5 times a week for 5 years, rather than just once a week for 6 months. The consistency and discipline of the psychoanalytic frame is powerfully motivational, just as is any other consistent and disciplined educational framework.